跳过导航链接
A B C D E F G H I J K L M N O P Q R S T U V W Y Z 

A

abs() - 类 ai.rapids.cudf.ColumnView 中的方法
计算绝对值,输出类型与输入类型相同。
AbstractStructBuilder(String, boolean) - 类 ai.rapids.cudf.ColumnWriterOptions.AbstractStructBuilder 的构造器
用于构建 Struct 元数据的构造器
AbstractStructBuilder(String, boolean, int) - 类 ai.rapids.cudf.ColumnWriterOptions.AbstractStructBuilder 的构造器
 
AbstractStructBuilder() - 类 ai.rapids.cudf.ColumnWriterOptions.AbstractStructBuilder 的构造器
 
add(BinaryOperable, DType) - 接口 ai.rapids.cudf.BinaryOperable 中的方法
将一个向量添加到另一个向量,并指定输出类型。
add(BinaryOperable) - 接口 ai.rapids.cudf.BinaryOperable 中的方法
加法 + 运算符。
addBatch(long, long, ByteBuffer, ByteBuffer, ByteBuffer) - 类 ai.rapids.cudf.ArrowColumnBuilder 中的方法
添加 Arrow 缓冲区。
addCalendricalMonths(ColumnView) - 类 ai.rapids.cudf.ColumnView 中的方法
将指定的月数添加到时间戳。
addCalendricalMonths(Scalar) - 类 ai.rapids.cudf.ColumnView 中的方法
将指定的月数添加到时间戳。
addColumn(DType, String, int) - 类 ai.rapids.cudf.Schema.Builder 中的方法
添加新列
addColumn(DType, String) - 类 ai.rapids.cudf.Schema.Builder 中的方法
 
addRef() - 类 ai.rapids.cudf.MemoryCleaner.Cleaner 中的方法
 
address - 类 ai.rapids.cudf.MemoryBuffer 中的变量
 
addressOutOfBoundsCheck(long, long, String) - 类 ai.rapids.cudf.MemoryBuffer 中的方法
 
aggregate(GroupByAggregationOnColumn...) - 类 ai.rapids.cudf.Table.GroupByOperation 中的方法
聚合以索引表示的列组。用法:aggregate(count(), max(2),...); 示例:输入: 1, 1, 1 1, 2, 1 2, 4, 5 table.groupBy(0, 2).count() col0, col1 输出: 1, 1 1, 2 2, 1 ==> 聚合计数
aggregateWindows(AggregationOverWindow...) - 类 ai.rapids.cudf.Table.GroupByOperation 中的方法
根据参数中指定的窗口,计算 Table/投影上的基于行的窗口聚合函数。
aggregateWindowsOverRanges(AggregationOverWindow...) - 类 ai.rapids.cudf.Table.GroupByOperation 中的方法
根据参数中指定的窗口,计算 Table/投影上的基于范围的窗口聚合函数。
Aggregation128Utils - ai.rapids.cudf 中的类
在聚合期间用于分解和重新组合 128 位值的实用方法,以实现基于哈希的聚合并检测溢出。
Aggregation128Utils() - 类 ai.rapids.cudf.Aggregation128Utils 的构造器
 
AggregationOverWindow - ai.rapids.cudf 中的类
一个 Aggregation 实例,同时包含列号和窗口元数据,以便聚合可以在特定窗口上进行。
ai.rapids.cudf - 包 ai.rapids.cudf
 
ai.rapids.cudf.ast - 包 ai.rapids.cudf.ast
 
ai.rapids.cudf.nvcomp - 包 ai.rapids.cudf.nvcomp
 
all() - 类 ai.rapids.cudf.ColumnView 中的方法
返回一个布尔标量,如果列中所有元素都为 true 或非零,则为 true,否则为 false。
all(DType) - 类 ai.rapids.cudf.ColumnView 中的方法
已弃用。
唯一支持的输出类型是 BOOL8。
all() - 类 ai.rapids.cudf.ReductionAggregation 中的静态方法
所有元素的归约。
all() - 类 ai.rapids.cudf.SegmentedReductionAggregation 中的静态方法
所有元素的归约。
alloc(long) - 类 ai.rapids.cudf.Rmm 中的静态方法
使用 stream 0 分配设备内存并返回设备内存指针。
alloc(long, Cuda.Stream) - 类 ai.rapids.cudf.Rmm 中的静态方法
分配设备内存并返回设备内存指针。
allocate(long) - 类 ai.rapids.cudf.CudaMemoryBuffer 中的静态方法
分配用于 GPU 的内存。
allocate(long, Cuda.Stream) - 类 ai.rapids.cudf.CudaMemoryBuffer 中的静态方法
分配用于 GPU 的内存。
allocate(long, boolean) - 类 ai.rapids.cudf.CuFileBuffer 中的静态方法
分配用于 GPU 上 cuFile 的内存。
allocate(long, boolean) - 类 ai.rapids.cudf.DefaultHostMemoryAllocator 中的方法
 
allocate(long) - 类 ai.rapids.cudf.DefaultHostMemoryAllocator 中的方法
 
allocate(long) - 类 ai.rapids.cudf.DeviceMemoryBuffer 中的静态方法
分配用于 GPU 的内存。
allocate(long, Cuda.Stream) - 类 ai.rapids.cudf.DeviceMemoryBuffer 中的静态方法
分配用于 GPU 的内存。
allocate(long, boolean) - 接口 ai.rapids.cudf.HostMemoryAllocator 中的方法
分配内存,但务必关闭返回的缓冲区以避免内存泄漏。
allocate(long) - 接口 ai.rapids.cudf.HostMemoryAllocator 中的方法
分配内存,但务必关闭返回的缓冲区以避免内存泄漏。
allocate(long, boolean) - 类 ai.rapids.cudf.HostMemoryBuffer 中的静态方法
分配内存,但务必关闭返回的缓冲区以避免内存泄漏。
allocate(long) - 类 ai.rapids.cudf.HostMemoryBuffer 中的静态方法
分配内存,但务必关闭返回的缓冲区以避免内存泄漏。
allocate(long, HostMemoryAllocator) - 类 ai.rapids.cudf.PinnedMemoryPool 中的静态方法
创建主机缓冲区(最好指向固定内存)的工厂方法。
allocate(long) - 类 ai.rapids.cudf.PinnedMemoryPool 中的静态方法
创建主机缓冲区(最好指向固定内存)的工厂方法。
allocateRaw(long) - 类 ai.rapids.cudf.HostMemoryBuffer 中的静态方法
绕过默认分配器分配主机内存。
allocCuda(long, Cuda.Stream) - 类 ai.rapids.cudf.Rmm 中的静态方法
使用 `cudaMalloc` 分配设备内存并返回设备内存指针。
allocFromFallbackPinnedPool(long) - 类 ai.rapids.cudf.Rmm 中的静态方法
 
allocFromPinnedPool(long, long) - 类 ai.rapids.cudf.Rmm 中的静态方法
 
allowSingleQuotes(boolean) - 类 ai.rapids.cudf.GetJsonObjectOptions.Builder 中的方法
 
and(BinaryOperable, DType) - 接口 ai.rapids.cudf.BinaryOperable 中的方法
执行逻辑与 (&&),并指定输出类型。
and(BinaryOperable) - 接口 ai.rapids.cudf.BinaryOperable 中的方法
执行逻辑与 (&&)。
any() - 类 ai.rapids.cudf.ColumnView 中的方法
返回一个布尔标量,如果列中任何元素都为 true 或非零,则为 true,否则为 false。
any(DType) - 类 ai.rapids.cudf.ColumnView 中的方法
如果列中任何元素都为 true 或非零,则根据指定的类型,返回 true 或 1,否则返回 false 或 0。
any() - 类 ai.rapids.cudf.ReductionAggregation 中的静态方法
任意元素的归约。
any() - 类 ai.rapids.cudf.SegmentedReductionAggregation 中的静态方法
任意元素的归约。
append(CuFileBuffer, long) - 类 ai.rapids.cudf.CuFileWriteHandle 中的方法
将指定的 cuFile 缓冲区同步附加到文件。
append(boolean) - 类 ai.rapids.cudf.HostColumnVector.Builder 中的方法
 
append(byte) - 类 ai.rapids.cudf.HostColumnVector.Builder 中的方法
 
append(byte, long) - 类 ai.rapids.cudf.HostColumnVector.Builder 中的方法
 
append(short) - 类 ai.rapids.cudf.HostColumnVector.Builder 中的方法
 
append(int) - 类 ai.rapids.cudf.HostColumnVector.Builder 中的方法
 
append(long) - 类 ai.rapids.cudf.HostColumnVector.Builder 中的方法
 
append(float) - 类 ai.rapids.cudf.HostColumnVector.Builder 中的方法
 
append(double) - 类 ai.rapids.cudf.HostColumnVector.Builder 中的方法
 
append(BigDecimal) - 类 ai.rapids.cudf.HostColumnVector.Builder 中的方法
使用 UNNECESSARY RoundingMode 将 java.math.BigDecimal 附加到 HostColumnVector 中。
append(BigDecimal, RoundingMode) - 类 ai.rapids.cudf.HostColumnVector.Builder 中的方法
使用用户定义的 RoundingMode 将 java.math.BigDecimal 附加到 HostColumnVector 中。
append(String) - 类 ai.rapids.cudf.HostColumnVector.Builder 中的方法
 
append(HostColumnVector) - 类 ai.rapids.cudf.HostColumnVector.Builder 中的方法
将此向量附加到此向量的末尾。
append(byte) - 类 ai.rapids.cudf.HostColumnVector.ColumnBuilder 中的方法
 
append(short) - 类 ai.rapids.cudf.HostColumnVector.ColumnBuilder 中的方法
 
append(int) - 类 ai.rapids.cudf.HostColumnVector.ColumnBuilder 中的方法
 
append(long) - 类 ai.rapids.cudf.HostColumnVector.ColumnBuilder 中的方法
 
append(float) - 类 ai.rapids.cudf.HostColumnVector.ColumnBuilder 中的方法
 
append(double) - 类 ai.rapids.cudf.HostColumnVector.ColumnBuilder 中的方法
 
append(boolean) - 类 ai.rapids.cudf.HostColumnVector.ColumnBuilder 中的方法
 
append(BigDecimal) - 类 ai.rapids.cudf.HostColumnVector.ColumnBuilder 中的方法
 
append(BigInteger) - 类 ai.rapids.cudf.HostColumnVector.ColumnBuilder 中的方法
 
append(String) - 类 ai.rapids.cudf.HostColumnVector.ColumnBuilder 中的方法
 
append(byte[], int, int) - 类 ai.rapids.cudf.HostColumnVector.ColumnBuilder 中的方法
附加多个非空字节值。
appendArray(byte...) - 类 ai.rapids.cudf.HostColumnVector.Builder 中的方法
 
appendArray(short...) - 类 ai.rapids.cudf.HostColumnVector.Builder 中的方法
 
appendArray(int...) - 类 ai.rapids.cudf.HostColumnVector.Builder 中的方法
 
appendArray(long...) - 类 ai.rapids.cudf.HostColumnVector.Builder 中的方法
 
appendArray(float...) - 类 ai.rapids.cudf.HostColumnVector.Builder 中的方法
 
appendArray(double...) - 类 ai.rapids.cudf.HostColumnVector.Builder 中的方法
 
appendBoxed(BigDecimal...) - 类 ai.rapids.cudf.HostColumnVector.Builder 中的方法
附加多个值。
appendBoxed(Byte...) - 类 ai.rapids.cudf.HostColumnVector.Builder 中的方法
附加多个值。
appendBoxed(Boolean...) - 类 ai.rapids.cudf.HostColumnVector.Builder 中的方法
附加多个值。
appendBoxed(Short...) - 类 ai.rapids.cudf.HostColumnVector.Builder 中的方法
附加多个值。
appendBoxed(Integer...) - 类 ai.rapids.cudf.HostColumnVector.Builder 中的方法
附加多个值。
appendBoxed(Long...) - 类 ai.rapids.cudf.HostColumnVector.Builder 中的方法
附加多个值。
appendBoxed(Float...) - 类 ai.rapids.cudf.HostColumnVector.Builder 中的方法
附加多个值。
appendBoxed(Double...) - 类 ai.rapids.cudf.HostColumnVector.Builder 中的方法
附加多个值。
appendBoxed(String...) - 类 ai.rapids.cudf.HostColumnVector.Builder 中的方法
附加多个值。
appendByteList(byte[]) - 类 ai.rapids.cudf.HostColumnVector.ColumnBuilder 中的方法
将字节附加到 INT8/UINT8 的 LIST 中。
appendByteList(byte[], int, int) - 类 ai.rapids.cudf.HostColumnVector.ColumnBuilder 中的方法
将字节附加到 INT8/UINT8 的 LIST 中。
appendBytes(byte) - 类 ai.rapids.cudf.Range 中的静态方法
将范围附加到构建器。
appendBytes(byte, byte) - 类 ai.rapids.cudf.Range 中的静态方法
将范围附加到构建器。
appendBytes(byte, byte, byte) - 类 ai.rapids.cudf.Range 中的静态方法
将范围附加到构建器。
appendDecimal128(byte[]) - 类 ai.rapids.cudf.HostColumnVector.ColumnBuilder 中的方法
接受一个字节数组,该数组包含无缩放值的补码表示,其字节顺序为大端。
appendDeviceBufferToFile(File, BaseDeviceMemoryBuffer) - 类 ai.rapids.cudf.CuFile 中的静态方法
将设备缓冲区同步附加到给定文件路径。
appendDeviceMemoryToFile(File, long, long) - 类 ai.rapids.cudf.CuFile 中的静态方法
将设备内存同步附加到给定文件路径。
appendDoubles(double, double, double) - 类 ai.rapids.cudf.Range 中的静态方法
将范围附加到构建器。
appendDoubles(double) - 类 ai.rapids.cudf.Range 中的静态方法
将范围附加到构建器。
appendDoubles(double, double) - 类 ai.rapids.cudf.Range 中的静态方法
将范围附加到构建器。
appendFloats(float, float, float) - 类 ai.rapids.cudf.Range 中的静态方法
将范围附加到构建器。
appendFloats(float) - 类 ai.rapids.cudf.Range 中的静态方法
将范围附加到构建器。
appendFloats(float, float) - 类 ai.rapids.cudf.Range 中的静态方法
将范围附加到构建器。
appendInts(int) - 类 ai.rapids.cudf.Range 中的静态方法
将范围附加到构建器。
appendInts(int, int) - 类 ai.rapids.cudf.Range 中的静态方法
将范围附加到构建器。
appendInts(int, int, int) - 类 ai.rapids.cudf.Range 中的静态方法
将范围附加到构建器。
appendLists(List...) - 类 ai.rapids.cudf.HostColumnVector.ColumnBuilder 中的方法
 
appendLongs(long, long, long) - 类 ai.rapids.cudf.Range 中的静态方法
将范围附加到构建器。
appendLongs(long) - 类 ai.rapids.cudf.Range 中的静态方法
将范围附加到构建器。
appendLongs(long, long) - 类 ai.rapids.cudf.Range 中的静态方法
将范围附加到构建器。
appendNull() - 类 ai.rapids.cudf.HostColumnVector.Builder 中的方法
附加 null 值。
appendNull() - 类 ai.rapids.cudf.HostColumnVector.ColumnBuilder 中的方法
 
appendShorts(short) - 类 ai.rapids.cudf.Range 中的静态方法
将范围附加到构建器。
appendShorts(short, short) - 类 ai.rapids.cudf.Range 中的静态方法
将范围附加到构建器。
appendShorts(short, short, short) - 类 ai.rapids.cudf.Range 中的静态方法
将范围附加到构建器。
appendStructValues(List<HostColumnVector.StructData>) - 类 ai.rapids.cudf.HostColumnVector.ColumnBuilder 中的方法
 
appendStructValues(HostColumnVector.StructData...) - 类 ai.rapids.cudf.HostColumnVector.ColumnBuilder 中的方法
 
appendUnscaledDecimal(int) - 类 ai.rapids.cudf.HostColumnVector.Builder 中的方法
 
appendUnscaledDecimal(long) - 类 ai.rapids.cudf.HostColumnVector.Builder 中的方法
 
appendUnscaledDecimal(BigInteger) - 类 ai.rapids.cudf.HostColumnVector.Builder 中的方法
 
appendUnscaledDecimalArray(int...) - 类 ai.rapids.cudf.HostColumnVector.Builder 中的方法
 
appendUnscaledDecimalArray(long...) - 类 ai.rapids.cudf.HostColumnVector.Builder 中的方法
 
appendUTF8String(byte[]) - 类 ai.rapids.cudf.HostColumnVector.Builder 中的方法
 
appendUTF8String(byte[], int, int) - 类 ai.rapids.cudf.HostColumnVector.Builder 中的方法
 
appendUTF8String(byte[]) - 类 ai.rapids.cudf.HostColumnVector.ColumnBuilder 中的方法
 
appendUTF8String(byte[], int, int) - 类 ai.rapids.cudf.HostColumnVector.ColumnBuilder 中的方法
 
apply(ColumnWriterOptions) - 接口 ai.rapids.cudf.ColumnWriterOptions.ByteArrayProducer 中的方法
 
apply(ColumnWriterOptions) - 接口 ai.rapids.cudf.ColumnWriterOptions.IntArrayProducer 中的方法
 
applyBooleanMask(ColumnView) - 类 ai.rapids.cudf.ColumnView 中的方法
使用 `booleanMaskView` 布尔值列表作为掩码,过滤此 LIST 列中每行的元素。
approxPercentile(double[]) - 类 ai.rapids.cudf.ColumnView 中的方法
计算此 ColumnVector 的各种百分位数,该向量必须包含 t-digest 聚合生成的质心。
approxPercentile(ColumnVector) - 类 ai.rapids.cudf.ColumnView 中的方法
计算此 ColumnVector 的各种百分位数,该向量必须包含 t-digest 聚合生成的质心。
arccos() - 类 ai.rapids.cudf.ColumnView 中的方法
计算反余弦,输出类型与输入类型相同。
arccosh() - 类 ai.rapids.cudf.ColumnView 中的方法
计算反双曲余弦,输出类型与输入类型相同。
arcsin() - 类 ai.rapids.cudf.ColumnView 中的方法
计算反正弦,输出类型与输入类型相同。
arcsinh() - 类 ai.rapids.cudf.ColumnView 中的方法
计算反双曲线正弦,输出类型与输入类型相同。
arctan() - 类 ai.rapids.cudf.ColumnView 中的方法
计算反正切,输出类型与输入类型相同。
arctan2(BinaryOperable, DType) - 接口 ai.rapids.cudf.BinaryOperable 中的方法
函数 arctan2(y,x) 或 atan2(y,x) 定义为欧几里得平面中,正 x 轴与点 (x, y) ≠ (0, 0) 射线之间的夹角(以弧度表示)。
arctan2(BinaryOperable) - 接口 ai.rapids.cudf.BinaryOperable 中的方法
函数 arctan2(y,x) 或 atan2(y,x) 定义为欧几里得平面中,正 x 轴与点 (x, y) ≠ (0, 0) 射线之间的夹角(以弧度表示)。
arctanh() - 类 ai.rapids.cudf.ColumnView 中的方法
计算反双曲线反正切,输出类型与输入类型相同。
ARENA - 类 ai.rapids.cudf.RmmAllocationMode 中的静态变量
使用竞技场子分配策略
argMax() - 类 ai.rapids.cudf.GroupByAggregation 中的静态方法
最大元素的索引。
argMax() - 类 ai.rapids.cudf.RollingAggregation 中的静态方法
最大元素的索引。
argMin() - 类 ai.rapids.cudf.GroupByAggregation 中的静态方法
最小元素的索引。
argMin() - 类 ai.rapids.cudf.RollingAggregation 中的静态方法
最小元素的索引。
ArrowColumnBuilder - ai.rapids.cudf 中的类
从 Arrow 数据构建列。
ArrowColumnBuilder(HostColumnVector.DataType) - 类 ai.rapids.cudf.ArrowColumnBuilder 的构造器
 
ArrowIPCOptions - ai.rapids.cudf 中的类
用于读取 Arrow IPC 格式数据的选项
ArrowIPCOptions.Builder - ai.rapids.cudf 中的类
 
ArrowIPCOptions.NeedGpu - ai.rapids.cudf 中的接口
 
ArrowIPCWriterOptions - ai.rapids.cudf 中的类
写入 Arrow IPC 数据时的设置。
ArrowIPCWriterOptions.Builder - ai.rapids.cudf 中的类
 
ArrowIPCWriterOptions.DoneOnGpu - ai.rapids.cudf 中的接口
 
asByteBuffer() - 类 ai.rapids.cudf.HostMemoryBuffer 中的方法
返回一个 ByteBuffer,用于访问底层内存。
asByteBuffer(long, int) - 类 ai.rapids.cudf.HostMemoryBuffer 中的方法
返回一个 ByteBuffer,用于访问底层内存。
asByteList() - 类 ai.rapids.cudf.ColumnView 中的方法
转换为字节列表。此方法转换 ColumnVector 提供的行,并将每行转换为字节列表,并反转字节序。
asByteList(boolean) - 类 ai.rapids.cudf.ColumnView 中的方法
转换为字节列表。此方法转换 ColumnVector 提供的行,并将每行转换为字节列表。
asBytes() - 类 ai.rapids.cudf.ColumnView 中的方法
转换为 Byte - ColumnVector。此方法获取 ColumnVector 提供的值并转换为字节。从 Date、Timestamp 或 Boolean 转换为 byte 类型时,将使用数据的底层数字表示进行转换。
asc(int) - 类 ai.rapids.cudf.OrderByArg 中的静态方法
 
asc(int, boolean) - 类 ai.rapids.cudf.OrderByArg 中的静态方法
 
asDoubles() - 类 ai.rapids.cudf.ColumnView 中的方法
转换为 Double - ColumnVector。此方法获取 ColumnVector 提供的值并转换为双精度浮点数。从 Date、Timestamp 或 Boolean 转换为 double 类型时,将使用数据的底层数字表示进行转换。
asFloats() - 类 ai.rapids.cudf.ColumnView 中的方法
转换为 Float - ColumnVector。此方法获取 ColumnVector 提供的值并转换为单精度浮点数。从 Date、Timestamp 或 Boolean 转换为 float 类型时,将使用数据的底层数字表示进行转换。
asHostDataType() - 类 ai.rapids.cudf.Schema 中的方法
 
asInts() - 类 ai.rapids.cudf.ColumnView 中的方法
转换为 Int - ColumnVector。此方法获取 ColumnVector 提供的值并转换为整数。从 Date、Timestamp 或 Boolean 转换为 int 类型时,将使用数据的底层数字表示进行转换。
asLongs() - 类 ai.rapids.cudf.ColumnView 中的方法
转换为 Long - ColumnVector。此方法获取 ColumnVector 提供的值并转换为长整数。从 Date、Timestamp 或 Boolean 转换为 long 类型时,将使用数据的底层数字表示进行转换。
AssertEmptyNulls - ai.rapids.cudf 中的类
此类是一个帮助类,用于断言 ColumnView 中没有非空的 null 值。此类存在的原因是,我们可以通过传递 “-da:ai.rapids.cudf.AssertEmptyNulls” 在需要时开启/关闭断言。
AssertEmptyNulls() - 类 ai.rapids.cudf.AssertEmptyNulls 的构造器
 
assertNullsAreEmpty(ColumnView) - 类 ai.rapids.cudf.AssertEmptyNulls 中的静态方法
 
asShorts() - 类 ai.rapids.cudf.ColumnView 中的方法
转换为 Short - ColumnVector。此方法获取 ColumnVector 提供的值并转换为短整数。从 Date、Timestamp 或 Boolean 转换为 short 类型时,将使用数据的底层数字表示进行转换。
asStrings() - 类 ai.rapids.cudf.ColumnView 中的方法
转换为字符串。
asStrings(String) - 类 ai.rapids.cudf.ColumnView 中的方法
用于解析时间戳列向量并将其转换为字符串列向量的方法。
AstExpression - ai.rapids.cudf.ast 中的类
AST 中每个节点的基类
AstExpression() - 类 ai.rapids.cudf.ast.AstExpression 的构造器
 
AstExpression.ExpressionType - ai.rapids.cudf.ast 中的枚举
AST 中可出现在序列化 AST 中的节点类型的枚举。
asTimestamp(DType, String) - 类 ai.rapids.cudf.ColumnView 中的方法
将字符串解析为时间戳。
asTimestampDays() - 类 ai.rapids.cudf.ColumnView 中的方法
转换为 TIMESTAMP_DAYS - ColumnVector。此方法获取 ColumnVector 提供的值并转换为 TIMESTAMP_DAYS。
asTimestampDays(String) - 类 ai.rapids.cudf.ColumnView 中的方法
转换为 TIMESTAMP_DAYS - ColumnVector。此方法获取 ColumnVector 提供的字符串值并转换为 TIMESTAMP_DAYS。
asTimestampMicroseconds() - 类 ai.rapids.cudf.ColumnView 中的方法
转换为 TIMESTAMP_MICROSECONDS - ColumnVector。此方法获取 ColumnVector 提供的值并转换为 TIMESTAMP_MICROSECONDS。
asTimestampMicroseconds(String) - 类 ai.rapids.cudf.ColumnView 中的方法
转换为 TIMESTAMP_MICROSECONDS - ColumnVector。此方法获取 ColumnVector 提供的字符串值并转换为 TIMESTAMP_MICROSECONDS。
asTimestampMilliseconds() - 类 ai.rapids.cudf.ColumnView 中的方法
转换为 TIMESTAMP_MILLISECONDS - ColumnVector。此方法获取 ColumnVector 提供的值并转换为 TIMESTAMP_MILLISECONDS。
asTimestampMilliseconds(String) - 类 ai.rapids.cudf.ColumnView 中的方法
转换为 TIMESTAMP_MILLISECONDS - ColumnVector。此方法获取 ColumnVector 提供的字符串值并转换为 TIMESTAMP_MILLISECONDS。
asTimestampNanoseconds() - 类 ai.rapids.cudf.ColumnView 中的方法
转换为 TIMESTAMP_NANOSECONDS - ColumnVector。此方法获取 ColumnVector 提供的值并转换为 TIMESTAMP_NANOSECONDS。
asTimestampNanoseconds(String) - 类 ai.rapids.cudf.ColumnView 中的方法
转换为 TIMESTAMP_NANOSECONDS - ColumnVector。此方法获取 ColumnVector 提供的字符串值并转换为 TIMESTAMP_NANOSECONDS。
asTimestampSeconds() - 类 ai.rapids.cudf.ColumnView 中的方法
转换为 TIMESTAMP_SECONDS - ColumnVector。此方法获取 ColumnVector 提供的值并转换为 TIMESTAMP_SECONDS。
asTimestampSeconds(String) - 类 ai.rapids.cudf.ColumnView 中的方法
转换为 TIMESTAMP_SECONDS - ColumnVector。此方法获取 ColumnVector 提供的字符串值并转换为 TIMESTAMP_SECONDS。
asUnsignedBytes() - 类 ai.rapids.cudf.ColumnView 中的方法
转换为无符号 Byte - ColumnVector。此方法获取 ColumnVector 提供的值并转换为字节。从 Date、Timestamp 或 Boolean 转换为 byte 类型时,将使用数据的底层数字表示进行转换。
asUnsignedInts() - 类 ai.rapids.cudf.ColumnView 中的方法
转换为无符号 Int - ColumnVector。此方法获取 ColumnVector 提供的值并转换为整数。从 Date、Timestamp 或 Boolean 转换为 int 类型时,将使用数据的底层数字表示进行转换。
asUnsignedLongs() - 类 ai.rapids.cudf.ColumnView 中的方法
转换为无符号 Long - ColumnVector。此方法获取 ColumnVector 提供的值并转换为长整数。从 Date、Timestamp 或 Boolean 转换为 long 类型时,将使用数据的底层数字表示进行转换。
asUnsignedShorts() - 类 ai.rapids.cudf.ColumnView 中的方法
转换为无符号 Short - ColumnVector。此方法获取 ColumnVector 提供的值并转换为短整数。从 Date、Timestamp 或 Boolean 转换为 short 类型时,将使用数据的底层数字表示进行转换。
asyncMemset(long, byte, long) - 类 ai.rapids.cudf.Cuda 中的静态方法
用 value 设置从 dst 指向的内存区域开始的 count 字节。
autoSetDevice() - 类 ai.rapids.cudf.Cuda 中的静态方法
为此线程设置适当的设备。
AvroOptions - ai.rapids.cudf 中的类
读取 Avro 文件的选项
AvroOptions.Builder - ai.rapids.cudf 中的类
 

B

BaseDeviceMemoryBuffer - ai.rapids.cudf 中的类
设备内存中所有 MemoryBuffer 的基类。
BaseDeviceMemoryBuffer(long, long, MemoryBuffer) - 类 ai.rapids.cudf.BaseDeviceMemoryBuffer 的构造器
 
BaseDeviceMemoryBuffer(long, long, MemoryBuffer.MemoryBufferCleaner) - 类 ai.rapids.cudf.BaseDeviceMemoryBuffer 的构造器
 
BasicType(boolean, DType) - 类 ai.rapids.cudf.HostColumnVector.BasicType 的构造器
 
batchedCompressAsync(long, long, long, long, long, long, long, long, long) - 类 ai.rapids.cudf.nvcomp.BatchedCompressor 中的方法
异步压缩一批缓冲区。
batchedCompressAsync(long, long, long, long, long, long, long, long, long) - 类 ai.rapids.cudf.nvcomp.BatchedLZ4Compressor 中的方法
 
batchedCompressAsync(long, long, long, long, long, long, long, long, long) - 类 ai.rapids.cudf.nvcomp.BatchedZstdCompressor 中的方法
 
batchedCompressGetTempSize(long, long) - 类 ai.rapids.cudf.nvcomp.BatchedCompressor 中的方法
获取执行整个批次压缩所需的临时工作空间大小(字节)。
batchedCompressGetTempSize(long, long) - 类 ai.rapids.cudf.nvcomp.BatchedLZ4Compressor 中的方法
 
batchedCompressGetTempSize(long, long) - 类 ai.rapids.cudf.nvcomp.BatchedZstdCompressor 中的方法
 
BatchedCompressor - ai.rapids.cudf.nvcomp 中的类
多缓冲区压缩器
BatchedCompressor(long, long, long) - 类 ai.rapids.cudf.nvcomp.BatchedCompressor 的构造器
构造一个批量压缩器实例。
batchedDecompressAsync(long, long, long, long, long, long, long, long) - 类 ai.rapids.cudf.nvcomp.BatchedDecompressor 中的方法
异步解压一批压缩数据缓冲区。
batchedDecompressAsync(long, long, long, long, long, long, long, long) - 类 ai.rapids.cudf.nvcomp.BatchedLZ4Decompressor 中的方法
 
batchedDecompressAsync(long, long, long, long, long, long, long, long) - 类 ai.rapids.cudf.nvcomp.BatchedZstdDecompressor 中的方法
 
batchedDecompressGetTempSize(long, long) - 类 ai.rapids.cudf.nvcomp.BatchedDecompressor 中的方法
计算解压压缩批次所需的临时存储大小(字节)。
batchedDecompressGetTempSize(long, long) - 类 ai.rapids.cudf.nvcomp.BatchedLZ4Decompressor 中的方法
 
batchedDecompressGetTempSize(long, long) - 类 ai.rapids.cudf.nvcomp.BatchedZstdDecompressor 中的方法
 
BatchedDecompressor - ai.rapids.cudf.nvcomp 中的类
对批量输入缓冲区进行操作的解压器
BatchedDecompressor(long) - 类 ai.rapids.cudf.nvcomp.BatchedDecompressor 的构造器
构造一个批量解压器实例。
BatchedLZ4Compressor - ai.rapids.cudf.nvcomp 中的类
多缓冲区 LZ4 压缩器
BatchedLZ4Compressor(long, long) - 类 ai.rapids.cudf.nvcomp.BatchedLZ4Compressor 的构造器
构造一个批量 LZ4 压缩器实例。
BatchedLZ4Decompressor - ai.rapids.cudf.nvcomp 中的类
对批量输入缓冲区进行操作的 LZ4 解压器
BatchedLZ4Decompressor(long) - 类 ai.rapids.cudf.nvcomp.BatchedLZ4Decompressor 的构造器
 
BatchedZstdCompressor - ai.rapids.cudf.nvcomp 中的类
多缓冲区 ZSTD 压缩器
BatchedZstdCompressor(long, long) - 类 ai.rapids.cudf.nvcomp.BatchedZstdCompressor 的构造器
构造一个批量 ZSTD 压缩器实例。
BatchedZstdDecompressor - ai.rapids.cudf.nvcomp 中的类
对批量输入缓冲区进行操作的 ZSTD 解压器
BatchedZstdDecompressor(long) - 类 ai.rapids.cudf.nvcomp.BatchedZstdDecompressor 的构造器
 
BinaryOp - ai.rapids.cudf 中的枚举
数学二元运算。
binaryOp(BinaryOp, BinaryOperable, DType) - 接口 ai.rapids.cudf.BinaryOperable 中的方法
多种不同的二元运算。
binaryOp(BinaryOp, BinaryOperable, DType) - 类 ai.rapids.cudf.ColumnView 中的方法
多种不同的二元运算。
binaryOp(BinaryOp, BinaryOperable, DType) - 类 ai.rapids.cudf.Scalar 中的方法
 
BinaryOperable - ai.rapids.cudf 中的接口
 
BinaryOperation - ai.rapids.cudf.ast 中的类
由运算符和两个操作数组成的二元运算。
BinaryOperation(BinaryOperator, AstExpression, AstExpression) - 类 ai.rapids.cudf.ast.BinaryOperation 的构造器
 
BinaryOperator - ai.rapids.cudf.ast 中的枚举
AST 中可出现在二元运算中的运算符枚举。
bitAnd(BinaryOperable, DType) - 接口 ai.rapids.cudf.BinaryOperable 中的方法
执行按位与 (&),并指定输出类型。
bitAnd(BinaryOperable) - 接口 ai.rapids.cudf.BinaryOperable 中的方法
执行按位与 (&)。
bitCastTo(DType) - 类 ai.rapids.cudf.ColumnView 中的方法
在具有相同底层长度的类型之间进行零拷贝转换。
bitInvert() - 类 ai.rapids.cudf.ColumnView 中的方法
反转位,输出类型与输入类型相同。
bitOr(BinaryOperable, DType) - 接口 ai.rapids.cudf.BinaryOperable 中的方法
执行按位或 (|),并指定输出类型。
bitOr(BinaryOperable) - 接口 ai.rapids.cudf.BinaryOperable 中的方法
执行按位或 (|)。
bitXor(BinaryOperable, DType) - 接口 ai.rapids.cudf.BinaryOperable 中的方法
执行按位异或 (^),并指定输出类型。
bitXor(BinaryOperable) - 接口 ai.rapids.cudf.BinaryOperable 中的方法
执行按位异或 (^)。
BOOL8 - 类 ai.rapids.cudf.DType 中的静态变量
 
boolFromBytes(byte...) - 类 ai.rapids.cudf.ColumnVector 中的静态方法
从给定值创建新的向量。
boolFromBytes(byte...) - 类中的静态方法 ai.rapids.cudf.HostColumnVector
从给定值创建新的向量。
bounds(int, int) - 类中的静态方法 ai.rapids.cudf.DecimalUtils
给定 Decimal 精度和缩放比例,返回当前 Decimal 类型的下界和上界。
BufferType - 包中的枚举 ai.rapids.cudf
ColumnVectors 和 HostColumnVectors 支持的缓冲区类型
build() - 类中的方法 ai.rapids.cudf.ArrowIPCOptions.Builder
 
build() - 类中的方法 ai.rapids.cudf.ArrowIPCWriterOptions.Builder
 
build() - 类中的方法 ai.rapids.cudf.AvroOptions.Builder
 
build(DType, int, Consumer<HostColumnVector.Builder>) - 类中的静态方法 ai.rapids.cudf.ColumnVector
创建新的向量。
build(int, long, Consumer<HostColumnVector.Builder>) - 类中的静态方法 ai.rapids.cudf.ColumnVector
 
build() - 类中的方法 ai.rapids.cudf.ColumnWriterOptions.ListBuilder
 
build() - 类中的方法 ai.rapids.cudf.ColumnWriterOptions.NestedBuilder
 
build() - 类中的方法 ai.rapids.cudf.ColumnWriterOptions.StructBuilder
 
build() - 类中的方法 ai.rapids.cudf.CSVOptions.Builder
 
build() - 类中的方法 ai.rapids.cudf.CSVWriterOptions.Builder
 
build() - 类中的方法 ai.rapids.cudf.GetJsonObjectOptions.Builder
 
build() - 类中的方法 ai.rapids.cudf.GroupByOptions.Builder
 
build(DType, int, Consumer<HostColumnVector.Builder>) - 类中的静态方法 ai.rapids.cudf.HostColumnVector
创建新的向量。
build(int, long, Consumer<HostColumnVector.Builder>) - 类中的静态方法 ai.rapids.cudf.HostColumnVector
 
build() - 类中的方法 ai.rapids.cudf.HostColumnVector.Builder
完成并创建不可变的 CudfColumn。
build() - 类中的方法 ai.rapids.cudf.HostColumnVector.ColumnBuilder
 
build() - 类中的方法 ai.rapids.cudf.JSONOptions.Builder
 
build() - 类中的方法 ai.rapids.cudf.ORCOptions.Builder
 
build() - 类中的方法 ai.rapids.cudf.ORCWriterOptions.Builder
 
build() - 类中的方法 ai.rapids.cudf.ParquetOptions.Builder
 
build() - 类中的方法 ai.rapids.cudf.ParquetWriterOptions.Builder
 
build() - 类中的方法 ai.rapids.cudf.Schema.Builder
 
build() - 类中的方法 ai.rapids.cudf.Table.TestBuilder
 
build() - 类中的方法 ai.rapids.cudf.TableDebug.Builder
 
build() - 类中的方法 ai.rapids.cudf.WindowOptions.Builder
 
buildAndPutOnDevice() - 类中的方法 ai.rapids.cudf.ArrowColumnBuilder
根据 Arrow 数据创建不可变的 ColumnVector,并复制到设备上。
buildAndPutOnDevice() - 类中的方法 ai.rapids.cudf.HostColumnVector.Builder
完成并创建不可变的 ColumnVector,并复制到设备上。
buildAndPutOnDevice() - 类中的方法 ai.rapids.cudf.HostColumnVector.ColumnBuilder
完成并创建不可变的 ColumnVector,并复制到设备上。
builder() - 类中的静态方法 ai.rapids.cudf.ArrowIPCOptions
 
Builder() - 类的构造方法 ai.rapids.cudf.ArrowIPCOptions.Builder
 
builder() - 类中的静态方法 ai.rapids.cudf.ArrowIPCWriterOptions
 
Builder() - 类的构造方法 ai.rapids.cudf.ArrowIPCWriterOptions.Builder
 
builder() - 类中的静态方法 ai.rapids.cudf.AvroOptions
 
Builder() - 类的构造方法 ai.rapids.cudf.AvroOptions.Builder
 
Builder() - 类的构造方法 ai.rapids.cudf.ColumnFilterOptions.Builder
 
Builder() - 类的构造方法 ai.rapids.cudf.CompressionMetadataWriterOptions.Builder
 
builder() - 类中的静态方法 ai.rapids.cudf.CSVOptions
 
Builder() - 类的构造方法 ai.rapids.cudf.CSVOptions.Builder
 
builder() - 类中的静态方法 ai.rapids.cudf.CSVWriterOptions
 
Builder() - 类的构造方法 ai.rapids.cudf.CSVWriterOptions.Builder
 
builder() - 类中的静态方法 ai.rapids.cudf.GetJsonObjectOptions
 
Builder() - 类的构造方法 ai.rapids.cudf.GetJsonObjectOptions.Builder
 
builder() - 类中的静态方法 ai.rapids.cudf.GroupByOptions
 
Builder() - 类的构造方法 ai.rapids.cudf.GroupByOptions.Builder
 
builder(DType, int) - 类中的静态方法 ai.rapids.cudf.HostColumnVector
创建一个新的 Builder 以容纳指定行数。
builder(int, long) - 类中的静态方法 ai.rapids.cudf.HostColumnVector
创建一个新的 Builder 以容纳指定行数,并留有足够的空间来容纳给定量的字符串数据。
builder() - 类中的静态方法 ai.rapids.cudf.JSONOptions
 
Builder() - 类的构造方法 ai.rapids.cudf.JSONOptions.Builder
 
builder() - 类中的静态方法 ai.rapids.cudf.ORCOptions
 
Builder() - 类的构造方法 ai.rapids.cudf.ORCOptions.Builder
 
builder() - 类中的静态方法 ai.rapids.cudf.ORCWriterOptions
 
Builder() - 类的构造方法 ai.rapids.cudf.ORCWriterOptions.Builder
 
builder() - 类中的静态方法 ai.rapids.cudf.ParquetOptions
 
Builder() - 类的构造方法 ai.rapids.cudf.ParquetOptions.Builder
 
builder() - 类中的静态方法 ai.rapids.cudf.ParquetWriterOptions
 
Builder() - 类的构造方法 ai.rapids.cudf.ParquetWriterOptions.Builder
 
builder() - 类中的静态方法 ai.rapids.cudf.Schema
 
builder() - 类中的静态方法 ai.rapids.cudf.TableDebug
 
Builder() - 类的构造方法 ai.rapids.cudf.TableDebug.Builder
 
builder() - 类中的静态方法 ai.rapids.cudf.WindowOptions
 
Builder() - 类的构造方法 ai.rapids.cudf.WindowOptions.Builder
 
buildMetadata() - 类中的方法 ai.rapids.cudf.ChunkedPack
生成不透明的表元数据,该元数据稍后可以通过 `cudf::unpack` 进行解包。

C

capitalize(Scalar) - 类中的方法 ai.rapids.cudf.ColumnView
返回一个将字符串首字母大写的列。
capture() - 类中的方法 ai.rapids.cudf.RegexProgram
获取用于创建此实例的捕获组设置
CaptureGroups - 包中的枚举 ai.rapids.cudf
捕获组设置,与 cudf::strings::capture_groups 紧密相关。
castTo(DType) - 类中的方法 ai.rapids.cudf.ColumnVector
用于转换 ColumnVector 的通用方法。当从 Date、Timestamp 或 Boolean 类型转换为数值类型时,将使用数据的底层数值表示进行转换。
castTo(DType) - 类中的方法 ai.rapids.cudf.ColumnView
用于转换 ColumnVector 的通用方法。当从 Date、Timestamp 或 Boolean 类型转换为数值类型时,将使用数据的底层数值表示进行转换。
cbrt() - 类中的方法 ai.rapids.cudf.ColumnView
计算立方根,输出类型与输入类型相同。
ceil() - 类中的方法 ai.rapids.cudf.ColumnView
计算上取整(ceil),输出类型与输入类型相同。
childColumnOptions - 类中的变量 ai.rapids.cudf.ColumnWriterOptions
 
children - 类中的变量 ai.rapids.cudf.ColumnWriterOptions.NestedBuilder
 
children - 类中的变量 ai.rapids.cudf.HostColumnVectorCore
 
children - 类中的变量 ai.rapids.cudf.JCudfSerialization.SerializedColumnHeader
 
ChunkedPack - 包中的类 ai.rapids.cudf
cudf::chunked_pack 的 JNI 接口。
ChunkedPack(long) - 类的构造方法 ai.rapids.cudf.ChunkedPack
创建原生 `cudf::chunked_pack` 后,由 `Table.makeChunkedPack` 调用此构造方法。
clamp(Scalar, Scalar) - 类中的方法 ai.rapids.cudf.ColumnView
将 `input` 中小于 `lo` 的值替换为 `lo`,将大于 `hi` 的值替换为 `hi`。
clamp(Scalar, Scalar, Scalar, Scalar) - 类中的方法 ai.rapids.cudf.ColumnView
将 `input` 中小于 `lo` 的值替换为 `lo_replace`,将大于 `hi` 的值替换为 `hi_replace`。
clean(boolean) - 类中的方法 ai.rapids.cudf.MemoryCleaner.Cleaner
清理任何之前未释放的资源。
cleaner - 类中的变量 ai.rapids.cudf.MemoryBuffer
 
Cleaner() - 类的构造方法 ai.rapids.cudf.MemoryCleaner.Cleaner
 
cleanImpl(boolean) - 类中的方法 ai.rapids.cudf.ColumnVector.OffHeapState
 
cleanImpl(boolean) - 类中的方法 ai.rapids.cudf.HostColumnVectorCore.OffHeapState
 
cleanImpl(boolean) - 类中的方法 ai.rapids.cudf.MemoryCleaner.Cleaner
清理任何之前未释放的资源。
cleanupDefaultCudaDevice() - 类中的静态方法 ai.rapids.cudf.Rmm
 
clearEventHandler() - 类中的静态方法 ai.rapids.cudf.Rmm
如果已设置活动 RMM 事件处理程序,则清除它。
close() - 类中的方法 ai.rapids.cudf.ArrowColumnBuilder
 
close() - 类中的方法 ai.rapids.cudf.ast.CompiledExpression
 
close() - 类中的方法 ai.rapids.cudf.ChunkedPack
 
close() - 类中的方法 ai.rapids.cudf.CloseableArray
 
close(Exception) - 类中的方法 ai.rapids.cudf.CloseableArray
 
close() - 类中的方法 ai.rapids.cudf.ColumnVector
关闭此 Vector 并释放为 HostMemoryBuffer 和 DeviceMemoryBuffer 分配的内存
close() - 类中的方法 ai.rapids.cudf.ColumnView
 
close() - 类中的方法 ai.rapids.cudf.ContigSplitGroupByResult
 
close() - 类中的方法 ai.rapids.cudf.ContiguousTable
关闭连续表实例及其底层资源。
close() - 类中的方法 ai.rapids.cudf.Cuda.Event
 
close() - 类中的方法 ai.rapids.cudf.Cuda.Stream
 
close() - 类中的方法 ai.rapids.cudf.CuFileBuffer
 
close() - 类中的方法 ai.rapids.cudf.DataSource
 
close() - 类中的方法 ai.rapids.cudf.GatherMap
关闭支持 GatherMap 数据的设备缓冲区。
close() - 类中的方法 ai.rapids.cudf.HashJoin
 
close() - 包中的接口 ai.rapids.cudf.HostBufferProvider
表示将不再提供缓冲区。
close() - 类中的方法 ai.rapids.cudf.HostColumnVector.Builder
关闭此 Builder 并释放内存(如果 CudfColumn 未生成)。
close() - 类中的方法 ai.rapids.cudf.HostColumnVector
关闭此 Vector 并释放为 HostMemoryBuffer 和 DeviceMemoryBuffer 分配的内存
close() - 类中的方法 ai.rapids.cudf.HostColumnVector.ColumnBuilder
 
close() - 类中的方法 ai.rapids.cudf.HostColumnVectorCore
列的关闭方法
close() - 类中的方法 ai.rapids.cudf.JCudfSerialization.HostConcatResult
 
close() - 类中的方法 ai.rapids.cudf.JCudfSerialization.TableAndRowCountPair
 
close() - 类中的方法 ai.rapids.cudf.MemoryBuffer
关闭此缓冲区并释放内存
close() - 类中的方法 ai.rapids.cudf.MixedJoinSize
 
close() - 类中的方法 ai.rapids.cudf.MultiBufferDataSource
 
close() - 类中的方法 ai.rapids.cudf.NvtxRange
 
close() - 类中的方法 ai.rapids.cudf.NvtxUniqueRange
 
close() - 类中的方法 ai.rapids.cudf.ORCChunkedReader
 
close() - 类中的方法 ai.rapids.cudf.PackedColumnMetadata
关闭 PackedColumnMetadata 实例及其底层资源。
close() - 类中的方法 ai.rapids.cudf.ParquetChunkedReader
 
close() - 类中的方法 ai.rapids.cudf.PartitionedTable
 
close() - 类中的方法 ai.rapids.cudf.PinnedMemoryPool
 
close() - 类中的方法 ai.rapids.cudf.RmmArenaMemoryResource
 
close() - 类中的方法 ai.rapids.cudf.RmmCudaAsyncMemoryResource
 
close() - 类中的方法 ai.rapids.cudf.RmmCudaMemoryResource
 
close() - 包中的接口 ai.rapids.cudf.RmmDeviceMemoryResource
 
close() - 类中的方法 ai.rapids.cudf.RmmEventHandlerResourceAdaptor
 
close() - 类中的方法 ai.rapids.cudf.RmmLimitingResourceAdaptor
 
close() - 类中的方法 ai.rapids.cudf.RmmLoggingResourceAdaptor
 
close() - 类中的方法 ai.rapids.cudf.RmmManagedMemoryResource
 
close() - 类中的方法 ai.rapids.cudf.RmmPoolMemoryResource
 
close() - 类中的方法 ai.rapids.cudf.RmmTrackingResourceAdaptor
 
close() - 类中的方法 ai.rapids.cudf.RmmWrappingDeviceMemoryResource
 
close() - 类中的方法 ai.rapids.cudf.Scalar
释放与标量关联的内存。
close() - 包中的接口 ai.rapids.cudf.StreamedTableReader
 
close() - 类中的方法 ai.rapids.cudf.Table
 
close() - 类中的方法 ai.rapids.cudf.TableWithMeta
 
close() - 类中的方法 ai.rapids.cudf.TableWriter
 
close() - 类中的方法 ai.rapids.cudf.WindowOptions
 
CloseableArray<T extends AutoCloseable> - 包中的类 ai.rapids.cudf
一个工具类,封装了一个可关闭实例数组,并可以关闭该数组。
closeAt(int) - 类中的方法 ai.rapids.cudf.CloseableArray
 
closed - 类中的变量 ai.rapids.cudf.MemoryBuffer
 
closeGroups() - 类中的方法 ai.rapids.cudf.ContigSplitGroupByResult
关闭分割的组表
closeUDFInstance(long) - 类中的静态方法 ai.rapids.cudf.HostUDFWrapper
关闭由 `createUDFInstance` 创建的派生 UDF 实例。
closeUniqKeyTable() - 类中的方法 ai.rapids.cudf.ContigSplitGroupByResult
关闭键表或键列
codePoints() - 类中的方法 ai.rapids.cudf.ColumnView
获取每个字符串中每个字符的码点值(整数)。
collectList() - 类中的静态方法 ai.rapids.cudf.GroupByAggregation
将值收集到列表中。
collectList(NullPolicy) - 类中的静态方法 ai.rapids.cudf.GroupByAggregation
将值收集到列表中。
collectList() - 类中的静态方法 ai.rapids.cudf.ReductionAggregation
 
collectList(NullPolicy) - 类中的静态方法 ai.rapids.cudf.ReductionAggregation
将值收集到列表中。
collectList() - 类中的静态方法 ai.rapids.cudf.RollingAggregation
将值收集到列表中。
collectList(NullPolicy) - 类中的静态方法 ai.rapids.cudf.RollingAggregation
将值收集到列表中。
collectSet() - 类中的静态方法 ai.rapids.cudf.GroupByAggregation
将值收集到集合中。
collectSet(NullPolicy, NullEquality, NaNEquality) - 类中的静态方法 ai.rapids.cudf.GroupByAggregation
将值收集到集合中。
collectSet() - 类中的静态方法 ai.rapids.cudf.ReductionAggregation
将值收集到集合中。
collectSet(NullPolicy, NullEquality, NaNEquality) - 类中的静态方法 ai.rapids.cudf.ReductionAggregation
将值收集到集合中。
collectSet() - 类中的静态方法 ai.rapids.cudf.RollingAggregation
将值收集到集合中。
collectSet(NullPolicy, NullEquality, NaNEquality) - 类中的静态方法 ai.rapids.cudf.RollingAggregation
将值收集到集合中。
column(DType, String, int) - 类中的方法 ai.rapids.cudf.Schema.Builder
向当前 schema 添加单个列。
column(DType, String) - 类中的方法 ai.rapids.cudf.Schema.Builder
 
column(String...) - 类中的方法 ai.rapids.cudf.Table.TestBuilder
 
column(Boolean...) - 类中的方法 ai.rapids.cudf.Table.TestBuilder
 
column(Byte...) - 类中的方法 ai.rapids.cudf.Table.TestBuilder
 
column(Short...) - 类中的方法 ai.rapids.cudf.Table.TestBuilder
 
column(Integer...) - 类中的方法 ai.rapids.cudf.Table.TestBuilder
 
column(Long...) - 类中的方法 ai.rapids.cudf.Table.TestBuilder
 
column(Float...) - 类中的方法 ai.rapids.cudf.Table.TestBuilder
 
column(Double...) - 类中的方法 ai.rapids.cudf.Table.TestBuilder
 
column(HostColumnVector.ListType, List<?>...) - 类中的方法 ai.rapids.cudf.Table.TestBuilder
 
column(String[]...) - 类中的方法 ai.rapids.cudf.Table.TestBuilder
 
column(Boolean[]...) - 类中的方法 ai.rapids.cudf.Table.TestBuilder
 
column(Byte[]...) - 类中的方法 ai.rapids.cudf.Table.TestBuilder
 
column(Short[]...) - 类中的方法 ai.rapids.cudf.Table.TestBuilder
 
column(Integer[]...) - 类中的方法 ai.rapids.cudf.Table.TestBuilder
 
column(Long[]...) - 类中的方法 ai.rapids.cudf.Table.TestBuilder
 
column(Float[]...) - 类中的方法 ai.rapids.cudf.Table.TestBuilder
 
column(Double[]...) - 类中的方法 ai.rapids.cudf.Table.TestBuilder
 
column(HostColumnVector.StructType, HostColumnVector.StructData...) - 类中的方法 ai.rapids.cudf.Table.TestBuilder
 
column(HostColumnVector.StructType, HostColumnVector.StructData[]...) - 类中的方法 ai.rapids.cudf.Table.TestBuilder
 
ColumnBuilder(HostColumnVector.DataType, long) - 类的构造方法 ai.rapids.cudf.HostColumnVector.ColumnBuilder
 
ColumnFilterOptions - 包中的类 ai.rapids.cudf
可过滤列的输入格式的基础选项类。
ColumnFilterOptions(ColumnFilterOptions.Builder<?>) - 类的构造方法 ai.rapids.cudf.ColumnFilterOptions
 
ColumnFilterOptions.Builder<T extends ColumnFilterOptions.Builder> - 包中的类 ai.rapids.cudf
 
columnIndex - 类中的变量 ai.rapids.cudf.GroupByAggregationOnColumn
 
columnIndex - 类中的变量 ai.rapids.cudf.GroupByScanAggregationOnColumn
 
columnIndex - 类中的变量 ai.rapids.cudf.RollingAggregationOnColumn
 
ColumnReference - 包中的类 ai.rapids.cudf.ast
输入表中列的引用。
ColumnReference(int) - 类的构造方法 ai.rapids.cudf.ast.ColumnReference
构造对唯一或最左侧输入表的列引用
ColumnReference(int, TableReference) - 类的构造方法 ai.rapids.cudf.ast.ColumnReference
构造对指定表中指定列索引的引用
ColumnVector - 包中的类 ai.rapids.cudf
此类表示不可变的数据向量。
ColumnVector(long) - 类的构造方法 ai.rapids.cudf.ColumnVector
将现有的设备上的 cudf::column 包装到相应的 ColumnVector 中。
ColumnVector(DType, long, Optional<Long>, DeviceMemoryBuffer, DeviceMemoryBuffer, DeviceMemoryBuffer) - 类的构造方法 ai.rapids.cudf.ColumnVector
根据已在设备上的数据创建新的列向量。
ColumnVector(DType, long, Optional<Long>, DeviceMemoryBuffer, DeviceMemoryBuffer, DeviceMemoryBuffer, List<DeviceMemoryBuffer>, long[]) - 类的构造方法 ai.rapids.cudf.ColumnVector
根据已在设备上的数据创建新的列向量,带子列。
ColumnVector.EventHandler - 包中的接口 ai.rapids.cudf
用于处理此 ColumnVector 事件的接口。
ColumnVector.OffHeapState - 包中的类 ai.rapids.cudf
保存列向量的堆外状态,以便即使发生内存泄漏也能进行清理。
ColumnView - 包中的类 ai.rapids.cudf
此类表示列的 column_view,类似于其相应的 cudf cpp counterpart。
ColumnView(ColumnVector.OffHeapState) - 类的构造方法 ai.rapids.cudf.ColumnView
旨在在 ColumnVector 构建时从其内部调用。
ColumnView(DType, long, Optional<Long>, BaseDeviceMemoryBuffer, BaseDeviceMemoryBuffer, ColumnView[]) - 类的构造方法 ai.rapids.cudf.ColumnView
根据已在设备上的数据创建新的列视图。
ColumnView(DType, long, Optional<Long>, BaseDeviceMemoryBuffer, BaseDeviceMemoryBuffer) - 类的构造方法 ai.rapids.cudf.ColumnView
根据已在设备上的数据创建新的列视图。
ColumnView(DType, long, Optional<Long>, BaseDeviceMemoryBuffer, BaseDeviceMemoryBuffer, BaseDeviceMemoryBuffer) - 类的构造方法 ai.rapids.cudf.ColumnView
根据已在设备上的数据创建新的列视图。
ColumnView.FindOptions - 包中的枚举 ai.rapids.cudf
用于选择 listIndexOf 函数行为的枚举:1。
ColumnWriterOptions - 包中的类 ai.rapids.cudf
写入 Parquet/ORC 文件的每列设置。
ColumnWriterOptions(String, boolean, int, boolean) - 类的构造方法 ai.rapids.cudf.ColumnWriterOptions
 
ColumnWriterOptions(String, boolean, int, boolean, int) - 类的构造方法 ai.rapids.cudf.ColumnWriterOptions
 
ColumnWriterOptions(String, boolean) - 类的构造方法 ai.rapids.cudf.ColumnWriterOptions
 
ColumnWriterOptions(String, boolean, int) - 类的构造方法 ai.rapids.cudf.ColumnWriterOptions
 
ColumnWriterOptions(String) - 类的构造方法 ai.rapids.cudf.ColumnWriterOptions
 
ColumnWriterOptions.AbstractStructBuilder<T extends ColumnWriterOptions.AbstractStructBuilder,V extends ColumnWriterOptions> - 包中的类 ai.rapids.cudf
 
ColumnWriterOptions.ByteArrayProducer - 包中的接口 ai.rapids.cudf
 
ColumnWriterOptions.IntArrayProducer - 包中的接口 ai.rapids.cudf
 
ColumnWriterOptions.ListBuilder - 包中的类 ai.rapids.cudf
 
ColumnWriterOptions.ListColumnWriterOptions - 包中的类 ai.rapids.cudf
 
ColumnWriterOptions.NestedBuilder<T extends ColumnWriterOptions.NestedBuilder,V extends ColumnWriterOptions> - 包中的类 ai.rapids.cudf
 
ColumnWriterOptions.StructBuilder - 包中的类 ai.rapids.cudf
 
ColumnWriterOptions.StructColumnWriterOptions - 包中的类 ai.rapids.cudf
 
combinedFlags() - 类中的方法 ai.rapids.cudf.RegexProgram
使用 'or' 组合正则表达式标志
combineInt64SumChunks(Table, DType) - 类中的静态方法 ai.rapids.cudf.Aggregation128Utils
从一个包含四个 64 位整数列的表中重组一个 128 位值的列,并检查是否溢出。
compile() - 类中的方法 ai.rapids.cudf.ast.AstExpression
 
CompiledExpression - 包中的类 ai.rapids.cudf.ast
此类封装了一个原生编译的 AST,必须关闭以避免原生内存泄漏。
compress(BaseDeviceMemoryBuffer[], Cuda.Stream) - 类中的方法 ai.rapids.cudf.nvcomp.BatchedCompressor
压缩一批缓冲区。
CompressionMetadataWriterOptions - 包中的类 ai.rapids.cudf
 
CompressionMetadataWriterOptions(CompressionMetadataWriterOptions.Builder) - 类的构造方法 ai.rapids.cudf.CompressionMetadataWriterOptions
 
CompressionMetadataWriterOptions.Builder<T extends CompressionMetadataWriterOptions.Builder,V extends CompressionMetadataWriterOptions> - 包中的类 ai.rapids.cudf
 
compressionRatio - 类中的变量 ai.rapids.cudf.TableWriter.WriteStatistics
 
CompressionType - 包中的枚举 ai.rapids.cudf
压缩格式的枚举。
CompressionType - 包中的枚举 ai.rapids.cudf.nvcomp
可压缩数据类型的枚举。
computeColumn(Table) - 类中的方法 ai.rapids.cudf.ast.CompiledExpression
通过将此 AST 表达式应用于指定表来计算新列。
computeHashCode() - 类中的方法 ai.rapids.cudf.HostUDFWrapper
 
concatenate(ColumnView...) - 类中的静态方法 ai.rapids.cudf.ColumnVector
通过连接多个列创建新的向量。
concatenate(Table...) - 类中的静态方法 ai.rapids.cudf.Table
连接多个表以形成一个表。
concatToContiguousTable(JCudfSerialization.SerializedTableHeader[], HostMemoryBuffer[]) - 类中的静态方法 ai.rapids.cudf.JCudfSerialization
将主机内存中的多个表连接到设备内存中的一个连续表中。
concatToHostBuffer(JCudfSerialization.SerializedTableHeader[], HostMemoryBuffer[], HostMemoryAllocator) - 类中的静态方法 ai.rapids.cudf.JCudfSerialization
将主机内存中的多个表连接到单个主机表缓冲区中。
concatToHostBuffer(JCudfSerialization.SerializedTableHeader[], HostMemoryBuffer[]) - 类中的静态方法 ai.rapids.cudf.JCudfSerialization
 
conditionalFullJoinGatherMaps(Table, CompiledExpression) - 类中的方法 ai.rapids.cudf.Table
计算 GatherMaps,当条件表达式为真时,可用于显示两个表之间全连接(full join)的结果。
conditionalInnerJoinGatherMaps(Table, CompiledExpression) - 类中的方法 ai.rapids.cudf.Table
计算 GatherMaps,当条件表达式为真时,可用于显示两个表之间内连接(inner join)的结果。
conditionalInnerJoinGatherMaps(Table, CompiledExpression, long) - 类中的方法 ai.rapids.cudf.Table
计算 GatherMaps,当条件表达式为真时,可用于显示两个表之间内连接(inner join)的结果。
conditionalInnerJoinRowCount(Table, CompiledExpression) - 类中的方法 ai.rapids.cudf.Table
计算当条件表达式为真时,两个表之间内连接(inner join)结果的行数。
conditionalLeftAntiJoinGatherMap(Table, CompiledExpression) - 类中的方法 ai.rapids.cudf.Table
计算 GatherMap,当条件表达式为真时,可用于显示两个表之间左反连接(left anti join)的结果。
conditionalLeftAntiJoinGatherMap(Table, CompiledExpression, long) - 类中的方法 ai.rapids.cudf.Table
计算 GatherMap,当条件表达式为真时,可用于显示两个表之间左反连接(left anti join)的结果。
conditionalLeftAntiJoinRowCount(Table, CompiledExpression) - 类中的方法 ai.rapids.cudf.Table
计算当条件表达式为真时,两个表之间左反连接(left anti join)结果的行数。
conditionalLeftJoinGatherMaps(Table, CompiledExpression) - 类中的方法 ai.rapids.cudf.Table
计算 GatherMaps,当条件表达式为真时,可用于显示两个表之间左连接(left join)的结果。
conditionalLeftJoinGatherMaps(Table, CompiledExpression, long) - 类中的方法 ai.rapids.cudf.Table
计算 GatherMaps,当条件表达式为真时,可用于显示两个表之间左连接(left join)的结果。
conditionalLeftJoinRowCount(Table, CompiledExpression) - 类中的方法 ai.rapids.cudf.Table
计算当条件表达式为真时,两个表之间左连接(left join)结果的行数。
conditionalLeftSemiJoinGatherMap(Table, CompiledExpression) - 类中的方法 ai.rapids.cudf.Table
计算 GatherMap,当条件表达式为真时,可用于显示两个表之间左半连接(left semi join)的结果。
conditionalLeftSemiJoinGatherMap(Table, CompiledExpression, long) - 类中的方法 ai.rapids.cudf.Table
计算 GatherMap,当条件表达式为真时,可用于显示两个表之间左半连接(left semi join)的结果。
conditionalLeftSemiJoinRowCount(Table, CompiledExpression) - 类中的方法 ai.rapids.cudf.Table
计算当条件表达式为真时,两个表之间左半连接(left semi join)结果的行数。
configuredDefaultShutdownHook() - 类中的静态方法 ai.rapids.cudf.MemoryCleaner
检查是否配置了在关闭时检查内存泄漏的 shutdown hook。
configureDefaultCudfPinnedPoolSize(long) - 类中的静态方法 ai.rapids.cudf.PinnedMemoryPool
设置 cuDF 默认 pinned 内存池的大小。
configureDefaultCudfPinnedPoolSize(long) - 类中的静态方法 ai.rapids.cudf.Rmm
设置 cuDF 默认 pinned 内存池的大小。
contains(Scalar) - 类中的方法 ai.rapids.cudf.ColumnView
查找此列中是否存在 `needle` 示例:单列:idx 0 1 2 3 4 col = { 10, 20, 20, 30, 50 } 标量:value = { 20 } result = true
contains(ColumnView) - 类中的方法 ai.rapids.cudf.ColumnView
返回一个与此列大小相同的新列,该列的元素类型为 DType.BOOL8,如果此列中的对应条目包含在给定的 searchSpace 列中,则每行值为 true,否则为 false。
containsRe(String) - 类中的方法 ai.rapids.cudf.ColumnView
已弃用。
containsRe(RegexProgram) - 类中的方法 ai.rapids.cudf.ColumnView
返回一个布尔类型的 ColumnVector,标识匹配给定 RegexProgram 模式(从任何位置开始)的行。
ContigSplitGroupByResult - 包中的类 ai.rapids.cudf
用于保存 `Table.contiguousSplitGroupsAndGenUniqKeys` 的分组和唯一键表。唯一键表中的每行对应一个分组 资源管理说明:此类拥有 `groups` 和 `uniqKeysTable`(如果未构造表,则是 uniqKeyColumns)。1:如果您想立即关闭资源,请使用 `closeGroups` 和 `closeUniqKeyTable` 分别关闭资源。
ContigSplitGroupByResult() - 类的构造方法 ai.rapids.cudf.ContigSplitGroupByResult
 
contiguousSplit(int...) - 类中的方法 ai.rapids.cudf.Table
在给定的边界处拆分表,但每个拆分的结果所占用的内存是连续布局的。
contiguousSplitGroups() - 类 ai.rapids.cudf.Table.GroupByOperation 中的方法
根据分组键将单个表中的组拆分成单独的表。
contiguousSplitGroupsAndGenUniqKeys() - 类 ai.rapids.cudf.Table.GroupByOperation 中的方法
Table.GroupByOperation.contiguousSplitGroups() 类似,返回一个额外的唯一键表,其中每行对应一个分组拆分。
ContiguousTable - 类位于 ai.rapids.cudf 包中
由单个连续设备缓冲区支持的表。
convertDecimal128FromJavaToCudf(byte[]) - 类 ai.rapids.cudf.HostColumnVectorCore 中的静态方法
 
copyFromDeviceBuffer(BaseDeviceMemoryBuffer) - 类 ai.rapids.cudf.HostMemoryBuffer 中的方法
将 DeviceMemoryBuffer 同步复制到 HostMemoryBuffer。
copyFromDeviceBuffer(BaseDeviceMemoryBuffer, Cuda.Stream) - 类 ai.rapids.cudf.HostMemoryBuffer 中的方法
使用指定的流将 DeviceMemoryBuffer 复制到 HostMemoryBuffer。
copyFromDeviceBufferAsync(long, BaseDeviceMemoryBuffer, long, long, Cuda.Stream) - 类 ai.rapids.cudf.BaseDeviceMemoryBuffer 中的方法
使用指定的 CUDA 流将 src 的子集从 destOffset 开始复制到此缓冲区。
copyFromDeviceBufferAsync(BaseDeviceMemoryBuffer, Cuda.Stream) - 类 ai.rapids.cudf.HostMemoryBuffer 中的方法
使用指定的流将 DeviceMemoryBuffer 复制到 HostMemoryBuffer。
copyFromHostBuffer(long, HostMemoryBuffer, long, long) - 类 ai.rapids.cudf.BaseDeviceMemoryBuffer 中的方法
将 src 的子集从 destOffset 开始复制到此缓冲区。
copyFromHostBuffer(long, HostMemoryBuffer, long, long, Cuda.Stream) - Class ai.rapids.cudf.BaseDeviceMemoryBuffer 中的方法
使用指定的 CUDA 流将 src 的子集从 destOffset 开始复制到此缓冲区。
copyFromHostBuffer(HostMemoryBuffer, long, long) - Class ai.rapids.cudf.BaseDeviceMemoryBuffer 中的方法
将 src 的子集复制到此缓冲区的开头。
copyFromHostBuffer(HostMemoryBuffer) - Class ai.rapids.cudf.BaseDeviceMemoryBuffer 中的方法
将 src 中的所有内容复制到此缓冲区的开头。
copyFromHostBuffer(HostMemoryBuffer, Cuda.Stream) - Class ai.rapids.cudf.BaseDeviceMemoryBuffer 中的方法
使用 CUDA 流将整个主机缓冲区复制到此缓冲区的开头。
copyFromHostBuffer(long, HostMemoryBuffer, long, long) - Class ai.rapids.cudf.HostMemoryBuffer 中的方法
将给定缓冲区的内容复制到此缓冲区
copyFromHostBufferAsync(long, HostMemoryBuffer, long, long, Cuda.Stream) - Class ai.rapids.cudf.BaseDeviceMemoryBuffer 中的方法
使用指定的 CUDA 流将 src 的子集从 destOffset 开始复制到此缓冲区。
copyFromHostBufferAsync(HostMemoryBuffer, Cuda.Stream) - Class ai.rapids.cudf.BaseDeviceMemoryBuffer 中的方法
使用 CUDA 流将整个主机缓冲区复制到此缓冲区的开头。
copyFromMemoryBuffer(long, MemoryBuffer, long, long, Cuda.Stream) - Class ai.rapids.cudf.MemoryBuffer 中的方法
使用指定的 CUDA 流将 src 的子集从 destOffset 开始复制到此缓冲区。
copyFromMemoryBufferAsync(long, MemoryBuffer, long, long, Cuda.Stream) - Class ai.rapids.cudf.MemoryBuffer 中的方法
使用指定的 CUDA 流将 src 的子集从 destOffset 开始复制到此缓冲区。
copyFromStream(long, InputStream, long) - Class ai.rapids.cudf.HostMemoryBuffer 中的方法
从 in 中复制 len 字节到此缓冲区。
copyToColumnVector() - Class ai.rapids.cudf.ColumnVector 中的方法
对于 ColumnVector,这实际上只是增加了引用计数。
copyToColumnVector() - Class ai.rapids.cudf.ColumnView 中的方法
从列视图句柄创建 ColumnVector
copyToDevice() - Class ai.rapids.cudf.HostColumnVector 中的方法
将数据复制到设备。
copyToHost(HostMemoryAllocator) - Class ai.rapids.cudf.ColumnView 中的方法
同步将数据复制到主机。
copyToHost() - Class ai.rapids.cudf.ColumnView 中的方法
同步将数据复制到主机内存
copyToHostAsync(Cuda.Stream, HostMemoryAllocator) - Class ai.rapids.cudf.ColumnView 中的方法
异步将数据复制到主机。
copyToHostAsync(Cuda.Stream) - Class ai.rapids.cudf.ColumnView 中的方法
异步将数据复制到主机。
cos() - Class ai.rapids.cudf.ColumnView 中的方法
计算余弦,输出类型与输入类型相同。
cosh() - Class ai.rapids.cudf.ColumnView 中的方法
计算双曲余弦,输出类型与输入类型相同。
count() - Class ai.rapids.cudf.GroupByAggregation 中的静态方法
计算有效元素的数量,即
count(NullPolicy) - Class ai.rapids.cudf.GroupByAggregation 中的静态方法
计算元素的数量。
count(NullPolicy) - Class ai.rapids.cudf.GroupByScanAggregation 中的静态方法
计算元素的数量。
count() - Class ai.rapids.cudf.RollingAggregation 中的静态方法
计算有效元素的数量,即
count(NullPolicy) - Class ai.rapids.cudf.RollingAggregation 中的静态方法
计算元素的数量。
countElements() - Class ai.rapids.cudf.ColumnView 中的方法
获取每个列表中的元素数量。
create(DType.DTypeEnum) - Class ai.rapids.cudf.DType 中的静态方法
非 Decimal DType 实例的工厂方法。
create(DType.DTypeEnum, int) - Class ai.rapids.cudf.DType 中的静态方法
专门用于 Decimal DType 实例的工厂方法。
createDecimalType(int, int) - Class ai.rapids.cudf.DecimalUtils 中的静态方法
使用精度和比例创建 cuDF Decimal 类型
createTDigest(int) - Class ai.rapids.cudf.GroupByAggregation 中的静态方法
从固定宽度数值输入列计算 t-digest。
createTDigest(int) - Class ai.rapids.cudf.ReductionAggregation 中的静态方法
tDigest 归约。
createUDFInstance() - Class ai.rapids.cudf.HostUDFWrapper 中的方法
创建派生的主机 UDF 本地实例。
crossJoin(Table) - Class ai.rapids.cudf.Table 中的方法
将两张表交叉连接,左表所有行与右表所有行连接。
CSVOptions - ai.rapids.cudf 中的类
读取 CSV 文件的选项
CSVOptions.Builder - ai.rapids.cudf 中的类
 
CSVWriterOptions - ai.rapids.cudf 中的类
写入 CSV 文件的选项
CSVWriterOptions.Builder - ai.rapids.cudf 中的类
 
Cuda - ai.rapids.cudf 中的类
 
Cuda() - Class ai.rapids.cudf.Cuda 的构造方法
 
Cuda.Event - ai.rapids.cudf 中的类
 
Cuda.Stream - ai.rapids.cudf 中的类
表示 CUDA 流的类
CUDA_ASYNC - Class ai.rapids.cudf.RmmAllocationMode 中的静态变量
使用 CUDA 异步子分配策略
CUDA_ASYNC_FABRIC - Class ai.rapids.cudf.RmmAllocationMode 中的静态变量
使用 CUDA 异步子分配策略,并使用具有读写访问权限的对等可访问 Fabric 句柄
CUDA_DEFAULT - Class ai.rapids.cudf.RmmAllocationMode 中的静态变量
使用 cudaMalloc 进行分配
CUDA_MANAGED_MEMORY - Class ai.rapids.cudf.RmmAllocationMode 中的静态变量
使用 cudaMallocManaged 而非 cudaMalloc
CudaComputeMode - ai.rapids.cudf 中的枚举
这是 CUDA 设备计算模式的 Java 映射。
CudaException - ai.rapids.cudf 中的异常
来自 cuda 语言/库的异常。
CudaException.CudaError - ai.rapids.cudf 中的枚举
cudaError 的 Java 镜像,用于在 JVM 中跟踪 CUDA 错误。
CudaFatalException - ai.rapids.cudf 中的异常
CudaFatalException 是一种 CudaException,它使进程处于不一致状态,任何进一步的 CUDA 工作都将返回相同的错误。
CudaMemInfo - ai.rapids.cudf 中的类
表示可用和总设备内存。
CudaMemoryBuffer - ai.rapids.cudf 中的类
此类表示直接使用 `cudaMalloc` 分配的数据,而不是默认的 RMM 内存资源。
CudaMemoryBuffer(long, long, Cuda.Stream) - Class ai.rapids.cudf.CudaMemoryBuffer 的构造方法
将现有 CUDA 分配包装到设备内存缓冲区中。
Cudf - ai.rapids.cudf 中的类
 
Cudf() - Class ai.rapids.cudf.Cudf 的构造方法
 
CudfColumnSizeOverflowException - ai.rapids.cudf 中的异常
当 CUDF 操作导致列大小超出 CUDF 列大小限制时抛出的异常
CudfException - ai.rapids.cudf 中的异常
由 cudf 本身抛出的异常。
CuFile - ai.rapids.cudf 中的类
用于访问 cuFile API 的 JNI 包装器。
CuFile() - Class ai.rapids.cudf.CuFile 的构造方法
 
CuFileBuffer - ai.rapids.cudf 中的类
表示 cuFile 缓冲区。
CuFileReadHandle - ai.rapids.cudf 中的类
表示用于读取的 cuFile 文件句柄。
CuFileReadHandle(String) - Class ai.rapids.cudf.CuFileReadHandle 的构造方法
使用指定的文件路径构造读取器。
CuFileWriteHandle - ai.rapids.cudf 中的类
表示用于读取的 cuFile 文件句柄。
CuFileWriteHandle(String) - Class ai.rapids.cudf.CuFileWriteHandle 的构造方法
使用指定的文件路径构造写入器。
currentRowFollowing() - Class ai.rapids.cudf.WindowOptions.Builder 中的方法
 
currentRowPreceding() - Class ai.rapids.cudf.WindowOptions.Builder 中的方法
 

D

data - Class ai.rapids.cudf.HostColumnVectorCore.OffHeapState 中的变量
 
DataSource - ai.rapids.cudf 中的类
可用于向 CUDF 动态提供数据的基类。
DataSource() - Class ai.rapids.cudf.DataSource 的构造方法
 
DataType() - Class ai.rapids.cudf.HostColumnVector.DataType 的构造方法
 
dateTimeCeil(DateTimeRoundingFrequency) - Class ai.rapids.cudf.ColumnView 中的方法
将时间戳向上取整到给定频率,保持类型不变。
DateTimeComponent - ai.rapids.cudf 中的枚举
可提取的日期时间组件类型。
dateTimeFloor(DateTimeRoundingFrequency) - Class ai.rapids.cudf.ColumnView 中的方法
将时间戳向下取整到给定频率,保持类型不变。
dateTimeRound(DateTimeRoundingFrequency) - Class ai.rapids.cudf.ColumnView 中的方法
将时间戳(四舍五入)取整到给定频率,保持类型不变。
DateTimeRoundingFrequency - ai.rapids.cudf 中的枚举
 
day() - Class ai.rapids.cudf.ColumnView 中的方法
从时间戳获取日期。
dayOfYear() - Class ai.rapids.cudf.ColumnView 中的方法
从时间戳获取一年中的天数。
daysFromInts(int...) - Class ai.rapids.cudf.ColumnVector 中的静态方法
从给定值创建新的向量。
daysFromInts(int...) - Class ai.rapids.cudf.HostColumnVector 中的静态方法
从给定值创建新的向量。
daysInMonth() - Class ai.rapids.cudf.ColumnView 中的方法
提取月份中的天数
debug(String, Table) - Class ai.rapids.cudf.TableDebug 中的方法
打印表的内容。
debug(String, ColumnView) - Class ai.rapids.cudf.TableDebug 中的方法
打印列的内容。
debug(String, HostColumnVectorCore) - Class ai.rapids.cudf.TableDebug 中的方法
打印列的内容。
DECIMAL128_MAX_PRECISION - Class ai.rapids.cudf.DType 中的静态变量
 
decimal128Column(String...) - Class ai.rapids.cudf.ORCOptions.Builder 中的方法
指定应读取为 DECIMAL128 的 Decimal 列。
decimal128Column(int, RoundingMode, BigInteger...) - Class ai.rapids.cudf.Table.TestBuilder 中的方法
 
DECIMAL32_MAX_PRECISION - Class ai.rapids.cudf.DType 中的静态变量
 
decimal32Column(int, Integer...) - Class ai.rapids.cudf.Table.TestBuilder 中的方法
 
decimal32Column(int, RoundingMode, Double...) - Class ai.rapids.cudf.Table.TestBuilder 中的方法
 
decimal32Column(int, RoundingMode, String...) - Class ai.rapids.cudf.Table.TestBuilder 中的方法
 
DECIMAL64_MAX_PRECISION - Class ai.rapids.cudf.DType 中的静态变量
 
decimal64Column(int, Long...) - Class ai.rapids.cudf.Table.TestBuilder 中的方法
 
decimal64Column(int, RoundingMode, Double...) - Class ai.rapids.cudf.Table.TestBuilder 中的方法
 
decimal64Column(int, RoundingMode, String...) - Class ai.rapids.cudf.Table.TestBuilder 中的方法
 
decimalFromBigInt(int, BigInteger...) - Class ai.rapids.cudf.ColumnVector 中的静态方法
从 BigInteger 创建新的 Decimal 向量。与 [[java.math.BigDecimal]] 的比例相比,这里的比例表示相反的含义。
decimalFromBigIntegers(int, BigInteger...) - Class ai.rapids.cudf.HostColumnVector 中的静态方法
从无标度值(BigInteger 数组)和标度创建新的 Decimal 向量。
decimalFromBoxedInts(int, Integer...) - Class ai.rapids.cudf.ColumnVector 中的静态方法
从封装的无标度值(Integer 数组)和标度创建新的 Decimal 向量。
decimalFromBoxedInts(int, Integer...) - Class ai.rapids.cudf.HostColumnVector 中的静态方法
从封装的无标度值(Integer 数组)和标度创建新的 Decimal 向量。
decimalFromBoxedLongs(int, Long...) - Class ai.rapids.cudf.ColumnVector 中的静态方法
从封装的无标度值(Long 数组)和标度创建新的 Decimal 向量。
decimalFromBoxedLongs(int, Long...) - Class ai.rapids.cudf.HostColumnVector 中的静态方法
从封装的无标度值(Long 数组)和标度创建新的 Decimal 向量。
decimalFromDoubles(DType, RoundingMode, double...) - Class ai.rapids.cudf.ColumnVector 中的静态方法
从具有特定 DecimalType 和 RoundingMode 的双精度浮点数创建新的 Decimal 向量。
decimalFromDoubles(DType, RoundingMode, double...) - Class ai.rapids.cudf.HostColumnVector 中的静态方法
从具有特定 DecimalType 和 RoundingMode 的双精度浮点数创建新的 Decimal 向量。
decimalFromInts(int, int...) - Class ai.rapids.cudf.ColumnVector 中的静态方法
从无标度值(int 数组)和标度创建新的 Decimal 向量。
decimalFromInts(int, int...) - Class ai.rapids.cudf.HostColumnVector 中的静态方法
从无标度值(int 数组)和标度创建新的 Decimal 向量。
decimalFromLongs(int, long...) - Class ai.rapids.cudf.ColumnVector 中的静态方法
从无标度值(long 数组)和标度创建新的 Decimal 向量。
decimalFromLongs(int, long...) - Class ai.rapids.cudf.HostColumnVector 中的静态方法
从无标度值(long 数组)和标度创建新的 Decimal 向量。
DecimalUtils - ai.rapids.cudf 中的类
 
DecimalUtils() - Class ai.rapids.cudf.DecimalUtils 的构造方法
 
decompressAsync(BaseDeviceMemoryBuffer[], BaseDeviceMemoryBuffer[], Cuda.Stream) - Class ai.rapids.cudf.nvcomp.BatchedDecompressor 中的方法
异步解压一批缓冲区
decompressAsync(long, BaseDeviceMemoryBuffer[], BaseDeviceMemoryBuffer[], Cuda.Stream) - Class ai.rapids.cudf.nvcomp.BatchedLZ4Decompressor 中的静态方法
异步解压一批缓冲区
DEFAULT - Class ai.rapids.cudf.ArrowIPCOptions 中的静态变量
 
DEFAULT - Class ai.rapids.cudf.ArrowIPCWriterOptions 中的静态变量
 
DEFAULT - Class ai.rapids.cudf.AvroOptions 中的静态变量
 
DEFAULT - Class ai.rapids.cudf.CSVOptions 中的静态变量
 
DEFAULT - Class ai.rapids.cudf.GetJsonObjectOptions 中的静态变量
 
DEFAULT - Class ai.rapids.cudf.GroupByOptions 中的静态变量
 
DEFAULT - Class ai.rapids.cudf.JSONOptions 中的静态变量
 
DEFAULT - Class ai.rapids.cudf.ORCOptions 中的静态变量
 
DEFAULT - Class ai.rapids.cudf.ParquetOptions 中的静态变量
 
DEFAULT_STREAM - Class ai.rapids.cudf.Cuda 中的静态变量
 
DefaultHostMemoryAllocator - ai.rapids.cudf 中的类
 
DefaultHostMemoryAllocator() - Class ai.rapids.cudf.DefaultHostMemoryAllocator 的构造方法
 
delRef() - Class ai.rapids.cudf.MemoryCleaner.Cleaner 中的方法
 
denseRank() - Class ai.rapids.cudf.GroupByScanAggregation 中的静态方法
获取行的密集排名。
denseRank() - Class ai.rapids.cudf.ScanAggregation 中的静态方法
获取行的密集排名。
desc(int) - Class ai.rapids.cudf.OrderByArg 中的静态方法
 
desc(int, boolean) - Class ai.rapids.cudf.OrderByArg 中的静态方法
 
DeviceMemoryBuffer - ai.rapids.cudf 中的类
此类表示 GPU 上某种形式的数据。
DeviceMemoryBufferView - ai.rapids.cudf 中的类
此类表示 GPU 上某种形式的数据。
DeviceMemoryBufferView(long, long) - Class ai.rapids.cudf.DeviceMemoryBufferView 的构造方法
 
deviceRead(long, DeviceMemoryBuffer, Cuda.Stream) - Class ai.rapids.cudf.DataSource 中的方法
从源的给定偏移量读取数据到目标。
deviceRead(long, DeviceMemoryBuffer, Cuda.Stream) - Class ai.rapids.cudf.MultiBufferDataSource 中的方法
 
deviceSynchronize() - Class ai.rapids.cudf.Cuda 中的静态方法
使用 cudaDeviceSynchronize 同步整个设备。
distinctCount(NullPolicy) - Class ai.rapids.cudf.ColumnView 中的方法
计算列中有多少行是互不相同的。
distinctCount() - Class ai.rapids.cudf.ColumnView 中的方法
计算列中有多少行是互不相同的。
distinctCount(NullEquality) - Class ai.rapids.cudf.Table 中的方法
计算表中有多少行是互不相同的。
distinctCount() - Class ai.rapids.cudf.Table 中的方法
计算表中有多少行是互不相同的。
div(BinaryOperable, DType) - Interface ai.rapids.cudf.BinaryOperable 中的方法
将一个向量除以另一个向量,并指定输出类型。
div(BinaryOperable) - Interface ai.rapids.cudf.BinaryOperable 中的方法
将一个向量除以另一个向量。
done() - Interface ai.rapids.cudf.HostBufferConsumer 中的方法
表示将不再提供缓冲区。
doneWithTheGpu(Table) - Interface ai.rapids.cudf.ArrowIPCWriterOptions.DoneOnGpu 中的方法
一个回调,用于指示表已从 GPU 上卸载,即使所有数据尚未写入,也可以关闭该表。
dropDuplicates(int[], Table.DuplicateKeepOption, boolean) - Class ai.rapids.cudf.Table 中的方法
将当前表的行复制到输出表中,使得忽略键列中的重复行(即,只复制重复行中的一行)。
dropListDuplicates() - Class ai.rapids.cudf.ColumnView 中的方法
通过复制当前 LIST 列中的元素并忽略重复元素来创建新的 LIST 列,生成一个每个列表只包含唯一元素的 LIST 列。
dropListDuplicatesWithKeysValues() - Class ai.rapids.cudf.ColumnView 中的方法
给定一个 LIST 列,其中每个元素是一个包含对的结构。
DType - ai.rapids.cudf 中的类
 
dtype - Class ai.rapids.cudf.JCudfSerialization.SerializedColumnHeader 中的变量
 
DType.DTypeEnum - ai.rapids.cudf 中的枚举
 
DURATION_DAYS - Class ai.rapids.cudf.DType 中的静态变量
 
DURATION_MICROSECONDS - Class ai.rapids.cudf.DType 中的静态变量
 
DURATION_MILLISECONDS - Class ai.rapids.cudf.DType 中的静态变量
 
DURATION_NANOSECONDS - Class ai.rapids.cudf.DType 中的静态变量
 
DURATION_SECONDS - Class ai.rapids.cudf.DType 中的静态变量
 
durationDaysFromBoxedInts(Integer...) - Class ai.rapids.cudf.ColumnVector 中的静态方法
从给定值创建新的向量。
durationDaysFromBoxedInts(Integer...) - Class ai.rapids.cudf.HostColumnVector 中的静态方法
从给定值创建新的向量。
durationDaysFromInt(int) - Class ai.rapids.cudf.Scalar 中的静态方法
返回 DURATION_DAYS 标量
durationDaysFromInt(Integer) - Class ai.rapids.cudf.Scalar 中的静态方法
返回 DURATION_DAYS 标量
durationDaysFromInts(int...) - Class ai.rapids.cudf.ColumnVector 中的静态方法
从给定值创建新的向量。
durationDaysFromInts(int...) - Class ai.rapids.cudf.HostColumnVector 中的静态方法
从给定值创建新的向量。
durationFromLong(DType, long) - Class ai.rapids.cudf.Scalar 中的静态方法
根据类型参数返回持续时间标量。
durationFromLong(DType, Long) - Class ai.rapids.cudf.Scalar 中的静态方法
根据类型参数返回持续时间标量。
durationMicroSecondsFromBoxedLongs(Long...) - Class ai.rapids.cudf.ColumnVector 中的静态方法
从给定值创建新的向量。
durationMicrosecondsFromBoxedLongs(Long...) - Class ai.rapids.cudf.HostColumnVector 中的静态方法
从给定值创建新的向量。
durationMicroSecondsFromLongs(long...) - Class ai.rapids.cudf.ColumnVector 中的静态方法
从给定值创建新的向量。
durationMicrosecondsFromLongs(long...) - Class ai.rapids.cudf.HostColumnVector 中的静态方法
从给定值创建新的向量。
durationMilliSecondsFromBoxedLongs(Long...) - Class ai.rapids.cudf.ColumnVector 中的静态方法
从给定值创建新的向量。
durationMillisecondsFromBoxedLongs(Long...) - Class ai.rapids.cudf.HostColumnVector 中的静态方法
从给定值创建新的向量。
durationMilliSecondsFromLongs(long...) - Class ai.rapids.cudf.ColumnVector 中的静态方法
从给定值创建新的向量。
durationMillisecondsFromLongs(long...) - Class ai.rapids.cudf.HostColumnVector 中的静态方法
从给定值创建新的向量。
durationNanoSecondsFromBoxedLongs(Long...) - Class ai.rapids.cudf.ColumnVector 中的静态方法
从给定值创建新的向量。
durationNanosecondsFromBoxedLongs(Long...) - Class ai.rapids.cudf.HostColumnVector 中的静态方法
从给定值创建新的向量。
durationNanoSecondsFromLongs(long...) - Class ai.rapids.cudf.ColumnVector 中的静态方法
从给定值创建新的向量。
durationNanosecondsFromLongs(long...) - Class ai.rapids.cudf.HostColumnVector 中的静态方法
从给定值创建新的向量。
durationSecondsFromBoxedLongs(Long...) - Class ai.rapids.cudf.ColumnVector 中的静态方法
从给定值创建新的向量。
durationSecondsFromBoxedLongs(Long...) - Class ai.rapids.cudf.HostColumnVector 中的静态方法
从给定值创建新的向量。
durationSecondsFromLongs(long...) - Class ai.rapids.cudf.ColumnVector 中的静态方法
从给定值创建新的向量。
durationSecondsFromLongs(long...) - Class ai.rapids.cudf.HostColumnVector 中的静态方法
从给定值创建新的向量。

E

empty(HostColumnVector.DataType) - Class ai.rapids.cudf.ColumnVector 中的静态方法
根据数据类型创建一个空列。
EMPTY - Class ai.rapids.cudf.DType 中的静态变量
 
emptyStructs(HostColumnVector.DataType, long) - Class ai.rapids.cudf.ColumnVector 中的静态方法
此方法正在演变,不稳定且目前仅用于测试。
emptyStructs(HostColumnVector.DataType, long) - Class ai.rapids.cudf.HostColumnVector 中的静态方法
 
endList() - Class ai.rapids.cudf.HostColumnVector.ColumnBuilder 中的方法
如果要构建列表列,可以获取 `builder.getChild(0)` 并追加到其后,然后在完成后调用 `endList`,追加到该 builder 的所有内容现在将位于下一个列表中。
endStruct() - Class ai.rapids.cudf.HostColumnVector.ColumnBuilder 中的方法
如果要构建结构列,可以获取每个子项 `builder.getChild(N)` 并全部追加到其后,然后在完成后调用 `endStruct` 以更新此 builder。
endsWith(Scalar) - Class ai.rapids.cudf.ColumnView 中的方法
检查列中的每个字符串是否以指定的比较字符串结尾,结果是一个布尔结果的并行列。
equals(Object) - Class ai.rapids.cudf.AggregationOverWindow 中的方法
 
equals(Object) - Class ai.rapids.cudf.DType 中的方法
 
equals(Object) - Class ai.rapids.cudf.GroupByAggregation 中的方法
 
equals(Object) - Class ai.rapids.cudf.GroupByAggregationOnColumn 中的方法
 
equals(Object) - Class ai.rapids.cudf.GroupByScanAggregation 中的方法
 
equals(Object) - Class ai.rapids.cudf.GroupByScanAggregationOnColumn 中的方法
 
equals(Object) - Class ai.rapids.cudf.HostUDFWrapper 中的方法
 
equals(Object) - Class ai.rapids.cudf.ReductionAggregation 中的方法
 
equals(Object) - Class ai.rapids.cudf.ReplacePolicyWithColumn 中的方法
 
equals(Object) - Class ai.rapids.cudf.RollingAggregation 中的方法
 
equals(Object) - Class ai.rapids.cudf.RollingAggregationOnColumn 中的方法
 
equals(Object) - Class ai.rapids.cudf.Scalar 中的方法
 
equals(Object) - Class ai.rapids.cudf.ScanAggregation 中的方法
 
equals(Object) - Class ai.rapids.cudf.SegmentedReductionAggregation 中的方法
 
equals(Object) - Class ai.rapids.cudf.WindowOptions 中的方法
 
equalTo(BinaryOperable, DType) - Interface ai.rapids.cudf.BinaryOperable 中的方法
this == rhs 1 为 true,0 为 false,输出转换为给定类型。
equalTo(BinaryOperable) - Interface ai.rapids.cudf.BinaryOperable 中的方法
this == rhs 1 为 true,0 为 false。
equalToNullAware(BinaryOperable, DType) - Interface ai.rapids.cudf.BinaryOperable 中的方法
类似于 equalTo,但 NULL == NULL 为 TRUE,NULL == 非 NULL 为 FALSE
equalToNullAware(BinaryOperable) - Interface ai.rapids.cudf.BinaryOperable 中的方法
类似于 equalTo,但 NULL == NULL 为 TRUE,NULL == 非 NULL 为 FALSE
Event() - Class ai.rapids.cudf.Cuda.Event 的构造方法
创建一个尽可能快的事件,禁用计时且不使用 blockingSync。
Event(boolean, boolean) - Class ai.rapids.cudf.Cuda.Event 的构造方法
创建一个用于 CUDA 同步的事件。
exp() - Class ai.rapids.cudf.ColumnView 中的方法
计算指数函数,输出类型与输入类型相同。
experimental() - Class ai.rapids.cudf.JSONOptions 中的方法
 
explode(int) - Class ai.rapids.cudf.Table 中的方法
展开列表列的元素。
explodeOuter(int) - Class ai.rapids.cudf.Table 中的方法
展开列表列的元素。
explodeOuterPosition(int) - Class ai.rapids.cudf.Table 中的方法
展开列表列的元素,保留任何 null 条目或空列表,并包含一个位置列。
explodePosition(int) - Class ai.rapids.cudf.Table 中的方法
展开列表列的元素,并包含一个位置列。
extractAllRecord(String, int) - Class ai.rapids.cudf.ColumnView 中的方法
已弃用。
extractAllRecord(RegexProgram, int) - Class ai.rapids.cudf.ColumnView 中的方法
提取与给定正则表达式程序模式匹配并对应于正则表达式组索引的所有字符串。
extractDateTimeComponent(DateTimeComponent) - Class ai.rapids.cudf.ColumnView 中的方法
从时间戳中提取特定的日期时间组件。
extractInt32Chunk(ColumnView, DType, int) - Class ai.rapids.cudf.Aggregation128Utils 中的静态方法
从 128 位值中提取 32 位块。
extractListElement(int) - Class ai.rapids.cudf.ColumnView 中的方法
对于此列中的每个列表,在给定索引处提取条目。
extractListElement(ColumnView) - Class ai.rapids.cudf.ColumnView 中的方法
对于此列中的每个列表,在索引列中指定的相应索引处提取条目。
extractRe(String) - Class ai.rapids.cudf.ColumnView 中的方法
已弃用。
extractRe(RegexProgram) - Class ai.rapids.cudf.ColumnView 中的方法
对于给定正则表达式程序中指定的每个捕获组,在表中返回一列。

F

filter(ColumnView) - Class ai.rapids.cudf.Table 中的方法
使用布尔值列作为掩码过滤此表,返回一个新表。
findAndReplaceAll(ColumnView, ColumnView) - Class ai.rapids.cudf.ColumnView 中的方法
返回一个向量,其中所有值 "oldValues[i]" 都被 "newValues[i]" 替换。
flags() - Class ai.rapids.cudf.RegexProgram 中的方法
获取用于创建此实例的正则表达式标志设置
flattenLists() - Class ai.rapids.cudf.ColumnView 中的方法
将列表的每个列表展开成一个列表。
flattenLists(boolean) - Class ai.rapids.cudf.ColumnView 中的方法
将列表的每个列表展开成一个列表。
FLOAT32 - Class ai.rapids.cudf.DType 中的静态变量
 
FLOAT64 - Class ai.rapids.cudf.DType 中的静态变量
 
floor() - Class ai.rapids.cudf.ColumnView 中的方法
计算下取整,输出类型与输入类型相同。
floorDiv(BinaryOperable, DType) - Interface ai.rapids.cudf.BinaryOperable 中的方法
将一个向量除以另一个向量,并计算结果的下取整,指定输出类型。
floorDiv(BinaryOperable) - Interface ai.rapids.cudf.BinaryOperable 中的方法
将一个向量除以另一个向量,并计算结果的下取整。
following(Scalar) - Class ai.rapids.cudf.WindowOptions.Builder 中的方法
设置范围窗口中当前行后的相对数量
free - Class ai.rapids.cudf.CudaMemInfo 中的变量
可用内存(字节)
freeFromFallbackPinnedPool(long, long) - Class ai.rapids.cudf.Rmm 中的静态方法
 
freeFromPinnedPool(long, long, long) - Class ai.rapids.cudf.Rmm 中的静态方法
 
freeZero() - Static method in class ai.rapids.cudf.Cuda
调用 cudaFree(0)。
fromArrow(DType, long, long, ByteBuffer, ByteBuffer, ByteBuffer) - Static method in class ai.rapids.cudf.ColumnVector
从传入的 Apache Arrow 字节缓冲区创建 ColumnVector。
fromBool(boolean) - Static method in class ai.rapids.cudf.Scalar
 
fromBool(Boolean) - Static method in class ai.rapids.cudf.Scalar
 
fromBooleans(boolean...) - Static method in class ai.rapids.cudf.ColumnVector
从给定值创建新的向量。
fromBoxedBooleans(Boolean...) - Static method in class ai.rapids.cudf.ColumnVector
从给定值创建新的向量。
fromBoxedBooleans(Boolean...) - Static method in class ai.rapids.cudf.HostColumnVector
从给定值创建新的向量。
fromBoxedBytes(Byte...) - Static method in class ai.rapids.cudf.ColumnVector
从给定值创建新的向量。
fromBoxedBytes(Byte...) - Static method in class ai.rapids.cudf.HostColumnVector
从给定值创建新的向量。
fromBoxedDoubles(Double...) - Static method in class ai.rapids.cudf.ColumnVector
从给定值创建新的向量。
fromBoxedDoubles(Double...) - Static method in class ai.rapids.cudf.HostColumnVector
从给定值创建新的向量。
fromBoxedFloats(Float...) - Static method in class ai.rapids.cudf.ColumnVector
从给定值创建新的向量。
fromBoxedFloats(Float...) - Static method in class ai.rapids.cudf.HostColumnVector
从给定值创建新的向量。
fromBoxedInts(Integer...) - Static method in class ai.rapids.cudf.ColumnVector
从给定值创建新的向量。
fromBoxedInts(Integer...) - Static method in class ai.rapids.cudf.HostColumnVector
从给定值创建新的向量。
fromBoxedLongs(Long...) - Static method in class ai.rapids.cudf.ColumnVector
从给定值创建新的向量。
fromBoxedLongs(Long...) - Static method in class ai.rapids.cudf.HostColumnVector
从给定值创建新的向量。
fromBoxedShorts(Short...) - Static method in class ai.rapids.cudf.ColumnVector
从给定值创建新的向量。
fromBoxedShorts(Short...) - Static method in class ai.rapids.cudf.HostColumnVector
从给定值创建新的向量。
fromBoxedUnsignedBytes(Byte...) - Static method in class ai.rapids.cudf.ColumnVector
从给定值创建新的向量。
fromBoxedUnsignedBytes(Byte...) - Static method in class ai.rapids.cudf.HostColumnVector
从给定值创建新的向量。
fromBoxedUnsignedInts(Integer...) - Static method in class ai.rapids.cudf.ColumnVector
从给定值创建新的向量。
fromBoxedUnsignedInts(Integer...) - Static method in class ai.rapids.cudf.HostColumnVector
从给定值创建新的向量。
fromBoxedUnsignedLongs(Long...) - Static method in class ai.rapids.cudf.ColumnVector
从给定值创建新的向量。
fromBoxedUnsignedLongs(Long...) - Static method in class ai.rapids.cudf.HostColumnVector
从给定值创建新的向量。
fromBoxedUnsignedShorts(Short...) - Static method in class ai.rapids.cudf.ColumnVector
从给定值创建新的向量。
fromBoxedUnsignedShorts(Short...) - Static method in class ai.rapids.cudf.HostColumnVector
从给定值创建新的向量。
fromByte(byte) - Static method in class ai.rapids.cudf.Scalar
 
fromByte(Byte) - Static method in class ai.rapids.cudf.Scalar
 
fromBytes(byte...) - Static method in class ai.rapids.cudf.ColumnVector
从给定值创建新的向量。
fromBytes(byte...) - Static method in class ai.rapids.cudf.HostColumnVector
从给定值创建新的向量。
fromDecimal(int, int) - Static method in class ai.rapids.cudf.Scalar
 
fromDecimal(int, long) - Static method in class ai.rapids.cudf.Scalar
 
fromDecimal(int, BigInteger) - Static method in class ai.rapids.cudf.Scalar
 
fromDecimal(BigDecimal) - Static method in class ai.rapids.cudf.Scalar
 
fromDecimal(BigInteger, DType) - Static method in class ai.rapids.cudf.Scalar
 
fromDecimals(BigDecimal...) - Static method in class ai.rapids.cudf.ColumnVector
从给定值创建新的向量。
fromDecimals(BigDecimal...) - Static method in class ai.rapids.cudf.HostColumnVector
从给定值创建新的向量。
fromDeviceBuffer(BaseDeviceMemoryBuffer, long, DType, int) - Static method in class ai.rapids.cudf.ColumnView
从原始设备缓冲区创建新的列视图。
fromDouble(double) - Static method in class ai.rapids.cudf.Scalar
 
fromDouble(Double) - Static method in class ai.rapids.cudf.Scalar
 
fromDoubles(double...) - Static method in class ai.rapids.cudf.ColumnVector
从给定值创建新的向量。
fromDoubles(double...) - Static method in class ai.rapids.cudf.HostColumnVector
从给定值创建新的向量。
fromFloat(float) - Static method in class ai.rapids.cudf.Scalar
 
fromFloat(Float) - Static method in class ai.rapids.cudf.Scalar
 
fromFloats(float...) - Static method in class ai.rapids.cudf.ColumnVector
从给定值创建新的向量。
fromFloats(float...) - Static method in class ai.rapids.cudf.HostColumnVector
从给定值创建新的向量。
fromInt(int) - Static method in class ai.rapids.cudf.Scalar
 
fromInt(Integer) - Static method in class ai.rapids.cudf.Scalar
 
fromInts(int...) - Static method in class ai.rapids.cudf.ColumnVector
从给定值创建新的向量。
fromInts(int...) - Static method in class ai.rapids.cudf.HostColumnVector
从给定值创建新的向量。
fromJavaBigDecimal(BigDecimal) - Static method in class ai.rapids.cudf.DType
使用 Java BigDecimal 的精度和小数位数创建类似 decimal 的 DType。
fromLists(HostColumnVector.DataType, List<T>...) - Static method in class ai.rapids.cudf.ColumnVector
此方法正在演变,不稳定且目前仅用于测试。
fromLists(HostColumnVector.DataType, List<T>...) - Static method in class ai.rapids.cudf.HostColumnVector
 
fromLong(long) - Static method in class ai.rapids.cudf.Scalar
 
fromLong(Long) - Static method in class ai.rapids.cudf.Scalar
 
fromLongs(long...) - Static method in class ai.rapids.cudf.ColumnVector
从给定值创建新的向量。
fromLongs(long...) - Static method in class ai.rapids.cudf.HostColumnVector
从给定值创建新的向量。
fromNative(int, int) - Static method in class ai.rapids.cudf.DType
DType 实例的工厂方法
fromNative(int) - Static method in enum ai.rapids.cudf.HashType
 
fromNative(int) - Static method in enum ai.rapids.cudf.PadSide
 
fromNativeId(int) - Static method in enum ai.rapids.cudf.nvcomp.CompressionType
查找与指定原生标识符对应的 CompressionType
fromNull(DType) - Static method in class ai.rapids.cudf.Scalar
 
fromPackedTable(ByteBuffer, DeviceMemoryBuffer) - Static method in class ai.rapids.cudf.Table
从打包表示构造表。
fromRmm(long, long, long) - Static method in class ai.rapids.cudf.DeviceMemoryBuffer
将现有 RMM 分配封装在设备内存缓冲区中。
fromScalar(Scalar, int) - Static method in class ai.rapids.cudf.ColumnVector
创建一个长度为 rows 的新向量,其中每行都填充 Scalar 的值
fromShort(short) - Static method in class ai.rapids.cudf.Scalar
 
fromShort(Short) - Static method in class ai.rapids.cudf.Scalar
 
fromShorts(short...) - Static method in class ai.rapids.cudf.ColumnVector
从给定值创建新的向量。
fromShorts(short...) - Static method in class ai.rapids.cudf.HostColumnVector
从给定值创建新的向量。
fromString(String) - Static method in class ai.rapids.cudf.Scalar
 
fromStrings(String...) - Static method in class ai.rapids.cudf.ColumnVector
从给定值创建新的字符串向量。
fromStrings(String...) - Static method in class ai.rapids.cudf.HostColumnVector
从给定值创建新的字符串向量。
fromStructs(HostColumnVector.DataType, List<HostColumnVector.StructData>) - Static method in class ai.rapids.cudf.ColumnVector
此方法正在演变,不稳定且目前仅用于测试。
fromStructs(HostColumnVector.DataType, HostColumnVector.StructData...) - Static method in class ai.rapids.cudf.ColumnVector
此方法正在演变,不稳定且目前仅用于测试。
fromStructs(HostColumnVector.DataType, List<HostColumnVector.StructData>) - Static method in class ai.rapids.cudf.HostColumnVector
 
fromStructs(HostColumnVector.DataType, HostColumnVector.StructData...) - Static method in class ai.rapids.cudf.HostColumnVector
 
fromUnsignedByte(byte) - Static method in class ai.rapids.cudf.Scalar
 
fromUnsignedByte(Byte) - Static method in class ai.rapids.cudf.Scalar
 
fromUnsignedBytes(byte...) - Static method in class ai.rapids.cudf.ColumnVector
从给定值创建新的向量。
fromUnsignedBytes(byte...) - Static method in class ai.rapids.cudf.HostColumnVector
从给定值创建新的向量。
fromUnsignedInt(int) - Static method in class ai.rapids.cudf.Scalar
 
fromUnsignedInt(Integer) - Static method in class ai.rapids.cudf.Scalar
 
fromUnsignedInts(int...) - Static method in class ai.rapids.cudf.ColumnVector
从给定值创建新的向量。
fromUnsignedInts(int...) - Static method in class ai.rapids.cudf.HostColumnVector
从给定值创建新的向量。
fromUnsignedLong(long) - Static method in class ai.rapids.cudf.Scalar
 
fromUnsignedLong(Long) - Static method in class ai.rapids.cudf.Scalar
 
fromUnsignedLongs(long...) - Static method in class ai.rapids.cudf.ColumnVector
从给定值创建新的向量。
fromUnsignedLongs(long...) - Static method in class ai.rapids.cudf.HostColumnVector
从给定值创建新的向量。
fromUnsignedShort(short) - Static method in class ai.rapids.cudf.Scalar
 
fromUnsignedShort(Short) - Static method in class ai.rapids.cudf.Scalar
 
fromUnsignedShorts(short...) - Static method in class ai.rapids.cudf.ColumnVector
从给定值创建新的向量。
fromUnsignedShorts(short...) - Static method in class ai.rapids.cudf.HostColumnVector
从给定值创建新的向量。
fromUTF8String(byte[]) - Static method in class ai.rapids.cudf.Scalar
从 UTF8 字节数组创建字符串标量。
fromUTF8Strings(byte[]...) - Static method in class ai.rapids.cudf.ColumnVector
从给定值创建新的字符串向量。
fromUTF8Strings(byte[]...) - Static method in class ai.rapids.cudf.HostColumnVector
从给定值创建新的字符串向量。
fromViewWithContiguousAllocation(long, DeviceMemoryBuffer) - Static method in class ai.rapids.cudf.ColumnVector
使用连续设备分配从原生 column_view 创建 ColumnVector。
fullJoinGatherMaps(Table, boolean) - Method in class ai.rapids.cudf.Table
计算可用于呈现两个表之间的完全等值连接结果的收集映射。
fullJoinGatherMaps(HashJoin) - Method in class ai.rapids.cudf.Table
计算可用于呈现两个表之间的完全等值连接结果的收集映射。
fullJoinGatherMaps(HashJoin, long) - Method in class ai.rapids.cudf.Table
计算可用于呈现两个表之间的完全等值连接结果的收集映射。
fullJoinRowCount(HashJoin) - Method in class ai.rapids.cudf.Table
计算两个表进行完全等值连接后产生的行数。

G

gather(ColumnView) - Method in class ai.rapids.cudf.Table
根据 `gatherMap` 收集此表的行,以便结果表中列的第 "i" 行将包含此表中 "gatherMap[i]" 行。
gather(ColumnView, OutOfBoundsPolicy) - Method in class ai.rapids.cudf.Table
根据 `gatherMap` 收集此表的行,以便结果表中列的第 "i" 行将包含此表中 "gatherMap[i]" 行。
GatherMap - Class in ai.rapids.cudf
此类跟踪与收集映射相关的数据,该映射是一个 INT32 元素的缓冲区,可索引源表并传递给表收集操作。
GatherMap(DeviceMemoryBuffer) - Constructor for class ai.rapids.cudf.GatherMap
从设备缓冲区构造收集映射实例。
generateListOffsets() - Method in class ai.rapids.cudf.ColumnView
根据每个列表的大小生成列表偏移。
get(int) - Method in class ai.rapids.cudf.CloseableArray
 
get() - Static method in class ai.rapids.cudf.DefaultHostMemoryAllocator
检索默认使用的当前主机内存分配器,如果未直接传递给 API,则使用此分配器
get() - Static method in class ai.rapids.cudf.TableDebug
 
getAddress() - Method in class ai.rapids.cudf.MemoryBuffer
返回此缓冲区指向的数据的位置
getAllocThresholds() - Method in interface ai.rapids.cudf.RmmEventHandler
获取内存阈值,当在内存分配期间超过一个或多个阈值时,这些阈值将触发调用 RmmEventHandler.onAllocThreshold(long)
getArray() - Method in class ai.rapids.cudf.CloseableArray
 
getBigDecimal(long) - Method in class ai.rapids.cudf.HostColumnVectorCore
获取索引处的 BigDecimal 值。
getBigDecimal() - Method in class ai.rapids.cudf.Scalar
以 BigDecimal 形式返回标量值。
getBigInteger() - Method in class ai.rapids.cudf.Scalar
以字节数组形式返回 BigDecimal 未缩放标量值。
getBoolean(long) - Method in class ai.rapids.cudf.HostColumnVectorCore
获取索引处的布尔值
getBoolean(long) - Method in class ai.rapids.cudf.HostMemoryBuffer
返回该偏移量处的布尔值
getBoolean() - Method in class ai.rapids.cudf.Scalar
以布尔值形式返回标量值。
getBuffer() - Method in class ai.rapids.cudf.ContiguousTable
获取支持连续表数据的设备缓冲区。
getByte(long) - Method in class ai.rapids.cudf.HostColumnVectorCore
获取索引处的值。
getByte(long) - Method in class ai.rapids.cudf.HostMemoryBuffer
返回该偏移量处的字节值
getByte() - Method in class ai.rapids.cudf.Scalar
以字节形式返回标量值。
getByteCount() - Method in class ai.rapids.cudf.ColumnView
检索每个字符串的字节数。
getBytes(byte[], long, long, long) - Method in class ai.rapids.cudf.HostMemoryBuffer
从缓冲区中从偏移量开始将一组字节复制到数组中。
getBytesFromList(long) - Method in class ai.rapids.cudf.HostColumnVectorCore
警告:int8 或 uint8 列表的特殊情况,不支持空列表值或列表 从 int8 或 uint8 列表列中获取索引处的字节数组。
getCallback() - Method in class ai.rapids.cudf.ArrowIPCOptions
 
getCallback() - Method in class ai.rapids.cudf.ArrowIPCWriterOptions
 
getCharLengths() - Method in class ai.rapids.cudf.ColumnView
检索每个字符串中的字符数。
getChild(int) - Method in class ai.rapids.cudf.HostColumnVector.BasicType
 
getChild(int) - Method in class ai.rapids.cudf.HostColumnVector.ColumnBuilder
 
getChild(int) - Method in class ai.rapids.cudf.HostColumnVector.DataType
 
getChild(int) - Method in class ai.rapids.cudf.HostColumnVector.ListType
 
getChild(int) - Method in class ai.rapids.cudf.HostColumnVector.StructType
 
getChild(int) - Method in class ai.rapids.cudf.Schema
获取子元素的模式。
getChild(int) - Method in class ai.rapids.cudf.TableWithMeta
 
getChild(int) - Method in class ai.rapids.cudf.TableWithMeta.NestedChildren
 
getChildColumnOptions() - Method in class ai.rapids.cudf.ColumnWriterOptions
返回此列的子 columnOptions
getChildColumnView(int) - Method in class ai.rapids.cudf.ColumnView
返回给定索引处的子列视图。
getChildColumnView(int) - Method in class ai.rapids.cudf.HostColumnVectorCore
 
getChildColumnViews() - Method in class ai.rapids.cudf.ColumnView
返回此视图的子列视图 请注意,调用者有责任关闭这些视图。
getChildren() - Method in class ai.rapids.cudf.JCudfSerialization.SerializedColumnHeader
获取任何子列的元数据,如果没有子列则返回 null
getChildrenFromStructScalar() - Method in class ai.rapids.cudf.Scalar
从 struct 标量中获取子列的视图。
getChildTypes() - Method in class ai.rapids.cudf.Schema
获取模式中顶级子列的类型。
getColumn(int) - Method in class ai.rapids.cudf.PartitionedTable
 
getColumn(int) - Method in class ai.rapids.cudf.Table
返回指定索引处的 ColumnVector
getColumnHeader(int) - Method in class ai.rapids.cudf.JCudfSerialization.SerializedTableHeader
获取相应列索引的列头
getColumnIndex() - Method in class ai.rapids.cudf.GroupByAggregationOnColumn
 
getColumnIndex() - Method in class ai.rapids.cudf.GroupByScanAggregationOnColumn
 
getColumnIndex() - Method in class ai.rapids.cudf.RollingAggregationOnColumn
 
getColumnName() - Method in class ai.rapids.cudf.ColumnWriterOptions
返回列是否可以包含 null 值
getColumnNames() - Method in class ai.rapids.cudf.CSVWriterOptions
 
getColumnNames() - Method in class ai.rapids.cudf.Schema
获取模式中顶级子列的名称。
getColumnNames() - Method in class ai.rapids.cudf.TableWithMeta
获取顶级列的名称。
getCompareNulls() - Method in class ai.rapids.cudf.HashJoin
如果哈希表是为匹配 null 而构建的,则返回 true,否则返回 false。
getCompressionType() - Method in class ai.rapids.cudf.CompressionMetadataWriterOptions
 
getComputeCapabilityMajor() - Static method in class ai.rapids.cudf.Cuda
获取当前设备的主 CUDA 计算能力。
getComputeCapabilityMinor() - Static method in class ai.rapids.cudf.Cuda
获取当前设备的次 CUDA 计算能力。
getComputeMode() - Static method in class ai.rapids.cudf.Cuda
获取当前设备的 CUDA 计算模式。
getContiguousTable() - Method in class ai.rapids.cudf.JCudfSerialization.TableAndRowCountPair
获取反序列化的 ContiguousTable,如果没有任何数据(例如:没有列的行)则返回 null。
getCSVBufferWriter(CSVWriterOptions, HostBufferConsumer, HostMemoryAllocator) - Static method in class ai.rapids.cudf.Table
 
getCSVBufferWriter(CSVWriterOptions, HostBufferConsumer) - Static method in class ai.rapids.cudf.Table
 
getCudaError() - Method in exception ai.rapids.cudf.CudaException
 
getCurrentByteIndex() - Method in class ai.rapids.cudf.HostColumnVector.ColumnBuilder
已弃用。
getCurrentDeviceResource() - Static method in class ai.rapids.cudf.Rmm
获取通过 Java API 最后设置的 RmmDeviceMemoryResource。
getCurrentIndex() - Method in class ai.rapids.cudf.HostColumnVector.ColumnBuilder
 
getData() - Method in class ai.rapids.cudf.ColumnVector.OffHeapState
 
getData() - Method in class ai.rapids.cudf.ColumnView
获取当前列视图 (viewHandle) 的数据缓冲区。
getData() - Method in class ai.rapids.cudf.HostColumnVectorCore
返回给定主机侧列向量的数据缓冲区
getDataLen() - Method in class ai.rapids.cudf.JCudfSerialization.SerializedTableHeader
返回将数据读入流所需的缓冲区大小。
getDeallocThresholds() - Method in interface ai.rapids.cudf.RmmEventHandler
获取内存阈值,当在内存释放期间超过一个或多个阈值时,这些阈值将触发调用 RmmEventHandler.onDeallocThreshold(long)
getDecimalMaxPrecision() - Method in class ai.rapids.cudf.DType
返回 Decimal 类型的最大精度。
getDevice() - Static method in class ai.rapids.cudf.Cuda
获取当前设备的 ID。
getDeviceBufferFor(BufferType) - Method in class ai.rapids.cudf.ColumnVector
获取此列的原始设备缓冲区的访问权限。
getDeviceCount() - Static method in class ai.rapids.cudf.Cuda
获取设备数量。
getDeviceMemorySize() - Method in class ai.rapids.cudf.ColumnView
返回使用的设备内存量。
getDeviceMemorySize() - Method in class ai.rapids.cudf.Table
返回设备内存缓冲区大小。
getDeviceReadCutoff() - Method in class ai.rapids.cudf.DataSource
获取支持设备读取时设备读取和主机读取之间的尺寸截止值。
getDevReadBytes() - Method in class ai.rapids.cudf.MultiBufferDataSource
 
getDevReads() - Method in class ai.rapids.cudf.MultiBufferDataSource
 
getDouble(long) - Method in class ai.rapids.cudf.HostColumnVectorCore
获取索引处的值。
getDouble(long) - Method in class ai.rapids.cudf.HostMemoryBuffer
返回该偏移量处的 Double 值
getDouble() - Method in class ai.rapids.cudf.Scalar
以 double 形式返回标量值。
getDriverVersion() - Static method in class ai.rapids.cudf.Cuda
获取 CUDA 驱动程序版本,这是驱动程序支持的最新 CUDA 版本。
getEndListOffset(long) - Method in class ai.rapids.cudf.HostColumnVectorCore
获取索引处列表或字符串的结束元素偏移量。
getEventHandler() - Method in class ai.rapids.cudf.ColumnVector
返回此 ColumnVector 的当前事件处理程序,如果未关联处理程序,则返回 null。
getEventHandler() - Method in class ai.rapids.cudf.HostColumnVector
返回此 HostColumnVector 的当前事件处理程序,如果未关联处理程序,则返回 null。
getEventHandler() - Method in class ai.rapids.cudf.MemoryBuffer
返回此缓冲区的当前事件处理程序,如果未关联处理程序或此缓冲区已关闭,则返回 null。
getFalseValue() - Method in class ai.rapids.cudf.CSVWriterOptions
 
getField(int) - Method in class ai.rapids.cudf.HostColumnVector.StructData
 
getFieldDelimiter() - Method in class ai.rapids.cudf.CSVWriterOptions
 
getFlatBooleans(boolean[], ColumnWriterOptions.ByteArrayProducer) - Method in class ai.rapids.cudf.ColumnWriterOptions
 
getFlatColumnNames(String[]) - Method in class ai.rapids.cudf.ColumnWriterOptions
 
getFlatInts(int[], ColumnWriterOptions.IntArrayProducer) - Method in class ai.rapids.cudf.ColumnWriterOptions
 
getFlattenedColumnNames() - Method in class ai.rapids.cudf.Schema
通过深度优先遍历获取模式中所有级别展平后的列名。
getFlattenedDecimalPrecisions() - Method in class ai.rapids.cudf.Schema
通过深度优先遍历获取模式中所有级别展平后的列类型的十进制精度。
getFlattenedNumChildren() - Method in class ai.rapids.cudf.Schema
获取模式中每个级别的子列数量。
getFlattenedTypeIds() - Method in class ai.rapids.cudf.Schema
通过深度优先遍历获取模式中所有级别展平后的列的类型 ID。
getFlattenedTypes() - Method in class ai.rapids.cudf.Schema
通过深度优先遍历获取模式中所有级别展平后的列类型。
getFlattenedTypeScales() - Method in class ai.rapids.cudf.Schema
通过深度优先遍历获取模式中所有级别展平后的列类型的小数位数(scale)。
getFloat(long) - Method in class ai.rapids.cudf.HostColumnVectorCore
获取索引处的值。
getFloat(long) - Method in class ai.rapids.cudf.HostMemoryBuffer
返回该偏移量处的 Float 值
getFloat() - Method in class ai.rapids.cudf.Scalar
以 float 形式返回标量值。
getGroups() - Method in class ai.rapids.cudf.ContigSplitGroupByResult
获取拆分的组表。
getHandle() - Method in class ai.rapids.cudf.RmmArenaMemoryResource
 
getHandle() - Method in class ai.rapids.cudf.RmmCudaAsyncMemoryResource
 
getHandle() - Method in class ai.rapids.cudf.RmmCudaMemoryResource
 
getHandle() - Method in interface ai.rapids.cudf.RmmDeviceMemoryResource
返回指向实现 rmm::mr::device_memory_resource 的底层 C++ 类的指针
getHandle() - Method in class ai.rapids.cudf.RmmEventHandlerResourceAdaptor
 
getHandle() - Method in class ai.rapids.cudf.RmmLimitingResourceAdaptor
 
getHandle() - Method in class ai.rapids.cudf.RmmLoggingResourceAdaptor
 
getHandle() - Method in class ai.rapids.cudf.RmmManagedMemoryResource
 
getHandle() - Method in class ai.rapids.cudf.RmmPoolMemoryResource
 
getHandle() - Method in class ai.rapids.cudf.RmmTrackingResourceAdaptor
 
getHostBuffer() - Method in class ai.rapids.cudf.JCudfSerialization.HostConcatResult
 
getHostBufferFor(BufferType) - Method in class ai.rapids.cudf.HostColumnVectorCore
获取此列的原始主机缓冲区的访问权限。
getHostBytesRequired() - Method in class ai.rapids.cudf.ColumnView
计算将数据复制到主机所需的总空间。
getHostMemorySize() - Method in class ai.rapids.cudf.HostColumnVectorCore
返回用于存储列/有效性数据(非元数据)的主机内存量。
getHostMemorySize() - Method in class ai.rapids.cudf.HostColumnVectorCore.OffHeapState
这返回为 ColumnVector 在主机上分配的总内存。
getHostReadBytes() - Method in class ai.rapids.cudf.MultiBufferDataSource
 
getHostReads() - Method in class ai.rapids.cudf.MultiBufferDataSource
 
getIncludeHeader() - Method in class ai.rapids.cudf.CSVWriterOptions
 
getInt(long) - Method in class ai.rapids.cudf.HostColumnVectorCore
获取索引处的值。
getInt(long) - Method in class ai.rapids.cudf.HostMemoryBuffer
返回该偏移量处的 Integer 值
getInt() - Method in class ai.rapids.cudf.Scalar
以 int 形式返回标量值。
getInts(int[], long, long, int) - Method in class ai.rapids.cudf.HostMemoryBuffer
从缓冲区中从偏移量开始将一组 ints 复制到数组中。
getJavaString(long) - Method in class ai.rapids.cudf.HostColumnVectorCore
获取索引处的值。
getJavaString() - Method in class ai.rapids.cudf.Scalar
以 Java 字符串形式返回标量值。
getJSONObject(Scalar, GetJsonObjectOptions) - Method in class ai.rapids.cudf.ColumnView
将 JSONPath 字符串应用于输入字符串列中的所有行。
getJSONObject(Scalar) - Method in class ai.rapids.cudf.ColumnView
将 JSONPath 字符串应用于输入字符串列中的所有行。
GetJsonObjectOptions - Class in ai.rapids.cudf
 
GetJsonObjectOptions.Builder - Class in ai.rapids.cudf
 
getLength() - Method in class ai.rapids.cudf.MemoryBuffer
返回此缓冲区的大小
getLineDelimiter() - Method in class ai.rapids.cudf.JSONOptions
 
getList(long) - Method in class ai.rapids.cudf.HostColumnVectorCore
警告:仅用于测试。
getListAsColumnView() - Method in class ai.rapids.cudf.Scalar
以 ColumnView 形式返回标量值。
getListOffsetsView() - Method in class ai.rapids.cudf.ColumnView
获取作为此列表偏移量的 ColumnView。
getLong(long) - Method in class ai.rapids.cudf.HostColumnVectorCore
获取索引处的值。
getLong(long) - Method in class ai.rapids.cudf.HostMemoryBuffer
返回该偏移量处的 Long 值
getLong() - Method in class ai.rapids.cudf.Scalar
以 long 形式返回标量值。
getLongs(long[], long, long, int) - Method in class ai.rapids.cudf.HostMemoryBuffer
从缓冲区中从偏移量开始将一组 longs 复制到数组中。
getMapKeyExistence(Scalar) - Method in class ai.rapids.cudf.ColumnView
对于 List 类型的列> 并传入 String 键,返回 struct 中所有键的布尔列,如果键存在于该行对应的 map 中则为 true,否则为 false。
getMapKeyExistence(ColumnView) - Method in class ai.rapids.cudf.ColumnView
对于 List 类型的列> 并传入键列,返回 map 中所有键的布尔列。
getMapValue(ColumnView) - Method in class ai.rapids.cudf.ColumnView
给定 List 类型的列> 并传入 X 类型的键列,返回 Y 类型的列,其中输出列中的每行是与 X 键对应的 Y 值。
getMapValue(Scalar) - Method in class ai.rapids.cudf.ColumnView
给定 List 类型的列> 并传入 X 类型的键,返回 Y 类型的列,其中输出列中的每行是与 X 键对应的 Y 值。
getMaxChunkSize() - Method in class ai.rapids.cudf.ArrowIPCWriterOptions
 
getMaximumTotalBytesAllocated() - Static method in class ai.rapids.cudf.Rmm
返回进程生命周期内未释放的最大 RMM 内存量(字节)。
getMaxSize() - Method in class ai.rapids.cudf.RmmPoolMemoryResource
 
getMaxTotalBytesAllocated() - Method in class ai.rapids.cudf.RmmTrackingResourceAdaptor
 
getMetadata() - Method in class ai.rapids.cudf.CompressionMetadataWriterOptions
 
getMetadataDirectBuffer() - Method in class ai.rapids.cudf.ContiguousTable
获取包含描述连续表模式和布局的主机元数据的字节缓冲区。
getMetadataDirectBuffer() - Method in class ai.rapids.cudf.PackedColumnMetadata
获取包含描述连续表模式和布局的主机元数据的字节缓冲区。
getNames() - Method in class ai.rapids.cudf.TableWithMeta.NestedChildren
 
getNativeHandle() - Method in class ai.rapids.cudf.ast.CompiledExpression
返回编译表达式的原生地址。
getNativeId() - Method in enum ai.rapids.cudf.DateTimeComponent
 
getNativeId() - Method in enum ai.rapids.cudf.DateTimeRoundingFrequency
 
getNativeId() - Method in enum ai.rapids.cudf.DType.DTypeEnum
 
getNativeId() - Method in enum ai.rapids.cudf.HashType
 
getNativeId() - Method in enum ai.rapids.cudf.PadSide
 
getNativeNullCount() - Method in class ai.rapids.cudf.ColumnVector.OffHeapState
 
getNativeStacktrace() - Method in exception ai.rapids.cudf.CudaException
 
getNativeStacktrace() - Method in exception ai.rapids.cudf.CudfException
 
getNativeView() - Method in class ai.rapids.cudf.ColumnView
请谨慎使用:此方法暴露原生 cudf::column_view 的地址。
getNativeView() - Method in class ai.rapids.cudf.Table
返回此表的原生表视图句柄
getNextIfAvailable() - Method in interface ai.rapids.cudf.StreamedTableReader
获取下一个可用表(如果存在)。
getNextIfAvailable(int) - Method in interface ai.rapids.cudf.StreamedTableReader
获取下一个可用表(如果存在)。
getNullCount() - Method in class ai.rapids.cudf.ColumnVector
返回数据中的 null 数量。
getNullCount() - Method in class ai.rapids.cudf.ColumnView
返回数据中的 null 数量。
getNullCount() - Method in class ai.rapids.cudf.HostColumnVectorCore
返回数据中的 null 数量。
getNullCount() - Method in class ai.rapids.cudf.JCudfSerialization.SerializedColumnHeader
获取列的 null 数量
getNullValue() - Method in class ai.rapids.cudf.CSVWriterOptions
 
getNumberOfColumns() - Method in class ai.rapids.cudf.HashJoin
获取用于生成哈希表的表的连接键列的数量。
getNumberOfColumns() - Method in class ai.rapids.cudf.PartitionedTable
 
getNumberOfColumns() - Method in class ai.rapids.cudf.Table
 
getNumChildren() - Method in class ai.rapids.cudf.ColumnView
 
getNumChildren() - Method in class ai.rapids.cudf.HostColumnVector.BasicType
 
getNumChildren() - Method in class ai.rapids.cudf.HostColumnVector.DataType
 
getNumChildren() - Method in class ai.rapids.cudf.HostColumnVector.ListType
 
getNumChildren() - Method in class ai.rapids.cudf.HostColumnVector.StructType
 
getNumChildren() - Method in class ai.rapids.cudf.HostColumnVectorCore
返回此列的子项数量
getNumChildren() - Method in class ai.rapids.cudf.JCudfSerialization.SerializedColumnHeader
获取子列的数量
getNumChildren() - Method in class ai.rapids.cudf.Schema
获取模式中顶级子列的数量。
getNumColumns() - Method in class ai.rapids.cudf.JCudfSerialization.SerializedTableHeader
返回此表中存储的列数
getNumFields() - Method in class ai.rapids.cudf.HostColumnVector.StructData
 
getNumRows() - Method in class ai.rapids.cudf.JCudfSerialization.SerializedTableHeader
返回此表中存储的行数。
getNumRows() - Method in class ai.rapids.cudf.JCudfSerialization.TableAndRowCountPair
获取已反序列化的行数。
getOffsets() - Method in class ai.rapids.cudf.ColumnVector.OffHeapState
 
getOffsets() - Method in class ai.rapids.cudf.ColumnView
 
getOffsets() - Method in class ai.rapids.cudf.HostColumnVectorCore
返回偏移量缓冲区
getOutputRowCount() - Method in class ai.rapids.cudf.MixedJoinSize
返回混合连接将生成的输出行数
getPartitions() - Method in class ai.rapids.cudf.PartitionedTable
此方法返回此表上的分区。
getPoolSize() - Static method in class ai.rapids.cudf.Rmm
获取最近设置的池大小,如果 RMM 未初始化或未启用池,则返回 -1。
getPrecision() - Method in class ai.rapids.cudf.ColumnWriterOptions
返回此列的精度
getPrecisionForInt() - Method in class ai.rapids.cudf.DType
获取保存整数类型所需的十进制位数。
getQuoteStyle() - Method in class ai.rapids.cudf.CSVWriterOptions
返回用于写入CSV的引用样式。
getRefCount() - Method in class ai.rapids.cudf.ColumnVector
返回此列当前的引用计数。
getRefCount() - Method in class ai.rapids.cudf.HostColumnVector
返回此列当前的引用计数。
getRefCount() - Method in class ai.rapids.cudf.MemoryBuffer
获取此缓冲区的当前引用计数。
getRowCount() - Method in class ai.rapids.cudf.ColumnView
返回此向量中的行数。
getRowCount() - Method in class ai.rapids.cudf.ContiguousTable
返回表中的行数。
getRowCount() - Method in class ai.rapids.cudf.GatherMap
返回 gather map 中的行数。
getRowCount() - Method in class ai.rapids.cudf.HostColumnVectorCore
返回给定主机端列向量的行数。
getRowCount() - Method in class ai.rapids.cudf.JCudfSerialization.SerializedColumnHeader
获取列的行数。
getRowCount() - Method in class ai.rapids.cudf.PartitionedTable
 
getRowCount() - Method in class ai.rapids.cudf.Table
 
getRowDelimiter() - Method in class ai.rapids.cudf.CSVWriterOptions
 
getRowGroupSizeBytes() - Method in class ai.rapids.cudf.ParquetWriterOptions
 
getRowGroupSizeRows() - Method in class ai.rapids.cudf.ParquetWriterOptions
 
getRuntimeVersion() - Static method in class ai.rapids.cudf.Cuda
获取当前CUDA运行时实例的CUDA运行时版本。
getScalarElement(int) - Method in class ai.rapids.cudf.ColumnView
以标量形式获取列中指定索引处的单个项。
getScalarHandle() - Method in class ai.rapids.cudf.Scalar
获取标量的原生句柄(原生指针地址)。
getScale() - Method in class ai.rapids.cudf.DType
返回Decimal类型的比例。
getScopedMaximumBytesAllocated() - Static method in class ai.rapids.cudf.Rmm
返回自上次调用 `resetScopedMaximumOutstanding` 以来未释放的RMM内存的最大量(字节)(它是“作用域内的”,因为它是自上次重置以来见到的最大量)。
getScopedMaxTotalBytesAllocated() - Method in class ai.rapids.cudf.RmmTrackingResourceAdaptor
 
getSerializedHeaderSizeInBytes() - Method in class ai.rapids.cudf.JCudfSerialization.SerializedColumnHeader
返回以序列化形式存储此列头部所需的字节数。
getSerializedHeaderSizeInBytes() - Method in class ai.rapids.cudf.JCudfSerialization.SerializedTableHeader
返回序列化此表头部所需的字节数。
getSerializedSizeInBytes(HostColumnVector[], long, long) - Static method in class ai.rapids.cudf.JCudfSerialization
获取序列化给定数据所需的字节大小。
getShort(long) - Method in class ai.rapids.cudf.HostColumnVectorCore
获取索引处的值。
getShort(long) - Method in class ai.rapids.cudf.HostMemoryBuffer
返回该偏移量处的Short值。
getShort() - Method in class ai.rapids.cudf.Scalar
返回标量值作为 short。
getSize() - Method in class ai.rapids.cudf.RmmArenaMemoryResource
 
getSize() - Method in class ai.rapids.cudf.RmmCudaAsyncMemoryResource
 
getSizeInBytes() - Method in class ai.rapids.cudf.DType
这只适用于固定宽度的类型。
getStartListOffset(long) - Method in class ai.rapids.cudf.HostColumnVectorCore
获取列表中或字符串在索引处的起始元素偏移量。
getStatisticsFrequency() - Method in class ai.rapids.cudf.ParquetWriterOptions
 
getStream() - Method in class ai.rapids.cudf.Cuda.Stream
 
getStripeSizeRows() - Method in class ai.rapids.cudf.ORCWriterOptions
 
getStruct(int) - Method in class ai.rapids.cudf.HostColumnVectorCore
警告:仅用于测试。
getTable() - Method in class ai.rapids.cudf.ContiguousTable
获取表实例,如有必要,从元数据中重建。
getTable() - Method in class ai.rapids.cudf.JCudfSerialization.TableAndRowCountPair
获取已反序列化的 Table,如果没有数据(例如:没有列的行),则返回 null。
getTable() - Method in class ai.rapids.cudf.PartitionedTable
 
getTableHeader() - Method in class ai.rapids.cudf.JCudfSerialization.HostConcatResult
 
getTopLevelChildren() - Method in class ai.rapids.cudf.CompressionMetadataWriterOptions
 
getTotalBytesAllocated() - Static method in class ai.rapids.cudf.Rmm
返回已分配的 RMM 内存量(字节)。
getTotalBytesAllocated() - Method in class ai.rapids.cudf.RmmTrackingResourceAdaptor
 
getTotalContiguousSize() - Method in class ai.rapids.cudf.ChunkedPack
获取我们正在打包的表的最终连续大小。
getTotalPoolSizeBytes() - Static method in class ai.rapids.cudf.PinnedMemoryPool
获取固定内存池分配的字节数。
getTotalSerializedSizeInBytes() - Method in class ai.rapids.cudf.JCudfSerialization.SerializedTableHeader
返回序列化此表头部和表数据所需的字节数。
getTracker() - Static method in class ai.rapids.cudf.Rmm
获取当前设置的 RmmTrackingResourceAdaptor。
getTrueValue() - Method in class ai.rapids.cudf.CSVWriterOptions
 
getType() - Method in interface ai.rapids.cudf.BinaryOperable
获取此数据的类型。
getType() - Method in class ai.rapids.cudf.ColumnView
 
getType() - Method in class ai.rapids.cudf.HostColumnVector.BasicType
 
getType() - Method in class ai.rapids.cudf.HostColumnVector.DataType
 
getType() - Method in class ai.rapids.cudf.HostColumnVector.ListType
 
getType() - Method in class ai.rapids.cudf.HostColumnVector.StructType
 
getType() - Method in class ai.rapids.cudf.HostColumnVectorCore
返回此向量的类型。
getType() - Method in class ai.rapids.cudf.JCudfSerialization.SerializedColumnHeader
获取列的数据类型。
getType() - Method in class ai.rapids.cudf.Scalar
 
getType() - Method in class ai.rapids.cudf.Schema
 
getTypeId() - Method in class ai.rapids.cudf.DType
返回此 DType 的枚举。
getUniqKeyTable() - Method in class ai.rapids.cudf.ContigSplitGroupByResult
获取键表,键表中的每一行对应一个组。
getUTF8(long) - Method in class ai.rapids.cudf.HostColumnVectorCore
获取索引处的原始 UTF8 字节。
getUTF8() - Method in class ai.rapids.cudf.Scalar
返回标量值作为 UTF-8 数据。
getValid() - Method in class ai.rapids.cudf.ColumnVector.OffHeapState
 
getValid() - Method in class ai.rapids.cudf.ColumnView
 
getValidity() - Method in class ai.rapids.cudf.HostColumnVectorCore
返回给定主机端列向量的有效性缓冲区。
getViewHandle() - Method in class ai.rapids.cudf.ColumnVector.OffHeapState
 
getWindowOptions() - Method in class ai.rapids.cudf.AggregationOverWindow
 
getWrapped() - Method in class ai.rapids.cudf.RmmWrappingDeviceMemoryResource
获取此对象正在包装的资源。
getWriteStatistics() - Method in class ai.rapids.cudf.TableWriter
获取截至上次写入调用时写入器的统计信息。
greaterOrEqualTo(BinaryOperable, DType) - Method in interface ai.rapids.cudf.BinaryOperable
this >= rhs,1 为 true,0 为 false,输出被强制转换为给定类型。
greaterOrEqualTo(BinaryOperable) - Method in interface ai.rapids.cudf.BinaryOperable
this >= rhs,1 为 true,0 为 false。
greaterThan(BinaryOperable, DType) - Method in interface ai.rapids.cudf.BinaryOperable
this > rhs,1 为 true,0 为 false,输出被强制转换为给定类型。
greaterThan(BinaryOperable) - Method in interface ai.rapids.cudf.BinaryOperable
this > rhs,1 为 true,0 为 false。
greaterThan(ColumnView, BigDecimal) - Static method in class ai.rapids.cudf.DecimalUtils
因为原生 greaterThan 运算符在精确比较具有不同精度和比例的十进制值时存在问题。
groupBy(GroupByOptions, int...) - Method in class ai.rapids.cudf.Table
返回按索引中提供的列分组的聚合操作。
groupBy(int...) - Method in class ai.rapids.cudf.Table
返回按索引中提供的列分组的聚合操作,使用如下默认选项:- 分组时 null 被视为键。
GroupByAggregation - Class in ai.rapids.cudf
可用于 reduce 的聚合。
GroupByAggregationOnColumn - Class in ai.rapids.cudf
表中特定列的 GroupByAggregation。
GroupByOptions - Class in ai.rapids.cudf
Groupby 的选项(参见 cudf::groupby::groupby 的构造函数)。
GroupByOptions.Builder - Class in ai.rapids.cudf
 
GroupByScanAggregation - Class in ai.rapids.cudf
可用于分组扫描的聚合。
GroupByScanAggregationOnColumn - Class in ai.rapids.cudf
表中特定列的 GroupByScanAggregation。

H

handleBuffer(HostMemoryBuffer, long) - Method in interface ai.rapids.cudf.HostBufferConsumer
消费一个缓冲区。
hasCompleted() - Method in class ai.rapids.cudf.Cuda.Event
检查事件是否已完成。
hashCode() - Method in class ai.rapids.cudf.AggregationOverWindow
 
hashCode() - Method in class ai.rapids.cudf.DType
 
hashCode() - Method in class ai.rapids.cudf.GroupByAggregation
 
hashCode() - Method in class ai.rapids.cudf.GroupByAggregationOnColumn
 
hashCode() - Method in class ai.rapids.cudf.GroupByScanAggregation
 
hashCode() - Method in class ai.rapids.cudf.GroupByScanAggregationOnColumn
 
hashCode() - Method in class ai.rapids.cudf.HostUDFWrapper
 
hashCode() - Method in class ai.rapids.cudf.ReductionAggregation
 
hashCode() - Method in class ai.rapids.cudf.ReplacePolicyWithColumn
 
hashCode() - Method in class ai.rapids.cudf.RollingAggregation
 
hashCode() - Method in class ai.rapids.cudf.RollingAggregationOnColumn
 
hashCode() - Method in class ai.rapids.cudf.Scalar
 
hashCode() - Method in class ai.rapids.cudf.ScanAggregation
 
hashCode() - Method in class ai.rapids.cudf.SegmentedReductionAggregation
 
hashCode() - Method in class ai.rapids.cudf.WindowOptions
 
hasHeader(boolean) - Method in class ai.rapids.cudf.CSVOptions.Builder
如果 hasHeader 为 true,则将头部行设置为 0(第一行),否则禁用头部。
hasHeader() - Method in class ai.rapids.cudf.CSVOptions.Builder
将头部行设置为 0(第一行)。
HashJoin - Class in ai.rapids.cudf
此类表示从连接操作中右侧表的连接键构建的哈希表。
HashJoin(Table, boolean) - Constructor for class ai.rapids.cudf.HashJoin
从表示连接中右侧表连接键列的表中,为连接构造一个哈希表。
hashPartition(int) - Method in class ai.rapids.cudf.Table.TableOperation
将表哈希分区到指定数量的分区。
hashPartition(HashType, int) - Method in class ai.rapids.cudf.Table.TableOperation
将表哈希分区到指定数量的分区。
hashPartition(HashType, int, int) - Method in class ai.rapids.cudf.Table.TableOperation
将表哈希分区到指定数量的分区。
HashType - Enum in ai.rapids.cudf
哈希算法标识符,镜像原生枚举 cudf::hash_id。
hasNestedChildren() - Method in class ai.rapids.cudf.Schema
这主要用于顶级结构体 schema,其中它本身是嵌套的,但对于 CSV 等情况,我们关注它没有任何也是嵌套的子项。
hasNext() - Method in class ai.rapids.cudf.ChunkedPack
调用此方法以确保 `ChunkedPack` 仍有工作待处理。
hasNext() - Method in class ai.rapids.cudf.ORCChunkedReader
检查给定文件是否还有内容可读。
hasNext() - Method in class ai.rapids.cudf.ParquetChunkedReader
检查给定文件是否还有内容可读。
hasNonEmptyNulls() - Method in class ai.rapids.cudf.ColumnView
精确检查列或其后代是否包含非空 null 行。
hasNulls() - Method in class ai.rapids.cudf.ColumnVector
返回向量是否包含 null 值。
hasNulls() - Method in class ai.rapids.cudf.HostColumnVectorCore
返回向量是否包含 null 值。
hasOffsets() - Method in class ai.rapids.cudf.DType
如果该类型使用偏移量向量,则返回 true。
hasParquetFieldId - Variable in class ai.rapids.cudf.ColumnWriterOptions.NestedBuilder
 
hasTimeResolution() - Method in class ai.rapids.cudf.DType
对于具有时间级别分辨率的时间戳返回 true,与日级别分辨率相反。
hasValidityVector() - Method in class ai.rapids.cudf.ColumnVector
返回向量是否已分配有效性向量。
hasValidityVector() - Method in class ai.rapids.cudf.HostColumnVectorCore
返回向量是否已分配有效性向量。
histogram() - Static method in class ai.rapids.cudf.GroupByAggregation
直方图聚合,计算每行唯一值的频率。
histogram() - Static method in class ai.rapids.cudf.ReductionAggregation
创建 HistogramAggregation,计算每行唯一值的频率。
HostBufferConsumer - Interface in ai.rapids.cudf
提供用于消费主机缓冲区的一组API。
HostBufferProvider - Interface in ai.rapids.cudf
提供用于提供主机缓冲区以供读取的一组API。
HostColumnVector - Class in ai.rapids.cudf
类似于 ColumnVector,但数据存储在主机内存中,可直接从 JVM 访问。
HostColumnVector(DType, long, Optional<Long>, HostMemoryBuffer, HostMemoryBuffer, HostMemoryBuffer, List<HostColumnVectorCore>) - Constructor for class ai.rapids.cudf.HostColumnVector
创建一个新的列向量,数据填充在主机上。
HostColumnVector.BasicType - Class in ai.rapids.cudf
 
HostColumnVector.Builder - Class in ai.rapids.cudf
 
HostColumnVector.ColumnBuilder - Class in ai.rapids.cudf
Build
HostColumnVector.DataType - Class in ai.rapids.cudf
 
HostColumnVector.EventHandler - Interface in ai.rapids.cudf
用于处理此 HostColumnVector 事件的接口。
HostColumnVector.ListType - Class in ai.rapids.cudf
 
HostColumnVector.StructData - Class in ai.rapids.cudf
 
HostColumnVector.StructType - Class in ai.rapids.cudf
 
HostColumnVectorCore - Class in ai.rapids.cudf
一个类,包含主机端列向量 API 和 OffHeapState。
HostColumnVectorCore(DType, long, Optional<Long>, HostMemoryBuffer, HostMemoryBuffer, HostMemoryBuffer, List<HostColumnVectorCore>) - Constructor for class ai.rapids.cudf.HostColumnVectorCore
 
HostColumnVectorCore.OffHeapState - Class in ai.rapids.cudf
保存列向量的堆外状态,以便即使发生内存泄漏也能进行清理。
HostConcatResult(JCudfSerialization.SerializedTableHeader, HostMemoryBuffer) - Constructor for class ai.rapids.cudf.JCudfSerialization.HostConcatResult
 
HostMemoryAllocator - Interface in ai.rapids.cudf
 
HostMemoryBuffer - Class in ai.rapids.cudf
此类在主机/CPU内存中持有一个堆外缓冲区。
hostPaddingSizeInBytes() - Static method in class ai.rapids.cudf.ColumnView
获取主机将内存分配对齐的字节大小。
hostRead(long, long) - Method in class ai.rapids.cudf.DataSource
从源的给定偏移量读取数据。
hostRead(long, HostMemoryBuffer) - Method in class ai.rapids.cudf.DataSource
从源的给定偏移量读取数据到目标。
hostRead(long, long) - Method in class ai.rapids.cudf.MultiBufferDataSource
 
hostRead(long, HostMemoryBuffer) - Method in class ai.rapids.cudf.MultiBufferDataSource
 
hostUDF(HostUDFWrapper) - Static method in class ai.rapids.cudf.GroupByAggregation
使用主机端用户定义函数 (UDF) 执行聚合。
hostUDF(HostUDFWrapper) - Static method in class ai.rapids.cudf.ReductionAggregation
使用主机端用户定义函数 (UDF) 执行归约。
hostUDF(HostUDFWrapper) - Static method in class ai.rapids.cudf.SegmentedReductionAggregation
使用主机端用户定义函数 (UDF) 执行归约。
HostUDFWrapper - Class in ai.rapids.cudf
原生主机 UDF 聚合的包装器。
HostUDFWrapper() - Constructor for class ai.rapids.cudf.HostUDFWrapper
 
hour() - Method in class ai.rapids.cudf.ColumnView
从具有时间分辨率的时间戳中获取小时。

I

id - Variable in class ai.rapids.cudf.MemoryBuffer
 
id - Variable in class ai.rapids.cudf.MemoryCleaner.Cleaner
 
ifElse(ColumnView, ColumnView) - Method in class ai.rapids.cudf.ColumnView
对于 BOOL8 向量,根据此向量对应行中的布尔值,计算一个其行从另外两个向量中选择的向量。
ifElse(ColumnView, Scalar) - Method in class ai.rapids.cudf.ColumnView
对于 BOOL8 向量,根据此向量对应行中的布尔值,计算一个其行从另外两个输入中选择的向量。
ifElse(Scalar, ColumnView) - Method in class ai.rapids.cudf.ColumnView
对于 BOOL8 向量,根据此向量对应行中的布尔值,计算一个其行从另外两个输入中选择的向量。
ifElse(Scalar, Scalar) - Method in class ai.rapids.cudf.ColumnView
对于 BOOL8 向量,根据此向量对应行中的布尔值,计算一个其行从另外两个输入中选择的向量。
implicitConversion(BinaryOp, BinaryOperable, BinaryOperable) - Static method in interface ai.rapids.cudf.BinaryOperable
找到隐式输出的正确 DType。
includeColumn(String...) - Method in class ai.rapids.cudf.ColumnFilterOptions.Builder
包含一个或多个特定列。
includeColumn(Collection<String>) - Method in class ai.rapids.cudf.ColumnFilterOptions.Builder
包含一个或多个特定列。
includeColumn(String...) - Method in class ai.rapids.cudf.JSONOptions.Builder
 
includeColumn(Collection<String>) - Method in class ai.rapids.cudf.JSONOptions.Builder
 
includeColumn(String...) - Method in class ai.rapids.cudf.ParquetOptions.Builder
包含一个或多个特定列。
includeColumn(String, boolean) - Method in class ai.rapids.cudf.ParquetOptions.Builder
包含此列。
includeColumn(Collection<String>) - Method in class ai.rapids.cudf.ParquetOptions.Builder
包含一个或多个特定列。
incrCurrentIndex() - Method in class ai.rapids.cudf.HostColumnVector.ColumnBuilder
已弃用。
incRefCount() - Method in class ai.rapids.cudf.ColumnVector
增加此列的引用计数。
incRefCount() - Method in class ai.rapids.cudf.HostColumnVector
增加此列的引用计数。
incRefCount() - Method in class ai.rapids.cudf.MemoryBuffer
增加此列的引用计数。
incRefCount() - Method in class ai.rapids.cudf.Scalar
增加此标量的引用计数。
incRefCount() - Method in class ai.rapids.cudf.WindowOptions
 
INFERRED - Static variable in class ai.rapids.cudf.Schema
 
initDefaultCudaDevice() - Static method in class ai.rapids.cudf.Rmm
 
initialize(long) - Static method in class ai.rapids.cudf.PinnedMemoryPool
初始化池。
initialize(long, int) - Static method in class ai.rapids.cudf.PinnedMemoryPool
初始化池。
initialize(long, int, boolean) - Static method in class ai.rapids.cudf.PinnedMemoryPool
初始化池。
initialize(int, Rmm.LogConf, long) - Static method in class ai.rapids.cudf.Rmm
初始化内存管理器状态和存储。
innerDistinctJoinGatherMaps(Table, boolean) - Method in class ai.rapids.cudf.Table
计算可用于显示两个表之间内等值连接结果的 gather map,其中右表保证不包含任何重复的连接键。
innerJoinGatherMaps(Table, boolean) - Method in class ai.rapids.cudf.Table
计算可用于显示两个表之间内等值连接结果的 gather map。
innerJoinGatherMaps(HashJoin) - Method in class ai.rapids.cudf.Table
计算可用于显示两个表之间内等值连接结果的 gather map。
innerJoinGatherMaps(HashJoin, long) - Method in class ai.rapids.cudf.Table
计算可用于显示两个表之间内等值连接结果的 gather map。
innerJoinRowCount(HashJoin) - Method in class ai.rapids.cudf.Table
计算两个表之间内等值连接产生的行数。
INT16 - Static variable in class ai.rapids.cudf.DType
 
INT32 - Static variable in class ai.rapids.cudf.DType
 
INT64 - Static variable in class ai.rapids.cudf.DType
 
INT8 - Static variable in class ai.rapids.cudf.DType
 
interleaveColumns() - Method in class ai.rapids.cudf.Table
将所有列交错为一个单一列。
isAllowSingleQuotes() - Method in class ai.rapids.cudf.GetJsonObjectOptions
 
isBackedByByte() - Method in class ai.rapids.cudf.DType
如果此类型由字节类型支持,则返回 true。具体而言,此方法对以下类型返回 true:DType.INT8, DType.UINT8, DType.BOOL8。
isBackedByInt() - Method in class ai.rapids.cudf.DType
如果此类型由 int 类型支持,则返回 true。具体而言,此方法对以下类型返回 true:DType.INT32, DType.UINT32, DType.DURATION_DAYS, DType.TIMESTAMP_DAYS, DType.DECIMAL32。
isBackedByLong() - Method in class ai.rapids.cudf.DType
如果此类型由 long 类型支持,则返回 true。具体而言,此方法对以下类型返回 true:DType.INT64, DType.UINT64, DType.DURATION_SECONDS, DType.DURATION_MILLISECONDS, DType.DURATION_MICROSECONDS, DType.DURATION_NANOSECONDS, DType.TIMESTAMP_SECONDS, DType.TIMESTAMP_MILLISECONDS, DType.TIMESTAMP_MICROSECONDS, DType.TIMESTAMP_NANOSECONDS, DType.DECIMAL64。
isBackedByShort() - Method in class ai.rapids.cudf.DType
如果此类型由 short 类型支持,则返回 true。具体而言,此方法对以下类型返回 true:DType.INT16, DType.UINT16。
isChildNested(int) - Method in class ai.rapids.cudf.TableWithMeta
 
isChildNested(int) - Method in class ai.rapids.cudf.TableWithMeta.NestedChildren
 
isClean() - Method in class ai.rapids.cudf.ColumnVector.OffHeapState
 
isClean() - Method in class ai.rapids.cudf.HostColumnVectorCore.OffHeapState
 
isClean() - Method in class ai.rapids.cudf.MemoryCleaner.Cleaner
检查底层内存是否已清理。
isDayFirst() - Method in class ai.rapids.cudf.JSONOptions
 
isDecimalType() - Method in enum ai.rapids.cudf.DType.DTypeEnum
 
isDecimalType() - Method in class ai.rapids.cudf.DType
如果此类型为 decimal 类型,则返回 true。具体而言,此方法对以下类型返回 true:DType.DECIMAL32, DType.DECIMAL64。
isDurationType() - Method in class ai.rapids.cudf.DType
对于 duration 类型返回 true。
isEnvCompatibleForTesting() - Static method in class ai.rapids.cudf.Cuda
这只应用于测试,用于在当前环境与此库版本不兼容时启用或禁用测试。
isEqual(Object) - Method in class ai.rapids.cudf.HostUDFWrapper
 
isFixedPoint(DType) - Method in class ai.rapids.cudf.ColumnView
返回一个布尔向量,其行数与此实例相同,对于任何固定点条目为 TRUE,如果不是固定点则为 FALSE。
isFloat() - Method in class ai.rapids.cudf.ColumnView
返回一个布尔向量,其行数与此实例相同,对于任何浮点条目为 TRUE,如果不是浮点则为 FALSE。
isInitialized() - Static method in class ai.rapids.cudf.PinnedMemoryPool
检查池是否已初始化。
isInitialized() - Static method in class ai.rapids.cudf.Rmm
检查 RMM 是否已初始化。
isInteger() - Method in class ai.rapids.cudf.ColumnView
返回一个布尔向量,其行数与此实例相同,对于任何整数条目为 TRUE,如果不是整数则为 FALSE。
isInteger(DType) - Method in class ai.rapids.cudf.ColumnView
返回一个布尔向量,其行数与此实例相同,对于任何整数条目为 TRUE,如果不是整数则为 FALSE。
isIntegral() - Method in class ai.rapids.cudf.DType
对于严格的整数类型返回 true,而不是由 int 支持的类型。
isLeakExpected() - Method in class ai.rapids.cudf.MemoryCleaner.Cleaner
如果此对象预计会发生内存泄漏则返回 true,否则返回 false。
isLeapYear() - Method in class ai.rapids.cudf.ColumnView
检查此时间戳的年份是否为闰年。
isLines() - Method in class ai.rapids.cudf.JSONOptions
 
isMissingFieldsAsNulls() - Method in class ai.rapids.cudf.GetJsonObjectOptions
 
isMixedTypesAsStrings() - Method in class ai.rapids.cudf.JSONOptions
 
isNan() - Method in class ai.rapids.cudf.ColumnView
返回一个布尔向量,其行数与此实例相同,对于任何 NaN 条目为 TRUE,如果为 null 或有效的浮点值则为 FALSE。
isNested() - Method in class ai.rapids.cudf.Schema
检查 schema 是否嵌套(即顶级类型是 LIST 或 STRUCT)。
isNestedType() - Method in class ai.rapids.cudf.DType
对于嵌套类型返回 true。
isNormalizeSingleQuotes() - Method in class ai.rapids.cudf.JSONOptions
 
isNormalizeWhitespace() - Method in class ai.rapids.cudf.JSONOptions
 
isNotNan() - Method in class ai.rapids.cudf.ColumnView
返回一个布尔向量,其行数与此实例相同,对于任何为 null 或有效的浮点值条目为 TRUE,否则为 FALSE。
isNotNull() - Method in class ai.rapids.cudf.ColumnView
返回一个布尔向量,其行数与此实例相同,对于任何非 null 条目为 TRUE,对于任何 null 条目(根据有效性掩码)为 FALSE。
isNull() - Method in class ai.rapids.cudf.ColumnView
返回一个布尔向量,其行数与此实例相同,对于任何非 null 条目为 FALSE,对于任何 null 条目(根据有效性掩码)为 TRUE。
isNull() - Method in class ai.rapids.cudf.HostColumnVector.StructData
 
isNull(long) - Method in class ai.rapids.cudf.HostColumnVectorCore
返回布尔值以指示给定行索引处的元素是否为 null 的方法。
isNullable() - Method in class ai.rapids.cudf.ColumnWriterOptions
返回列是否可以包含 null 值
isNullable - Variable in class ai.rapids.cudf.ColumnWriterOptions.NestedBuilder
 
isNullable() - Method in class ai.rapids.cudf.HostColumnVector.BasicType
 
isNullable() - Method in class ai.rapids.cudf.HostColumnVector.DataType
 
isNullable() - Method in class ai.rapids.cudf.HostColumnVector.ListType
 
isNullable() - Method in class ai.rapids.cudf.HostColumnVector.StructType
 
isPoolingEnabled() - Static method in class ai.rapids.cudf.Rmm
如果 rmm 已初始化且已启用池,则返回 true,否则返回 false。
isPtdsEnabled() - Static method in class ai.rapids.cudf.Cuda
是否启用每线程默认流。
isRecoverWithNull() - Method in class ai.rapids.cudf.JSONOptions
返回 recoverWithNull 选项的值。
isStripQuotesFromSingleStrings() - Method in class ai.rapids.cudf.GetJsonObjectOptions
 
isStructOrHasStructDescendant() - Method in class ai.rapids.cudf.Schema
检查 schema 是否包含结构体。
isTimestamp(String) - Method in class ai.rapids.cudf.ColumnView
验证字符串列是否可以使用提供的格式模式解析为时间戳。
isTimestamp() - Method in class ai.rapids.cudf.DType
已弃用。
isTimestampType() - Method in class ai.rapids.cudf.DType
 
isTimestampTypeInt96() - Method in class ai.rapids.cudf.ColumnWriterOptions
如果 writer 预计以 INT96 写入时间戳,则返回 true。
isValid() - Method in class ai.rapids.cudf.Scalar
 

J

JCudfSerialization - 类在 ai.rapids.cudf
使用自定义格式序列化和反序列化 CUDF 表和列。
JCudfSerialization() - 类 ai.rapids.cudf.JCudfSerialization 的构造函数
 
JCudfSerialization.HostConcatResult - 类在 ai.rapids.cudf
用于保存主机端连接产生的头信息和缓冲区对结果的类
JCudfSerialization.SerializedColumnHeader - 类在 ai.rapids.cudf
保存关于序列化列的元数据。
JCudfSerialization.SerializedTableHeader - 类在 ai.rapids.cudf
保存关于序列化表的元数据。
JCudfSerialization.TableAndRowCountPair - 类在 ai.rapids.cudf
保存反序列化表的结果。
joinStrings(Scalar, Scalar) - 类 ai.rapids.cudf.ColumnView 中的方法
将列中的所有字符串连接成一个新字符串,可选地使用分隔符字符串分隔。
JSONOptions - 类在 ai.rapids.cudf
读取 JSON 编码数据的选项。
JSONOptions.Builder - 类在 ai.rapids.cudf
 

K

keepStringQuotes() - 类 ai.rapids.cudf.JSONOptions 中的方法
 

L

lag(int) - 类 ai.rapids.cudf.RollingAggregation 中的静态方法
在滚动窗口中,返回偏移量位于当前项后或超出窗口的值,否则返回 null。
lag(int, ColumnVector) - 类 ai.rapids.cudf.RollingAggregation 中的静态方法
在滚动窗口中,返回偏移量位于当前项后或超出窗口的值,否则返回 defaultOutput 中的对应值。
lastDayOfMonth() - 类 ai.rapids.cudf.ColumnView 中的方法
获取此时间戳所在月份的最后一天日期。
lead(int) - 类 ai.rapids.cudf.RollingAggregation 中的静态方法
在滚动窗口中,返回偏移量位于当前项前或超出窗口的值,否则返回 null。
lead(int, ColumnVector) - 类 ai.rapids.cudf.RollingAggregation 中的静态方法
在滚动窗口中,返回偏移量位于当前项前或超出窗口的值,否则返回 defaultOutput 中的对应值。
leadingZerosAllowed() - 类 ai.rapids.cudf.JSONOptions 中的方法
 
leftAntiJoinGatherMap(Table, boolean) - 类 ai.rapids.cudf.Table 中的方法
计算可用于显现两个表之间左反连接结果的收集映射。
leftDistinctJoinGatherMap(Table, boolean) - 类 ai.rapids.cudf.Table 中的方法
计算可用于显现两个表之间左等值连接结果的收集映射,其中保证右表不包含任何重复的连接键。
leftJoinGatherMaps(Table, boolean) - 类 ai.rapids.cudf.Table 中的方法
计算可用于显现两个表之间左等值连接结果的收集映射。
leftJoinGatherMaps(HashJoin) - 类 ai.rapids.cudf.Table 中的方法
计算可用于显现两个表之间左等值连接结果的收集映射。
leftJoinGatherMaps(HashJoin, long) - 类 ai.rapids.cudf.Table 中的方法
计算可用于显现两个表之间左等值连接结果的收集映射。
leftJoinRowCount(HashJoin) - 类 ai.rapids.cudf.Table 中的方法
计算两个表之间左等值连接产生的行数。
leftSemiJoinGatherMap(Table, boolean) - 类 ai.rapids.cudf.Table 中的方法
计算可用于显现两个表之间左半连接结果的收集映射。
length - 类 ai.rapids.cudf.MemoryBuffer 中的变量
 
lessOrEqualTo(BinaryOperable, DType) - 接口 ai.rapids.cudf.BinaryOperable 中的方法
this <= rhs,如果为 true 则为 1,否则为 0,并将输出强制转换为给定类型。
lessOrEqualTo(BinaryOperable) - 接口 ai.rapids.cudf.BinaryOperable 中的方法
this <= rhs,如果为 true 则为 1,否则为 0。
lessThan(BinaryOperable, DType) - 接口 ai.rapids.cudf.BinaryOperable 中的方法
this < rhs,如果为 true 则为 1,否则为 0,并将输出强制转换为给定类型。
lessThan(BinaryOperable) - 接口 ai.rapids.cudf.BinaryOperable 中的方法
this < rhs,如果为 true 则为 1,否则为 0。
lessThan(ColumnView, BigDecimal) - 类 ai.rapids.cudf.DecimalUtils 中的静态方法
因为原生 lessThan 运算符在精确比较具有不同精度和小数位数的 decimal 值时存在问题。
lessThan(BinaryOperable, BigDecimal, int) - 类 ai.rapids.cudf.DecimalUtils 中的静态方法
因为原生 lessThan 运算符在精确比较具有不同精度和小数位数的 decimal 值时存在问题。
libraryLoaded() - 类 ai.rapids.cudf.CuFile 中的静态方法
检查 libcufilejni 库是否已加载。
libraryLoaded() - 类 ai.rapids.cudf.NativeDepsLoader 中的静态方法
 
like(Scalar, Scalar) - 类 ai.rapids.cudf.ColumnView 中的方法
返回一个布尔 ColumnVector,用于标识匹配给定 like 模式的行。
LIST - 类 ai.rapids.cudf.DType 中的静态变量
 
listBuilder(String) - 类 ai.rapids.cudf.ColumnWriterOptions 中的静态方法
为名为 'name' 的列创建 ListBuilder
listBuilder(String, boolean) - 类 ai.rapids.cudf.ColumnWriterOptions 中的静态方法
为名为 'name' 的列创建 ListBuilder
ListBuilder(String, boolean) - 类 ai.rapids.cudf.ColumnWriterOptions.ListBuilder 的构造函数
 
ListColumnWriterOptions(ColumnWriterOptions.ListBuilder) - 类 ai.rapids.cudf.ColumnWriterOptions.ListColumnWriterOptions 的构造函数
 
listConcatenateByRow(ColumnView...) - 类 ai.rapids.cudf.ColumnVector 中的静态方法
水平(按行)连接列表列,将每列中的对应行合并到新列的单个列表行中。
listConcatenateByRow(boolean, ColumnView...) - 类 ai.rapids.cudf.ColumnVector 中的静态方法
水平(按行)连接列表列,将每列中的对应行合并到新列的单个列表行中。
listContains(Scalar) - 类 ai.rapids.cudf.ColumnView 中的方法
创建一个布尔值列,指示指定的标量是否是列表列中每行的元素。
listContainsColumn(ColumnView) - 类 ai.rapids.cudf.ColumnView 中的方法
创建一个布尔值列,指示第一列的列表行是否包含第二列中的相应值。
listContainsNulls() - 类 ai.rapids.cudf.ColumnView 中的方法
创建一个布尔值列,指示指定列的列表行是否包含 null 元素。
listFromColumnView(ColumnView) - 类 ai.rapids.cudf.Scalar 中的静态方法
从 ColumnView 创建一个列表标量。
listFromNull(HostColumnVector.DataType) - 类 ai.rapids.cudf.Scalar 中的静态方法
创建一个列表类型的 null 标量。
listIndexOf(Scalar, ColumnView.FindOptions) - 类 ai.rapids.cudf.ColumnView 中的方法
创建一个 int32 索引列,指示标量搜索键在每个列表行中的位置。
listIndexOf(ColumnView, ColumnView.FindOptions) - 类 ai.rapids.cudf.ColumnView 中的方法
创建一个 int32 索引列,指示搜索键列中每行在其对应列表列行中的位置。
listReduce(SegmentedReductionAggregation) - 类 ai.rapids.cudf.ColumnView 中的方法
对列表中的值执行归约。
listReduce(SegmentedReductionAggregation, DType) - 类 ai.rapids.cudf.ColumnView 中的方法
对列表中的值执行归约。
listReduce(SegmentedReductionAggregation, NullPolicy, DType) - 类 ai.rapids.cudf.ColumnView 中的方法
对列表中的值执行归约。
listsDifferenceDistinct(ColumnView, ColumnView) - 类 ai.rapids.cudf.ColumnView 中的静态方法
找到左列列表与右列列表的差集。
listsHaveOverlap(ColumnView, ColumnView) - 类 ai.rapids.cudf.ColumnView 中的静态方法
对于输入列表列中的每对列表,检查它们是否包含任何共同的非 null 元素。
listsIntersectDistinct(ColumnView, ColumnView) - 类 ai.rapids.cudf.ColumnView 中的静态方法
找到给定列表列中每行列表之间的无重复交集。
listSortRows(boolean, boolean) - 类 ai.rapids.cudf.ColumnView 中的方法
对列表列中每行列表内的元素进行分段排序。
listsUnionDistinct(ColumnView, ColumnView) - 类 ai.rapids.cudf.ColumnView 中的静态方法
找到给定列表列中每行列表之间的无重复并集。
ListType(boolean, HostColumnVector.DataType) - 类 ai.rapids.cudf.HostColumnVector.ListType 的构造函数
 
Literal - 类在 ai.rapids.cudf.ast
AST 表达式中的文字值。
loadNativeDep(String, boolean) - 类 ai.rapids.cudf.NativeDepsLoader 中的静态方法
允许其他库重用相同的原生依赖加载逻辑。
loadNativeDeps() - 类 ai.rapids.cudf.NativeDepsLoader 中的静态方法
如果尚未加载,则加载 libcudf 所需的原生库。
loadNativeDeps(String[]) - 类 ai.rapids.cudf.NativeDepsLoader 中的静态方法
允许其他库重用相同的原生依赖加载逻辑。
loadNativeDeps(String[], boolean) - 类 ai.rapids.cudf.NativeDepsLoader 中的静态方法
允许其他库重用相同的原生依赖加载逻辑。
log(BinaryOperable, DType) - 接口 ai.rapids.cudf.BinaryOperable 中的方法
计算指定底数的对数
log(BinaryOperable) - 接口 ai.rapids.cudf.BinaryOperable 中的方法
计算指定底数的对数,输出类型与此相同。
log() - 类 ai.rapids.cudf.ColumnView 中的方法
计算对数,输出类型与输入相同。
log10() - 类 ai.rapids.cudf.ColumnView 中的方法
计算以 10 为底的对数,输出类型与输入相同。
log2() - 类 ai.rapids.cudf.ColumnView 中的方法
计算以 2 为底的对数,输出类型与输入相同。
logicalCastTo(DType) - 类 ai.rapids.cudf.ColumnView 中的方法
已弃用。
在 C++ 中,此已更改为 bit_cast,因此请改用该名称。
logRefCountDebug(String) - 类 ai.rapids.cudf.MemoryCleaner.Cleaner 中的方法
 
logTo(File) - 类 ai.rapids.cudf.Rmm 中的静态方法
创建一个将分配/释放日志写入文件的配置。
logToStderr() - 类 ai.rapids.cudf.Rmm 中的静态方法
创建一个将分配/释放日志写入 stderr 的配置。
logToStdout() - 类 ai.rapids.cudf.Rmm 中的静态方法
创建一个将分配/释放日志写入 stdout 的配置。
lower() - 类 ai.rapids.cudf.ColumnView 中的方法
将字符串转换为小写。
lowerBound(boolean[], Table, boolean[]) - 类 ai.rapids.cudf.Table 中的方法
在排序表中找到应插入值以保持顺序的最小索引。
lowerBound(Table, OrderByArg...) - 类 ai.rapids.cudf.Table 中的方法
在排序表中找到应插入值以保持顺序的最小索引。
lstrip() - 类 ai.rapids.cudf.ColumnView 中的方法
移除字符串开头的空格。
lstrip(Scalar) - 类 ai.rapids.cudf.ColumnView 中的方法
移除每个字符串开头指定的字符。

M

M2() - 类 ai.rapids.cudf.GroupByAggregation 中的静态方法
与均方差之和。
makeChunkedPack(long, RmmDeviceMemoryResource) - 类 ai.rapids.cudf.Table 中的方法
创建一个 `ChunkedPack` 实例,可用于在内存中连续打包此表,利用大小为 `bounceBufferSize` 的跳板缓冲区。
makeChunkedPack(long) - 类 ai.rapids.cudf.Table 中的方法
创建一个 `ChunkedPack` 实例,可用于在内存中连续打包此表,利用大小为 `bounceBufferSize` 的跳板缓冲区。
makeList(ColumnView...) - 类 ai.rapids.cudf.ColumnVector 中的静态方法
从给定列创建 LIST 列。
makeList(long, DType, ColumnView...) - 类 ai.rapids.cudf.ColumnVector 中的静态方法
从给定列创建 LIST 列。
makeListFromOffsets(long, ColumnView) - 类 ai.rapids.cudf.ColumnVector 中的方法
从当前列和给定的偏移量列创建 LIST 列。
makeStruct(ColumnView...) - 类 ai.rapids.cudf.ColumnVector 中的静态方法
创建由现有列组成的新结构向量。
makeStruct(long, ColumnView...) - 类 ai.rapids.cudf.ColumnVector 中的静态方法
创建由现有列组成的新结构向量。
makeStructView(long, ColumnView...) - 类 ai.rapids.cudf.ColumnView 中的静态方法
创建现有列视图的新结构列视图。
makeStructView(ColumnView...) - 类 ai.rapids.cudf.ColumnView 中的静态方法
创建现有列视图的新结构列视图。
mapColumn(String, ColumnWriterOptions, ColumnWriterOptions) - 类 ai.rapids.cudf.ColumnWriterOptions 中的静态方法
已弃用。
mapColumn(String, ColumnWriterOptions, ColumnWriterOptions, Boolean) - 类 ai.rapids.cudf.ColumnWriterOptions 中的静态方法
向模式添加 Map 列。
mapFile(File, FileChannel.MapMode, long, long) - 类 ai.rapids.cudf.HostMemoryBuffer 中的静态方法
创建一个内存映射到文件的主机缓冲区。
matchesRe(String) - 类 ai.rapids.cudf.ColumnView 中的方法
已弃用。
matchesRe(RegexProgram) - 类 ai.rapids.cudf.ColumnView 中的方法
返回一个布尔 ColumnVector,用于标识匹配给定正则表达式程序模式(但仅在字符串开头)的行。
max() - 类 ai.rapids.cudf.ColumnView 中的方法
返回列中所有值的最大值,返回与此列相同类型的标量。
max(DType) - 类 ai.rapids.cudf.ColumnView 中的方法
已弃用。
max 归约不再内部允许设置输出类型,作为一种解决方法,此 API 将为您将输入类型强制转换为输出类型,但这并非适用于所有情况。
max() - 类 ai.rapids.cudf.GroupByAggregation 中的静态方法
最大值聚合
max() - 类 ai.rapids.cudf.GroupByScanAggregation 中的静态方法
最大值聚合
max() - 类 ai.rapids.cudf.ReductionAggregation 中的静态方法
最大值聚合
max() - 类 ai.rapids.cudf.RollingAggregation 中的静态方法
滚动窗口最大值
max() - 类 ai.rapids.cudf.ScanAggregation 中的静态方法
最大值聚合
max() - 类 ai.rapids.cudf.SegmentedReductionAggregation 中的静态方法
最大值聚合
maxNullAware(BinaryOperable, DType) - 接口 ai.rapids.cudf.BinaryOperable 中的方法
返回最大的非 null 值。
maxNullAware(BinaryOperable) - 接口 ai.rapids.cudf.BinaryOperable 中的方法
返回最大的非 null 值。
md5Hash(ColumnView...) - 类 ai.rapids.cudf.ColumnVector 中的静态方法
创建一个新向量,包含表中每行的 MD5 哈希值。
mean() - 类 ai.rapids.cudf.ColumnView 中的方法
返回列中所有值的算术平均值,返回 FLOAT64 标量,除非列类型为 FLOAT32,此时返回 FLOAT32 标量。
mean(DType) - 类 ai.rapids.cudf.ColumnView 中的方法
返回列中所有值的算术平均值,返回指定类型的标量。
mean() - 类 ai.rapids.cudf.GroupByAggregation 中的静态方法
算术平均值归约。
mean() - 类 ai.rapids.cudf.ReductionAggregation 中的静态方法
算术平均值归约。
mean() - 类 ai.rapids.cudf.RollingAggregation 中的静态方法
算术平均值
median() - 类 ai.rapids.cudf.GroupByAggregation 中的静态方法
中位数归约。
median() - 类 ai.rapids.cudf.ReductionAggregation 中的静态方法
中位数归约。
memGetInfo() - 类 ai.rapids.cudf.Cuda 中的静态方法
映射:cudaMemGetInfo(size_t *free, size_t *total)
MemoryBuffer - 类在 ai.rapids.cudf
表示内存缓冲区的抽象类 注意:MemoryBuffer 是公共的,以便更轻松地处理类层次结构,不建议且不支持超出 CUDF 包含范围的子类化。
MemoryBuffer(long, long, MemoryBuffer.MemoryBufferCleaner) - 类 ai.rapids.cudf.MemoryBuffer 的构造函数
构造函数
MemoryBuffer(long, long) - 类 ai.rapids.cudf.MemoryBuffer 的构造函数
构造函数
MemoryBuffer(long, long, MemoryBuffer) - 类 ai.rapids.cudf.MemoryBuffer 的构造函数
创建切片时使用的内部构造函数。
MemoryBuffer.EventHandler - 接口在 ai.rapids.cudf
用于处理此 MemoryBuffer 事件的接口。
MemoryBuffer.MemoryBufferCleaner - 类在 ai.rapids.cudf
 
MemoryBufferCleaner() - 类 ai.rapids.cudf.MemoryBuffer.MemoryBufferCleaner 的构造函数
 
MemoryCleaner - 类在 ai.rapids.cudf
ColumnVectors 可能在堆外存储数据,由于处理复杂,单个向量的生命周期可能会有很大差异。
MemoryCleaner() - 类 ai.rapids.cudf.MemoryCleaner 的构造函数
 
MemoryCleaner.Cleaner - 类在 ai.rapids.cudf
可用于清理向量资源的 API,即使发生内存泄漏
memset(long, byte, long) - 类 ai.rapids.cudf.Cuda 中的静态方法
用 value 设置从 dst 指向的内存区域开始的 count 字节。
merge(Table[], OrderByArg...) - 类 ai.rapids.cudf.Table 中的静态方法
合并多个已排序表,保持排序顺序不变。
merge(List<Table>, OrderByArg...) - 类 ai.rapids.cudf.Table 中的静态方法
合并多个已排序表,保持排序顺序不变。
mergeAndSetValidity(BinaryOp, ColumnView...) - 类 ai.rapids.cudf.ColumnView 中的方法
创建列的深层副本,同时替换 null 掩码。
mergeHistogram() - 类 ai.rapids.cudf.GroupByAggregation 中的静态方法
MergeHistogram 聚合,用于合并多个直方图。
mergeHistogram() - 类 ai.rapids.cudf.ReductionAggregation 中的静态方法
创建 MergeHistogramAggregation,用于合并多个直方图。
mergeLists() - 类 ai.rapids.cudf.GroupByAggregation 中的静态方法
合并多个 CollectListAggregations 产生的局部列表。
mergeLists() - 类 ai.rapids.cudf.ReductionAggregation 中的静态方法
合并多个 CollectListAggregations 产生的局部列表。
mergeM2() - 类 ai.rapids.cudf.GroupByAggregation 中的静态方法
合并由多个 M2Aggregation 实例产生的局部 M2 值。
mergeSets() - 类 ai.rapids.cudf.GroupByAggregation 中的静态方法
合并由多个 CollectSetAggregations 产生的局部集。
mergeSets(NullEquality, NaNEquality) - 类 ai.rapids.cudf.GroupByAggregation 中的静态方法
合并由多个 CollectSetAggregations 产生的局部集。
mergeSets() - 类 ai.rapids.cudf.ReductionAggregation 中的静态方法
合并由多个 CollectSetAggregations 产生的局部集。
mergeSets(NullEquality, NaNEquality) - 类 ai.rapids.cudf.ReductionAggregation 中的静态方法
合并由多个 CollectSetAggregations 产生的局部集。
mergeTDigest(int) - 类 ai.rapids.cudf.GroupByAggregation 中的静态方法
合并 t-digest。
mergeTDigest(int) - 类 ai.rapids.cudf.ReductionAggregation 中的静态方法
tDigest 合并归约。
min() - 类 ai.rapids.cudf.ColumnView 中的方法
返回列中所有值的最小值,返回与此列相同类型的标量。
min(DType) - 类 ai.rapids.cudf.ColumnView 中的方法
已弃用。
min 归约不再内部允许设置输出类型,作为一种解决方法,此 API 将为您将输入类型强制转换为输出类型,但这并非适用于所有情况。
min() - 类 ai.rapids.cudf.GroupByAggregation 中的静态方法
最小值聚合
min() - 类 ai.rapids.cudf.GroupByScanAggregation 中的静态方法
最小值聚合
min() - 类 ai.rapids.cudf.ReductionAggregation 中的静态方法
最小值聚合
min() - 类 ai.rapids.cudf.RollingAggregation 中的静态方法
滚动窗口最小值
min() - 类 ai.rapids.cudf.ScanAggregation 中的静态方法
最小值聚合
min() - 类 ai.rapids.cudf.SegmentedReductionAggregation 中的静态方法
最小值聚合
minNullAware(BinaryOperable, DType) - 接口 ai.rapids.cudf.BinaryOperable 中的方法
返回最小的非 null 值。
minNullAware(BinaryOperable) - 接口 ai.rapids.cudf.BinaryOperable 中的方法
返回最小的非 null 值。
minPeriods(int) - 类 ai.rapids.cudf.WindowOptions.Builder 中的方法
设置评估元素所需的最小观测数。
minute() - 类 ai.rapids.cudf.ColumnView 中的方法
从具有时间分辨率的时间戳获取分钟。
missingFieldsAsNulls(boolean) - 类 ai.rapids.cudf.GetJsonObjectOptions.Builder 中的方法
 
mixedFullJoinGatherMaps(Table, Table, Table, Table, CompiledExpression, NullEquality) - 类 ai.rapids.cudf.Table 中的静态方法
计算可用于显现两个表之间使用混合相等和不等条件进行全连接结果的收集映射。
mixedInnerJoinGatherMaps(Table, Table, Table, Table, CompiledExpression, NullEquality) - 类 ai.rapids.cudf.Table 中的静态方法
计算可用于显现两个表之间使用混合相等和不等条件进行内连接结果的收集映射。
mixedInnerJoinGatherMaps(Table, Table, Table, Table, CompiledExpression, NullEquality, MixedJoinSize) - 类 ai.rapids.cudf.Table 中的静态方法
计算可用于显现两个表之间使用混合相等和不等条件进行内连接结果的收集映射。
mixedInnerJoinSize(Table, Table, Table, Table, CompiledExpression, NullEquality) - 类 ai.rapids.cudf.Table 中的静态方法
计算使用混合相等和不等条件进行两个表之间内连接的输出大小信息。
MixedJoinSize - 类在 ai.rapids.cudf
此类跟踪与混合表连接相关的大小信息。
mixedLeftAntiJoinGatherMap(Table, Table, Table, Table, CompiledExpression, NullEquality) - 类 ai.rapids.cudf.Table 中的静态方法
计算可用于显现两个表之间使用混合相等和不等条件进行左反连接结果的收集映射。
mixedLeftJoinGatherMaps(Table, Table, Table, Table, CompiledExpression, NullEquality) - 类 ai.rapids.cudf.Table 中的静态方法
计算可用于显现两个表之间使用混合相等和不等条件进行左连接结果的收集映射。
mixedLeftJoinGatherMaps(Table, Table, Table, Table, CompiledExpression, NullEquality, MixedJoinSize) - 类 ai.rapids.cudf.Table 中的静态方法
计算可用于显现两个表之间使用混合相等和不等条件进行左连接结果的收集映射。
mixedLeftJoinSize(Table, Table, Table, Table, CompiledExpression, NullEquality) - 类 ai.rapids.cudf.Table 中的静态方法
计算使用混合相等和不等条件进行两个表之间左连接的输出大小信息。
mixedLeftSemiJoinGatherMap(Table, Table, Table, Table, CompiledExpression, NullEquality) - 类 ai.rapids.cudf.Table 中的静态方法
计算可用于显现两个表之间使用混合相等和不等条件进行左半连接结果的收集映射。
mod(BinaryOperable, DType) - 接口 ai.rapids.cudf.BinaryOperable 中的方法
计算指定输出类型的模数。
mod(BinaryOperable) - 接口 ai.rapids.cudf.BinaryOperable 中的方法
计算模数。
month() - 类 ai.rapids.cudf.ColumnView 中的方法
从时间戳获取月份。
mul(BinaryOperable, DType) - 接口 ai.rapids.cudf.BinaryOperable 中的方法
将两个向量相乘,并指定输出类型。
mul(BinaryOperable) - 接口 ai.rapids.cudf.BinaryOperable 中的方法
将两个向量相乘。
multiBufferCopyAsync(long[], long[], long[], Cuda.Stream) - 类 ai.rapids.cudf.Cuda 中的静态方法
将数据从多个设备缓冲区源复制到多个设备缓冲区目的地。
MultiBufferDataSource - 类在 ai.rapids.cudf
这是一个可以接受多个 HostMemoryBuffers 的 DataSource。
MultiBufferDataSource(HostMemoryBuffer...) - 类 ai.rapids.cudf.MultiBufferDataSource 的构造函数
创建由多个缓冲区支持的新数据源。
MultiBufferDataSource(HostMemoryAllocator, HostMemoryBuffer...) - 类 ai.rapids.cudf.MultiBufferDataSource 的构造函数
创建由多个缓冲区支持的新数据源。

N

name - 类 ai.rapids.cudf.ColumnWriterOptions.NestedBuilder 中的变量
 
NaNEquality - 枚举在 ai.rapids.cudf
在操作中应如何比较 NaN。
nansToNulls() - 类 ai.rapids.cudf.ColumnView 中的方法
返回一个新的 ColumnVector,其中 NaN 已转换为 null,同时保留现有的 null 值。
NativeDepsLoader - 类在 ai.rapids.cudf
此类将加载原生依赖项。
NativeDepsLoader() - 类 ai.rapids.cudf.NativeDepsLoader 的构造函数
 
needTheGpu() - 接口 ai.rapids.cudf.ArrowIPCOptions.NeedGpu 中的方法
一个回调,用于指示我们即将开始将数据放在 GPU 上。
NestedBuilder(String, boolean) - 类 ai.rapids.cudf.ColumnWriterOptions.NestedBuilder 的构造函数
用于构建 Struct 元数据的构造器
NestedBuilder(String, boolean, int) - 类 ai.rapids.cudf.ColumnWriterOptions.NestedBuilder 的构造函数
 
NestedBuilder() - 类 ai.rapids.cudf.ColumnWriterOptions.NestedBuilder 的构造函数
 
newPinnedPoolMemoryResource(long, long) - 类 ai.rapids.cudf.Rmm 中的静态方法
 
next(DeviceMemoryBuffer) - 类 ai.rapids.cudf.ChunkedPack 中的方法
将表的下一个连续块放入 `userPtr` 中。
nonNumericNumbersAllowed() - 类 ai.rapids.cudf.JSONOptions 中的方法
 
normalizeNANsAndZeros() - 类 ai.rapids.cudf.ColumnView 中的方法
创建一个“标准化”值的新向量,其中:1.
not() - 类 ai.rapids.cudf.ColumnView 中的方法
返回输入列(this)中每个值的逻辑 `not` 的向量
notEqualTo(BinaryOperable, DType) - 接口 ai.rapids.cudf.BinaryOperable 中的方法
this != rhs,如果为 true 则为 1,否则为 0,并将输出强制转换为给定类型。
notEqualTo(BinaryOperable) - 接口 ai.rapids.cudf.BinaryOperable 中的方法
this != rhs,如果为 true 则为 1,否则为 0。
notEqualToNullAware(BinaryOperable, DType) - 接口 ai.rapids.cudf.BinaryOperable 中的方法
类似于 notEqualTo,但 NULL != NULL 为 TRUE,NULL != 非 NULL 为 FALSE
notEqualToNullAware(BinaryOperable) - 接口 ai.rapids.cudf.BinaryOperable 中的方法
类似于 notEqualTo,但 NULL != NULL 为 TRUE,NULL != 非 NULL 为 FALSE
noWarnLeakExpected() - 类 ai.rapids.cudf.ColumnVector 中的方法
这是一个非常难看的 API,但由于 Java 和 GC 的原因,一列数据的生命周期可能不明确。
noWarnLeakExpected() - 类 ai.rapids.cudf.ColumnVector.OffHeapState 中的方法
 
noWarnLeakExpected() - 类 ai.rapids.cudf.HostColumnVector 中的方法
这是一个非常难看的 API,但由于 Java 和 GC 的原因,一列数据的生命周期可能不明确。
noWarnLeakExpected() - 类 ai.rapids.cudf.HostColumnVectorCore.OffHeapState 中的方法
 
noWarnLeakExpected() - 类 ai.rapids.cudf.MemoryBuffer 中的方法
这是一个非常难看的 API,但由于 Java 和 GC 的原因,一列数据的生命周期可能不明确。
noWarnLeakExpected() - 类 ai.rapids.cudf.MemoryCleaner.Cleaner 中的方法
 
nth(int) - 类 ai.rapids.cudf.GroupByAggregation 中的静态方法
获取组中第 n 个非 null 元素。
nth(int, NullPolicy) - 类 ai.rapids.cudf.GroupByAggregation 中的静态方法
获取组中第 n 个元素。
nth(int) - 类 ai.rapids.cudf.ReductionAggregation 中的静态方法
获取组中第 n 个非 null 元素。
nth(int, NullPolicy) - 类 ai.rapids.cudf.ReductionAggregation 中的静态方法
获取组中第 n 个元素。
nth(int, NullPolicy) - 类 ai.rapids.cudf.RollingAggregation 中的静态方法
从指定的窗口中选择第 n 个元素。
nullCount - 类 ai.rapids.cudf.ColumnView 中的变量
 
nullCount - 类 ai.rapids.cudf.HostColumnVectorCore 中的变量
 
nullCount - 类 ai.rapids.cudf.JCudfSerialization.SerializedColumnHeader 中的变量
 
NullEquality - 枚举在 ai.rapids.cudf
在操作中应如何比较 null。
NullPolicy - 枚举在 ai.rapids.cudf
指定在操作中是否包含 null 或排除 null。
numCompressedBytes - 类 ai.rapids.cudf.TableWriter.WriteStatistics 中的变量
 
numFailedBytes - 类 ai.rapids.cudf.TableWriter.WriteStatistics 中的变量
 
numSkippedBytes - 类 ai.rapids.cudf.TableWriter.WriteStatistics 中的变量
 
nunique() - 类 ai.rapids.cudf.GroupByAggregation 中的静态方法
唯一非 null 元素的数量。
nunique(NullPolicy) - 类 ai.rapids.cudf.GroupByAggregation 中的静态方法
唯一元素的数量。
nunique() - 类 ai.rapids.cudf.ReductionAggregation 中的静态方法
唯一非 null 元素的数量。
nunique(NullPolicy) - 类 ai.rapids.cudf.ReductionAggregation 中的静态方法
唯一元素的数量。
NvcompCudaException - 异常在 ai.rapids.cudf.nvcomp
nvcomp 抛出的异常,指示发生了 CUDA 错误。
NvcompException - 异常在 ai.rapids.cudf.nvcomp
所有 nvcomp 特定异常的基类。
NvtxColor - 枚举在 ai.rapids.cudf
 
NvtxRange - 类在 ai.rapids.cudf
此类支持 push/pop NVTX 分析范围,或“作用域内”范围。
NvtxRange(String, NvtxColor) - 类 ai.rapids.cudf.NvtxRange 的构造函数
 
NvtxRange(String, int) - 类 ai.rapids.cudf.NvtxRange 的构造函数
 
NvtxUniqueRange - 类位于 ai.rapids.cudf
此类支持 NVTX 分析范围的开始/结束标记。
NvtxUniqueRange(String, NvtxColor) - 类 ai.rapids.cudf.NvtxUniqueRange 的构造函数
 
NvtxUniqueRange(String, int) - 类 ai.rapids.cudf.NvtxUniqueRange 的构造函数
 

O

ofBoolean(boolean) - 类 ai.rapids.cudf.ast.Literal 中的静态方法
使用指定值构造布尔文字。
ofBoolean(Boolean) - 类 ai.rapids.cudf.ast.Literal 中的静态方法
使用指定值或 null 构造布尔文字。
ofByte(byte) - 类 ai.rapids.cudf.ast.Literal 中的静态方法
使用指定值构造字节文字。
ofByte(Byte) - 类 ai.rapids.cudf.ast.Literal 中的静态方法
使用指定值或 null 构造字节文字。
ofDouble(double) - 类 ai.rapids.cudf.ast.Literal 中的静态方法
使用指定值构造双精度文字。
ofDouble(Double) - 类 ai.rapids.cudf.ast.Literal 中的静态方法
使用指定值或 null 构造双精度文字。
ofDurationDaysFromInt(int) - 类 ai.rapids.cudf.ast.Literal 中的静态方法
使用指定值构造天数时长文字。
ofDurationDaysFromInt(Integer) - 类 ai.rapids.cudf.ast.Literal 中的静态方法
使用指定值或 null 构造天数时长文字。
ofDurationFromLong(DType, long) - 类 ai.rapids.cudf.ast.Literal 中的静态方法
使用指定值构造基于 long 的时长文字。
ofDurationFromLong(DType, Long) - 类 ai.rapids.cudf.ast.Literal 中的静态方法
使用指定值或 null 构造基于 long 的时长文字。
offHeap - 类 ai.rapids.cudf.ColumnView 中的变量
 
offHeap - 类 ai.rapids.cudf.HostColumnVectorCore 中的变量
 
OffHeapState(long) - 类 ai.rapids.cudf.ColumnVector.OffHeapState 的构造函数
从现有的 cudf::column * 创建列。
OffHeapState(DeviceMemoryBuffer, DeviceMemoryBuffer, DeviceMemoryBuffer, List<DeviceMemoryBuffer>, long) - 类 ai.rapids.cudf.ColumnVector.OffHeapState 的构造函数
从现有的 cudf::column_view 和缓冲区创建。
OffHeapState(long, DeviceMemoryBuffer) - 类 ai.rapids.cudf.ColumnVector.OffHeapState 的构造函数
从现有的 cudf::column_view 和连续缓冲区创建。
ofFloat(float) - 类 ai.rapids.cudf.ast.Literal 中的静态方法
使用指定值构造浮点文字。
ofFloat(Float) - 类 ai.rapids.cudf.ast.Literal 中的静态方法
使用指定值或 null 构造浮点文字。
offsets - 类 ai.rapids.cudf.HostColumnVectorCore.OffHeapState 中的变量
 
ofInt(int) - 类 ai.rapids.cudf.ast.Literal 中的静态方法
使用指定值构造整数文字。
ofInt(Integer) - 类 ai.rapids.cudf.ast.Literal 中的静态方法
使用指定值或 null 构造整数文字。
ofLong(long) - 类 ai.rapids.cudf.ast.Literal 中的静态方法
使用指定值构造 long 文字。
ofLong(Long) - 类 ai.rapids.cudf.ast.Literal 中的静态方法
使用指定值或 null 构造 long 文字。
ofNull(DType) - 类 ai.rapids.cudf.ast.Literal 中的静态方法
构造指定类型的 null 文字。
ofShort(short) - 类 ai.rapids.cudf.ast.Literal 中的静态方法
使用指定值构造 short 文字。
ofShort(Short) - 类 ai.rapids.cudf.ast.Literal 中的静态方法
使用指定值或 null 构造 short 文字。
ofString(String) - 类 ai.rapids.cudf.ast.Literal 中的静态方法
使用指定值或 null 构造字符串文字。
ofTimestampDaysFromInt(int) - 类 ai.rapids.cudf.ast.Literal 中的静态方法
使用指定值构造天数时间戳文字。
ofTimestampDaysFromInt(Integer) - 类 ai.rapids.cudf.ast.Literal 中的静态方法
使用指定值或 null 构造天数时间戳文字。
ofTimestampFromLong(DType, long) - 类 ai.rapids.cudf.ast.Literal 中的静态方法
使用指定值构造基于 long 的时间戳文字。
ofTimestampFromLong(DType, Long) - 类 ai.rapids.cudf.ast.Literal 中的静态方法
使用指定值或 null 构造基于 long 的时间戳文字。
ofUTF8String(byte[]) - 类 ai.rapids.cudf.ast.Literal 中的静态方法
直接使用字节数组构造字符串文字,以跳过转码。
onAllocated(long) - 接口 ai.rapids.cudf.RmmEventHandler 中的方法
在启用调试模式时,每次内存分配后调用。
onAllocFailure(long) - 接口 ai.rapids.cudf.RmmEventHandler 中的方法
已弃用。
已弃用,请使用 onAllocFailure(long, boolean)
onAllocFailure(long, int) - 接口 ai.rapids.cudf.RmmEventHandler 中的方法
在内存分配失败时调用。
onAllocThreshold(long) - 接口 ai.rapids.cudf.RmmEventHandler 中的方法
当分配阈值被突破时,在 RMM 内存分配操作后调用。
onClosed(ColumnVector, int) - 接口 ai.rapids.cudf.ColumnVector.EventHandler 中的方法
`onClosed` 在 `close` 期间使用更新的 `refCount` 调用。
onClosed(HostColumnVector, int) - 接口 ai.rapids.cudf.HostColumnVector.EventHandler 中的方法
`onClosed` 在 `close` 期间使用更新的 `refCount` 调用。
onClosed(int) - 接口 ai.rapids.cudf.MemoryBuffer.EventHandler 中的方法
`onClosed` 在 `close` 期间使用更新的 `refCount` 调用。
onColumn(int) - 类 ai.rapids.cudf.GroupByAggregation 中的方法
向聚合中添加列,以便可以在特定数据列上使用它。
onColumn(int) - 类 ai.rapids.cudf.GroupByScanAggregation 中的方法
向聚合中添加列,以便可以在特定数据列上使用它。
onColumn(int) - 枚举 ai.rapids.cudf.ReplacePolicy 中的方法
指示应在哪个列上进行替换。
onColumn(int) - 类 ai.rapids.cudf.RollingAggregation 中的方法
向聚合中添加列,以便可以在特定数据列上使用它。
onColumns(int...) - 类 ai.rapids.cudf.Table 中的方法
 
onDeallocated(long) - 接口 ai.rapids.cudf.RmmEventHandler 中的方法
在启用调试模式时,每次内存释放后调用。
onDeallocThreshold(long) - 接口 ai.rapids.cudf.RmmEventHandler 中的方法
当释放阈值被突破时,在 RMM 内存释放操作后调用。
onHostBufferDone(HostMemoryBuffer) - 类 ai.rapids.cudf.DataSource 中的方法
当 hostRead 返回的缓冲区处理完毕时调用。
or(BinaryOperable, DType) - 接口 ai.rapids.cudf.BinaryOperable 中的方法
使用给定输出类型进行逻辑或 (||) 运算。
or(BinaryOperable) - 接口 ai.rapids.cudf.BinaryOperable 中的方法
逻辑或 (||) 运算。
ORCChunkedReader - 类位于 ai.rapids.cudf
提供一种迭代读取 ORC 文件的接口。
ORCChunkedReader(long, long, ORCOptions, HostMemoryBuffer, long, long) - 类 ai.rapids.cudf.ORCChunkedReader 的构造函数
从读取限制、输出行粒度以及已加载到内存缓冲区中的文件构造读取器实例。
ORCChunkedReader(long, long, long, ORCOptions, HostMemoryBuffer, long, long) - 类 ai.rapids.cudf.ORCChunkedReader 的构造函数
构造一个分块 ORC 读取器实例,类似于 ORCChunkedReader(long, long, ORCOptions, HostMemoryBuffer, long, long),并带有附加参数来控制输出表的粒度。
ORCOptions - 类位于 ai.rapids.cudf
读取 ORC 文件的选项
ORCOptions.Builder - 类位于 ai.rapids.cudf
 
ORCWriterOptions - 类位于 ai.rapids.cudf
此类表示用于写入 ORC 文件的设置。
ORCWriterOptions.Builder - 类位于 ai.rapids.cudf
 
orderBy(OrderByArg...) - 类 ai.rapids.cudf.Table 中的方法
使用排序键对表进行排序,返回一个新的已分配表。
OrderByArg - 类位于 ai.rapids.cudf
为特定列提供排序。
orderByAscending() - 类 ai.rapids.cudf.WindowOptions.Builder 中的方法
 
orderByColumnIndex(int) - 类 ai.rapids.cudf.WindowOptions.Builder 中的方法
 
orderByDescending() - 类 ai.rapids.cudf.WindowOptions.Builder 中的方法
 
outOfBounds(ColumnView, int, int) - 类 ai.rapids.cudf.DecimalUtils 中的静态方法
根据精度和标度,检查输入 decimal 列的每个值是否超出范围。
OutOfBoundsPolicy - 枚举位于 ai.rapids.cudf
处理可能的越界索引的策略。`NULLIFY` 表示将对应于越界 gather map 值的输出值置为 null。
OUTPUT_STREAM - 类 ai.rapids.cudf.TableDebug 中的静态变量
指定以下选项之一:-Dai.rapids.cudf.debug.output=stderr 直接打印到标准错误 (默认) -Dai.rapids.cudf.debug.output=stdout 直接打印到标准输出 -Dai.rapids.cudf.debug.output=log[_level] 重定向到可进一步配置的日志记录子系统。
overWindow(WindowOptions) - 类 ai.rapids.cudf.RollingAggregationOnColumn 中的方法
 

P

PackedColumnMetadata - 类位于 ai.rapids.cudf
由单个连续设备缓冲区支持的表的元数据。
pad(int) - 类 ai.rapids.cudf.ColumnView 中的方法
用空格 " " 右填充字符串列,直到达到所需长度。
pad(int, PadSide) - 类 ai.rapids.cudf.ColumnView 中的方法
用空格 " " 填充字符串列,直到达到所需长度。
pad(int, PadSide, String) - 类 ai.rapids.cudf.ColumnView 中的方法
填充字符串列,直到达到所需长度。
PadSide - 枚举位于 ai.rapids.cudf
 
ParquetChunkedReader - 类位于 ai.rapids.cudf
提供一种迭代读取 Parquet 文件的接口。
ParquetChunkedReader(long, File) - 类 ai.rapids.cudf.ParquetChunkedReader 的构造函数
从读取限制和文件路径构造读取器实例。
ParquetChunkedReader(long, ParquetOptions, File) - 类 ai.rapids.cudf.ParquetChunkedReader 的构造函数
从读取限制、ParquetOptions 对象和文件路径构造读取器实例。
ParquetChunkedReader(long, long, ParquetOptions, File) - 类 ai.rapids.cudf.ParquetChunkedReader 的构造函数
从读取限制、ParquetOptions 对象和文件路径构造读取器实例。
ParquetChunkedReader(long, ParquetOptions, HostMemoryBuffer, long, long) - 类 ai.rapids.cudf.ParquetChunkedReader 的构造函数
从读取限制和已读入内存缓冲区的文件构造读取器实例。
ParquetChunkedReader(long, long, ParquetOptions, HostMemoryBuffer, long, long) - 类 ai.rapids.cudf.ParquetChunkedReader 的构造函数
从读取限制和已读入内存缓冲区的文件构造读取器实例。
ParquetChunkedReader(long, long, ParquetOptions, HostMemoryBuffer...) - 类 ai.rapids.cudf.ParquetChunkedReader 的构造函数
从读取限制和主机内存缓冲区中的数据构造读取器实例。
ParquetChunkedReader(long, ParquetOptions, DataSource) - 类 ai.rapids.cudf.ParquetChunkedReader 的构造函数
从 DataSource 构造读取器实例
parquetFieldId - 类 ai.rapids.cudf.ColumnWriterOptions.NestedBuilder 中的变量
 
ParquetOptions - 类位于 ai.rapids.cudf
读取 parquet 文件的选项
ParquetOptions.Builder - 类位于 ai.rapids.cudf
 
ParquetWriterOptions - 类位于 ai.rapids.cudf
此类表示用于写入 Parquet 文件的设置。
ParquetWriterOptions.Builder - 类位于 ai.rapids.cudf
 
ParquetWriterOptions.StatisticsFrequency - 枚举位于 ai.rapids.cudf
 
parseErrorCode(int) - 枚举 ai.rapids.cudf.CudaException.CudaError 中的静态方法
 
partition(ColumnView, int) - 类 ai.rapids.cudf.Table 中的方法
使用 partitionMap 中的映射对表进行分区。
PartitionedTable - 类位于 ai.rapids.cudf
提供 PartitionedTable 的类
pattern() - 类 ai.rapids.cudf.RegexProgram 中的方法
获取用于创建此实例的模式
percentRank() - 类 ai.rapids.cudf.GroupByScanAggregation 中的静态方法
获取行的百分位排名。
percentRank() - 类 ai.rapids.cudf.ScanAggregation 中的静态方法
获取行的百分位排名。
PinnedMemoryPool - 类位于 ai.rapids.cudf
这是到 rmm::pool_memory_resource 的 JNI 接口.
pmod(BinaryOperable, DType) - 接口 ai.rapids.cudf.BinaryOperable 中的方法
返回 lhs 对 rhs 取模的正值。
pmod(BinaryOperable) - 接口 ai.rapids.cudf.BinaryOperable 中的方法
返回 lhs 对 rhs 取模的正值。
POOL - 类 ai.rapids.cudf.RmmAllocationMode 中的静态变量
使用池子分配策略
pow(BinaryOperable, DType) - 接口 ai.rapids.cudf.BinaryOperable 中的方法
使用给定输出类型计算幂。
pow(BinaryOperable) - 接口 ai.rapids.cudf.BinaryOperable 中的方法
计算幂。
preceding(Scalar) - 类 ai.rapids.cudf.WindowOptions.Builder 中的方法
为范围窗口设置当前行之前的相对数量
prefixSum() - 类 ai.rapids.cudf.ColumnView 中的方法
计算此列中值的前缀和(也称为累积和)。
printBuffer() - 类 ai.rapids.cudf.HostMemoryBuffer 中的方法
警告:仅用于调试的打印缓冲区方法。
printBuffer(int) - 类 ai.rapids.cudf.HostMemoryBuffer 中的方法
警告:仅用于调试的打印缓冲区方法。
product() - 类 ai.rapids.cudf.ColumnView 中的方法
返回列中所有值的乘积,返回的标量与此列类型相同。
product(DType) - 类 ai.rapids.cudf.ColumnView 中的方法
返回列中所有值的乘积,返回指定类型的标量。
product() - 类 ai.rapids.cudf.GroupByAggregation 中的静态方法
乘积聚合。
product() - 类 ai.rapids.cudf.GroupByScanAggregation 中的静态方法
乘积聚合。
product() - 类 ai.rapids.cudf.ReductionAggregation 中的静态方法
乘积聚合。
product() - 类 ai.rapids.cudf.ScanAggregation 中的静态方法
乘积聚合。
product() - 类 ai.rapids.cudf.SegmentedReductionAggregation 中的静态方法
乘积聚合。
profilerStart() - 类 ai.rapids.cudf.Cuda 中的静态方法
如果当前附加了分析器,则开始 Nsight 分析会话。
profilerStop() - 类 ai.rapids.cudf.Cuda 中的静态方法
停止一个活动的 Nsight 分析会话。
purgeNonEmptyNulls() - 类 ai.rapids.cudf.ColumnView 中的方法
将此列复制到输出中,同时清除列中或其后代中的任何非空 null 行。

Q

quantile(QuantileMethod, double[]) - 类 ai.rapids.cudf.ColumnView 中的方法
计算此 ColumnVector 的各种分位数。
quantile(double...) - 类 ai.rapids.cudf.GroupByAggregation 中的静态方法
聚合以计算指定的分位数。
quantile(QuantileMethod, double...) - 类 ai.rapids.cudf.GroupByAggregation 中的静态方法
聚合以计算各种分位数。
quantile(double...) - 类 ai.rapids.cudf.ReductionAggregation 中的静态方法
聚合以计算指定的分位数。
quantile(QuantileMethod, double...) - 类 ai.rapids.cudf.ReductionAggregation 中的静态方法
聚合以计算各种分位数。
QuantileMethod - 枚举位于 ai.rapids.cudf
当所需分位数位于两个数据点 i 和 j 之间时使用的插值方法。
quarterOfYear() - 类 ai.rapids.cudf.ColumnView 中的方法
从时间戳获取年份的季度。
QuoteStyle - 枚举位于 ai.rapids.cudf
CSV 记录的引用样式,紧密遵循 cudf::io::quote_style。

R

Range - 类位于 ai.rapids.cudf
创建范围的辅助工具。
Range() - 类 ai.rapids.cudf.Range 的构造函数
 
rank() - 类 ai.rapids.cudf.GroupByScanAggregation 中的静态方法
获取行的排名。
rank() - 类 ai.rapids.cudf.ScanAggregation 中的静态方法
获取行的排名。
read(CuFileBuffer, long) - 类 ai.rapids.cudf.CuFileReadHandle 中的方法
将文件内容读取到指定的 cuFile 缓冲区中。
readAndConcat(JCudfSerialization.SerializedTableHeader[], HostMemoryBuffer[]) - 类 ai.rapids.cudf.JCudfSerialization 中的静态方法
 
readAndInferJSON(JSONOptions, DataSource) - 类 ai.rapids.cudf.Table 中的静态方法
读取 JSON 格式数据并推断列名和 schema。
readArrowIPCChunked(ArrowIPCOptions, File) - 类 ai.rapids.cudf.Table 中的静态方法
获取将返回表的读取器。
readArrowIPCChunked(File) - 类 ai.rapids.cudf.Table 中的静态方法
获取将返回表的读取器。
readArrowIPCChunked(ArrowIPCOptions, HostBufferProvider, HostMemoryAllocator) - 类 ai.rapids.cudf.Table 中的静态方法
获取将返回表的读取器。
readArrowIPCChunked(ArrowIPCOptions, HostBufferProvider) - 类 ai.rapids.cudf.Table 中的静态方法
 
readArrowIPCChunked(HostBufferProvider) - 类 ai.rapids.cudf.Table 中的静态方法
获取将返回表的读取器。
readAvro(File) - 类 ai.rapids.cudf.Table 中的静态方法
使用默认 AvroOptions 读取 Avro 文件。
readAvro(AvroOptions, File) - 类 ai.rapids.cudf.Table 中的静态方法
读取 Avro 文件。
readAvro(byte[]) - 类 ai.rapids.cudf.Table 中的静态方法
读取 Avro 格式数据。
readAvro(AvroOptions, byte[]) - 类 ai.rapids.cudf.Table 中的静态方法
读取 Avro 格式数据。
readAvro(AvroOptions, byte[], long, long, HostMemoryAllocator) - 类 ai.rapids.cudf.Table 中的静态方法
读取 Avro 格式数据。
readAvro(AvroOptions, byte[], long, long) - 类 ai.rapids.cudf.Table 中的静态方法
 
readAvro(AvroOptions, HostMemoryBuffer, long, long) - 类 ai.rapids.cudf.Table 中的静态方法
读取 Avro 格式数据。
readAvro(AvroOptions, DataSource) - 类 ai.rapids.cudf.Table 中的静态方法
 
readChunk() - 类 ai.rapids.cudf.ORCChunkedReader 中的方法
读取给定 ORC 文件中的行块,使返回数据的总大小不超过给定的读取限制。
readChunk() - 类 ai.rapids.cudf.ParquetChunkedReader 中的方法
读取给定 Parquet 文件中的行块,使返回数据的总大小不超过给定的读取限制。
readCSV(Schema, File) - 类 ai.rapids.cudf.Table 中的静态方法
使用默认 CSVOptions 读取 CSV 文件。
readCSV(Schema, CSVOptions, File) - 类 ai.rapids.cudf.Table 中的静态方法
读取 CSV 文件。
readCSV(Schema, byte[]) - 类 ai.rapids.cudf.Table 中的静态方法
使用默认 CSVOptions 读取 CSV 格式数据。
readCSV(Schema, CSVOptions, byte[]) - 类 ai.rapids.cudf.Table 中的静态方法
读取 CSV 格式数据。
readCSV(Schema, CSVOptions, byte[], long, long, HostMemoryAllocator) - 类 ai.rapids.cudf.Table 中的静态方法
读取 CSV 格式数据。
readCSV(Schema, CSVOptions, byte[], long, long) - 类 ai.rapids.cudf.Table 中的静态方法
 
readCSV(Schema, CSVOptions, HostMemoryBuffer, long, long) - 类 ai.rapids.cudf.Table 中的静态方法
读取 CSV 格式数据。
readCSV(Schema, CSVOptions, DataSource) - 类 ai.rapids.cudf.Table 中的静态方法
 
readFileToDeviceBuffer(BaseDeviceMemoryBuffer, File, long) - 类 ai.rapids.cudf.CuFile 中的静态方法
将文件同步读取到设备缓冲区中。
readFileToDeviceMemory(long, long, File, long) - 类 ai.rapids.cudf.CuFile 中的静态方法
将文件同步读取到设备内存中。
readInto(HostMemoryBuffer, long) - 接口 ai.rapids.cudf.HostBufferProvider 中的方法
将数据放入给定的缓冲区中。
readJSON(Schema, File) - 类 ai.rapids.cudf.Table 中的静态方法
使用默认 JSONOptions 读取 JSON 文件。
readJSON(Schema, byte[]) - 类 ai.rapids.cudf.Table 中的静态方法
使用默认 JSONOptions 读取 JSON 格式数据。
readJSON(Schema, JSONOptions, byte[]) - 类 ai.rapids.cudf.Table 中的静态方法
读取 JSON 格式数据。
readJSON(Schema, JSONOptions, File) - 类 ai.rapids.cudf.Table 中的静态方法
读取 JSON 文件。
readJSON(Schema, JSONOptions, byte[], long, long, HostMemoryAllocator) - 类 ai.rapids.cudf.Table 中的静态方法
读取 JSON 格式数据。
readJSON(Schema, JSONOptions, byte[], long, long, HostMemoryAllocator, int) - 类 ai.rapids.cudf.Table 中的静态方法
已弃用。
此方法已弃用,因为未使用 emptyRowCount。请改用不带 emptyRowCount 的方法。
readJSON(Schema, JSONOptions, byte[], long, long, int) - 类 ai.rapids.cudf.Table 中的静态方法
 
readJSON(Schema, JSONOptions, byte[], long, long) - 类 ai.rapids.cudf.Table 中的静态方法
 
readJSON(JSONOptions, HostMemoryBuffer, long, long) - 类 ai.rapids.cudf.Table 中的静态方法
读取 JSON 格式数据并推断列名和 schema。
readJSON(Schema, JSONOptions, HostMemoryBuffer, long, long) - 类 ai.rapids.cudf.Table 中的静态方法
读取 JSON 格式数据。
readJSON(Schema, JSONOptions, HostMemoryBuffer, long, long, int) - 类 ai.rapids.cudf.Table 中的静态方法
已弃用。
此方法已弃用,因为未使用 emptyRowCount。请改用不带 emptyRowCount 的方法。
readJSON(Schema, JSONOptions, DataSource) - 类 ai.rapids.cudf.Table 中的静态方法
读取 JSON 格式数据。
readJSON(Schema, JSONOptions, DataSource, int) - 类 ai.rapids.cudf.Table 中的静态方法
已弃用。
此方法已弃用,因为未使用 emptyRowCount。请改用不带 emptyRowCount 的方法。
readORC(File) - 类 ai.rapids.cudf.Table 中的静态方法
使用默认 ORCOptions 读取 ORC 文件。
readORC(ORCOptions, File) - 类 ai.rapids.cudf.Table 中的静态方法
读取 ORC 文件。
readORC(byte[]) - 类 ai.rapids.cudf.Table 中的静态方法
读取 ORC 格式数据。
readORC(ORCOptions, byte[]) - 类 ai.rapids.cudf.Table 中的静态方法
读取 ORC 格式数据。
readORC(ORCOptions, byte[], long, long, HostMemoryAllocator) - 类 ai.rapids.cudf.Table 中的静态方法
读取 ORC 格式数据。
readORC(ORCOptions, byte[], long, long) - 类 ai.rapids.cudf.Table 中的静态方法
 
readORC(ORCOptions, HostMemoryBuffer, long, long) - 类 ai.rapids.cudf.Table 中的静态方法
读取 ORC 格式数据。
readORC(ORCOptions, DataSource) - 类 ai.rapids.cudf.Table 中的静态方法
 
readParquet(File) - 类 ai.rapids.cudf.Table 中的静态方法
使用默认 ParquetOptions 读取 Parquet 文件。
readParquet(ParquetOptions, File) - 类 ai.rapids.cudf.Table 中的静态方法
读取 Parquet 文件。
readParquet(byte[]) - 类 ai.rapids.cudf.Table 中的静态方法
读取 parquet 格式数据。
readParquet(ParquetOptions, byte[]) - 类 ai.rapids.cudf.Table 中的静态方法
读取 parquet 格式数据。
readParquet(ParquetOptions, byte[], long, long, HostMemoryAllocator) - 类 ai.rapids.cudf.Table 中的静态方法
读取 parquet 格式数据。
readParquet(ParquetOptions, byte[], long, long) - 类 ai.rapids.cudf.Table 中的静态方法
读取 parquet 格式数据。
readParquet(ParquetOptions, HostMemoryBuffer, long, long) - 类 ai.rapids.cudf.Table 中的静态方法
读取 parquet 格式数据。
readParquet(ParquetOptions, HostMemoryBuffer...) - 类 ai.rapids.cudf.Table 中的静态方法
读取 parquet 格式数据。
readParquet(ParquetOptions, DataSource) - 类 ai.rapids.cudf.Table 中的静态方法
读取 parquet 格式数据。
readTableFrom(JCudfSerialization.SerializedTableHeader, HostMemoryBuffer) - 类 ai.rapids.cudf.JCudfSerialization 中的静态方法
 
readTableFrom(InputStream, HostMemoryAllocator) - 类 ai.rapids.cudf.JCudfSerialization 中的静态方法
从给定 InputStream 读取序列化表。
readTableFrom(InputStream) - 类 ai.rapids.cudf.JCudfSerialization 中的静态方法
 
readTableIntoBuffer(InputStream, JCudfSerialization.SerializedTableHeader, HostMemoryBuffer) - 类 ai.rapids.cudf.JCudfSerialization 中的静态方法
读取表的头部后,将数据部分读取到主机侧缓冲区中。
record(Cuda.Stream) - 类 ai.rapids.cudf.Cuda.Event 中的方法
在此调用时捕获 stream 的内容。
record() - 类 ai.rapids.cudf.Cuda.Event 中的方法
在此调用时捕获默认 stream 的内容。
reduce(ReductionAggregation) - 类 ai.rapids.cudf.ColumnView 中的方法
计算列中所有行值的缩减结果。
reduce(ReductionAggregation, DType) - 类 ai.rapids.cudf.ColumnView 中的方法
计算列中所有行值的缩减结果。
ReductionAggregation - 类位于 ai.rapids.cudf
可用于 reduce 的聚合。
refCount - 类 ai.rapids.cudf.MemoryBuffer 中的变量
 
RegexFlag - 枚举位于 ai.rapids.cudf
Regex 标志设置,紧密遵循 cudf::strings::regex_flags。
RegexProgram - 类位于 ai.rapids.cudf
Regex 程序类,紧密遵循 cudf::strings::regex_program。
RegexProgram(String) - 类 ai.rapids.cudf.RegexProgram 的构造函数
RegexProgram 的构造函数
RegexProgram(String, EnumSet<RegexFlag>) - 类 ai.rapids.cudf.RegexProgram 的构造函数
RegexProgram 的构造函数
RegexProgram(String, CaptureGroups) - 类 ai.rapids.cudf.RegexProgram 的构造函数
RegexProgram 的构造函数
RegexProgram(String, EnumSet<RegexFlag>, CaptureGroups) - 类 ai.rapids.cudf.RegexProgram 的构造函数
RegexProgram 的构造函数
register(CompiledExpression, MemoryCleaner.Cleaner) - 类 ai.rapids.cudf.MemoryCleaner 中的静态方法
 
release() - 类 ai.rapids.cudf.CloseableArray 中的方法
 
releaseBuffer() - 类 ai.rapids.cudf.GatherMap 中的方法
释放底层设备缓冲区实例。
releasePinnedPoolMemoryResource(long) - 类 ai.rapids.cudf.Rmm 中的静态方法
 
releaseTable() - 类 ai.rapids.cudf.TableWithMeta 中的方法
从此元数据中获取表。
releaseWrapped() - 类 ai.rapids.cudf.RmmWrappingDeviceMemoryResource 中的方法
释放包装的设备内存资源并关闭此资源。
removeDefaultShutdownHook() - 类 ai.rapids.cudf.MemoryCleaner 中的静态方法
从 Java 默认 Runtime 中取消注册默认关闭钩子,然后返回相应的关闭 runnable。
repeat(int) - 类 ai.rapids.cudf.Table 中的方法
将表的每一行重复 count 次。
repeat(ColumnView) - 类 ai.rapids.cudf.Table 中的方法
通过重复此表的每一行来创建一个新表。
repeatString(int) - 类 ai.rapids.cudf.Scalar 中的方法
重复给定的字符串标量,重复次数由 repeatTimes 参数指定。
repeatStrings(int) - 类 ai.rapids.cudf.ColumnView 中的方法
给定一个字符串列,其中每个字符串的重复次数由 repeatTimes 参数指定。
repeatStrings(ColumnView) - 类 ai.rapids.cudf.ColumnView 中的方法
给定一个字符串列,通过根据 repeatTimes 数值列中对应行给定的次数重复每个输入字符串,生成一个输出字符串列。
replaceChildrenWithViews(int[], ColumnView[]) - 类 ai.rapids.cudf.ColumnView 中的方法
此方法接受嵌套类型,并将其子节点替换为给定的视图。注意:确保叶节点中的行数与替换它的子节点相同,否则列表可能指向列值之外的元素。
replaceListChild(ColumnView) - 类 ai.rapids.cudf.ColumnView 中的方法
此方法接受一个列表,并返回一个新列表,其中叶节点被给定的视图替换。
replaceMultiRegex(String[], ColumnView) - 类 ai.rapids.cudf.ColumnView 中的方法
对于每个字符串,将其匹配任何正则表达式模式的字符序列替换为相应的替换字符串。
replaceNulls(Scalar) - 类 ai.rapids.cudf.ColumnView 中的方法
返回一个 ColumnVector,其中任何 null 值都已替换为标量。
replaceNulls(ColumnView) - Method in class ai.rapids.cudf.ColumnView
返回一个 ColumnVector,其中的所有 null 值都被指定替换列中相应的行替换。
replaceNulls(ReplacePolicy) - Method in class ai.rapids.cudf.ColumnView
 
replaceNulls(ReplacePolicyWithColumn...) - Method in class ai.rapids.cudf.Table.GroupByOperation
 
ReplacePolicy - Enum in ai.rapids.cudf
策略,用于指定替换值相对于 null 行的位置。
ReplacePolicyWithColumn - Class in ai.rapids.cudf
特定列的替换策略
replaceRegex(String, Scalar) - Method in class ai.rapids.cudf.ColumnView
已弃用。
replaceRegex(RegexProgram, Scalar) - Method in class ai.rapids.cudf.ColumnView
对于每个字符串,使用替换字符串标量替换与给定正则表达式程序模式匹配的任何字符序列。
replaceRegex(String, Scalar, int) - Method in class ai.rapids.cudf.ColumnView
已弃用。
replaceRegex(RegexProgram, Scalar, int) - Method in class ai.rapids.cudf.ColumnView
对于每个字符串,使用替换字符串标量替换与给定正则表达式程序模式匹配的任何字符序列。
resetScopedMaximumBytesAllocated(long) - Static method in class ai.rapids.cudf.Rmm
重置用于在调试时跟踪代码段之间使用情况的 RMM 内存作用域最大计数器。
resetScopedMaximumBytesAllocated() - Static method in class ai.rapids.cudf.Rmm
重置用于在调试时跟踪代码段之间使用情况的 RMM 内存作用域最大计数器。
resetScopedMaxTotalBytesAllocated(long) - Method in class ai.rapids.cudf.RmmTrackingResourceAdaptor
 
reverseStringsOrLists() - Method in class ai.rapids.cudf.ColumnView
将当前列复制到新列,输出列中的每个字符串或列表将具有字符或元素的相反顺序。
rint() - Method in class ai.rapids.cudf.ColumnView
将浮点参数四舍五入到最接近的整数值,但将其作为浮点数返回。
Rmm - Class in ai.rapids.cudf
这是 rmm 库的绑定类。
Rmm() - Constructor for class ai.rapids.cudf.Rmm
 
Rmm.LogConf - Class in ai.rapids.cudf
RMM 分配和释放日志的发送目标。
RmmAllocationMode - Class in ai.rapids.cudf
 
RmmAllocationMode() - Constructor for class ai.rapids.cudf.RmmAllocationMode
 
RmmArenaMemoryResource<C extends RmmDeviceMemoryResource> - Class in ai.rapids.cudf
一种设备内存资源,它将预分配资源池,并从该池中进行子分配,以提高内存性能。
RmmArenaMemoryResource(C, long, boolean) - Constructor for class ai.rapids.cudf.RmmArenaMemoryResource
创建一个新的 arena 内存资源,接管其包装的 RmmDeviceMemoryResource 的所有权。
RmmCudaAsyncMemoryResource - Class in ai.rapids.cudf
一种设备内存资源,使用 `cudaMallocAsync` 和 `cudaFreeAsync` 进行分配和释放。
RmmCudaAsyncMemoryResource(long, long) - Constructor for class ai.rapids.cudf.RmmCudaAsyncMemoryResource
创建一个新的异步内存资源
RmmCudaAsyncMemoryResource(long, long, boolean) - Constructor for class ai.rapids.cudf.RmmCudaAsyncMemoryResource
创建一个新的异步内存资源
RmmCudaMemoryResource - Class in ai.rapids.cudf
一种设备内存资源,使用 `cudaMalloc` 和 `cudaFree` 进行分配和释放。
RmmCudaMemoryResource() - Constructor for class ai.rapids.cudf.RmmCudaMemoryResource
 
RmmDeviceMemoryResource - Interface in ai.rapids.cudf
分配/释放设备内存的资源。
RmmEventHandler - Interface in ai.rapids.cudf
 
RmmEventHandlerResourceAdaptor<C extends RmmDeviceMemoryResource> - Class in ai.rapids.cudf
一种设备内存资源,在特定情况下会提供回调。
RmmEventHandlerResourceAdaptor(C, RmmTrackingResourceAdaptor<?>, RmmEventHandler, boolean) - Constructor for class ai.rapids.cudf.RmmEventHandlerResourceAdaptor
创建一个新的日志记录资源适配器。
RmmException - Exception in ai.rapids.cudf
来自 RMM 分配器的异常。
RmmLimitingResourceAdaptor<C extends RmmDeviceMemoryResource> - Class in ai.rapids.cudf
一种设备内存资源,将限制分配的最大数量。
RmmLimitingResourceAdaptor(C, long, long) - Constructor for class ai.rapids.cudf.RmmLimitingResourceAdaptor
创建一个新的限制资源适配器。
RmmLoggingResourceAdaptor<C extends RmmDeviceMemoryResource> - Class in ai.rapids.cudf
一种设备内存资源,将记录交互。
RmmLoggingResourceAdaptor(C, Rmm.LogConf, boolean) - Constructor for class ai.rapids.cudf.RmmLoggingResourceAdaptor
创建一个新的日志记录资源适配器。
RmmManagedMemoryResource - Class in ai.rapids.cudf
一种设备内存资源,使用 `cudaMallocManaged` 和 `cudaFreeManaged` 进行分配和释放。
RmmManagedMemoryResource() - Constructor for class ai.rapids.cudf.RmmManagedMemoryResource
 
RmmPoolMemoryResource<C extends RmmDeviceMemoryResource> - Class in ai.rapids.cudf
一种设备内存资源,它将预分配资源池,并从该池中进行子分配,以提高内存性能。
RmmPoolMemoryResource(C, long, long) - Constructor for class ai.rapids.cudf.RmmPoolMemoryResource
创建一个新的 pooled 内存资源,接管其包装的 RmmDeviceMemoryResource 的所有权。
RmmTrackingResourceAdaptor<C extends RmmDeviceMemoryResource> - Class in ai.rapids.cudf
一种设备内存资源,将跟踪一些关于内存使用情况的基本统计信息。
RmmTrackingResourceAdaptor(C, long) - Constructor for class ai.rapids.cudf.RmmTrackingResourceAdaptor
创建一个新的跟踪资源适配器。
RmmWrappingDeviceMemoryResource<C extends RmmDeviceMemoryResource> - Class in ai.rapids.cudf
包装另一个 RmmDeviceMemoryResource 的资源
RmmWrappingDeviceMemoryResource(C) - Constructor for class ai.rapids.cudf.RmmWrappingDeviceMemoryResource
 
RollingAggregation - Class in ai.rapids.cudf
可用于滚动窗口的聚合。
RollingAggregationOnColumn - Class in ai.rapids.cudf
表中特定列的 RollingAggregation。
rollingWindow(RollingAggregation, WindowOptions) - Method in class ai.rapids.cudf.ColumnView
此函数聚合输入列中每个元素 i 周围窗口中的值。
round(int, RoundMode) - Method in class ai.rapids.cudf.ColumnView
将列中的所有值四舍五入到指定的十进制位数。
round(RoundMode) - Method in class ai.rapids.cudf.ColumnView
将列中的所有值四舍五入,十进制位数为 0。
round(int) - Method in class ai.rapids.cudf.ColumnView
将列中的所有值四舍五入到指定的十进制位数,使用 HALF_UP(默认)作为舍入方法。
round() - Method in class ai.rapids.cudf.ColumnView
使用以下默认值将列中的所有值四舍五入:十进制位数 = 0,舍入方法 = RoundMode.HALF_UP
RoundMode - Enum in ai.rapids.cudf
round 方法支持的舍入模式。
roundRobinPartition(int, int) - Method in class ai.rapids.cudf.Table
将表以轮询方式分区到指定的分区数。
rowBitCount() - Method in class ai.rapids.cudf.Table
返回 `table_view` 中所有列每行的近似累计大小(以位为单位)。
rowCount - Variable in class ai.rapids.cudf.JCudfSerialization.SerializedColumnHeader
 
rowNumber() - Static method in class ai.rapids.cudf.RollingAggregation
获取行号。
rows - Variable in class ai.rapids.cudf.ColumnView
 
rows - Variable in class ai.rapids.cudf.HostColumnVectorCore
 
rstrip() - Method in class ai.rapids.cudf.ColumnView
移除字符串末尾的空白字符。
rstrip(Scalar) - Method in class ai.rapids.cudf.ColumnView
移除每个字符串末尾指定的字符。

S

sample(long, boolean, long) - Method in class ai.rapids.cudf.Table
从表中随机采样 `n` 个样本。注意:不保留顺序。示例:输入:{col1: {1, 2, 3, 4, 5}, col2: {6, 7, 8, 9, 10}} n: 3 replacement: false 输出:{col1: {3, 1, 4}, col2: {8, 6, 9}} replacement: true 输出:{col1: {3, 1, 1}, col2: {8, 6, 6}} 如果 `n` > 表行数且 `replacement` == FALSE,则抛出“logic_error”。
Scalar - Class in ai.rapids.cudf
一个标量值。
Scalar(DType, long) - Constructor for class ai.rapids.cudf.Scalar
从原生句柄和类型创建标量值的构造函数。
scan(ScanAggregation, ScanType, NullPolicy) - Method in class ai.rapids.cudf.ColumnView
计算列的 scan(扫描)操作。
scan(ScanAggregation, ScanType) - Method in class ai.rapids.cudf.ColumnView
计算排除 null 的列的 scan(扫描)操作。
scan(ScanAggregation) - Method in class ai.rapids.cudf.ColumnView
计算排除 null 的列的 inclusive scan(包含式扫描)操作。
scan(GroupByScanAggregationOnColumn...) - Method in class ai.rapids.cudf.Table.GroupByOperation
 
ScanAggregation - Class in ai.rapids.cudf
可用于 scan(扫描)的聚合。
ScanType - Enum in ai.rapids.cudf
扫描操作类型。
scatter(ColumnView, Table) - Method in class ai.rapids.cudf.Table
将源表中的值scatter(分散)到目标表中,返回新的结果表。
scatter(Scalar[], ColumnView, Table) - Static method in class ai.rapids.cudf.Table
将源行中的值scatter(分散)到目标表中,返回新的结果表。
Schema - Class in ai.rapids.cudf
要读取的数据的 schema(模式)。
Schema.Builder - Class in ai.rapids.cudf
 
second() - Method in class ai.rapids.cudf.ColumnView
从具有时间分辨率的时间戳中获取秒。
segmentedGather(ColumnView) - Method in class ai.rapids.cudf.ColumnView
对列表列中每行列表元素内的元素进行分段 gather(收集)。
segmentedGather(ColumnView, OutOfBoundsPolicy) - Method in class ai.rapids.cudf.ColumnView
对列表列中每行列表元素内的元素进行分段 gather(收集)。
segmentedReduce(ColumnView, SegmentedReductionAggregation) - Method in class ai.rapids.cudf.ColumnView
执行分段 reduce(归约),其中偏移量列指示要组合的本列中的哪些组。
segmentedReduce(ColumnView, SegmentedReductionAggregation, DType) - Method in class ai.rapids.cudf.ColumnView
执行分段 reduce(归约),其中偏移量列指示要组合的本列中的哪些组。
segmentedReduce(ColumnView, SegmentedReductionAggregation, NullPolicy, DType) - Method in class ai.rapids.cudf.ColumnView
执行分段 reduce(归约),其中偏移量列指示要组合的本列中的哪些组。
SegmentedReductionAggregation - Class in ai.rapids.cudf
可用于 reduce 的聚合。
sequence(Scalar, Scalar, int) - Static method in class ai.rapids.cudf.ColumnVector
创建一个长度为 rows 的新向量,从 initialValue 开始,每次增加 step。
sequence(Scalar, int) - Static method in class ai.rapids.cudf.ColumnVector
创建一个长度为 rows 的新向量,从 initialValue 开始,每次增加 1。
sequence(ColumnView, ColumnView) - Static method in class ai.rapids.cudf.ColumnVector
创建一个列表列,其中每一行都是一个值序列,从 `start` 值开始,每次增加 1,其基数由 `size` 值指定。
sequence(ColumnView, ColumnView, ColumnView) - Static method in class ai.rapids.cudf.ColumnVector
创建一个列表列,其中每一行都是一个值序列,从 `start` 值开始,每次增加 `step` 值,其基数由 `size` 值指定。
SerializedTableHeader(DataInputStream) - Constructor for class ai.rapids.cudf.JCudfSerialization.SerializedTableHeader
 
SerializedTableHeader(int) - Constructor for class ai.rapids.cudf.JCudfSerialization.SerializedTableHeader
仅包含行数的表的构造函数(无列)
set(int, T) - Method in class ai.rapids.cudf.CloseableArray
 
set(HostMemoryAllocator) - Static method in class ai.rapids.cudf.DefaultHostMemoryAllocator
默认设置新的默认主机内存分配器实现。
setBoolean(long, boolean) - Method in class ai.rapids.cudf.HostMemoryBuffer
在该偏移量处设置布尔值
setByte(long, byte) - Method in class ai.rapids.cudf.HostMemoryBuffer
在该偏移量处设置字节值
setBytes(long, byte[], long, long) - Method in class ai.rapids.cudf.HostMemoryBuffer
将数组中的一组字节复制到缓冲区偏移量处。
setCapture(CaptureGroups) - Method in class ai.rapids.cudf.RegexProgram
重置此实例的捕获组设置
setCudfPinnedPoolMemoryResource(long) - Static method in class ai.rapids.cudf.Rmm
 
setCurrentDeviceResource(RmmDeviceMemoryResource, RmmDeviceMemoryResource, boolean) - Static method in class ai.rapids.cudf.Rmm
设置 RMM 应在所有分配和释放中使用的当前设备资源。
setDevice(int) - Static method in class ai.rapids.cudf.Cuda
设置当前设备的 ID。
setDouble(long, double) - Method in class ai.rapids.cudf.HostMemoryBuffer
在该偏移量处设置 Double 值
setDoubles(long, double[], long, long) - Method in class ai.rapids.cudf.HostMemoryBuffer
将数组中的一组双精度浮点数复制到缓冲区偏移量处。
setEventHandler(ColumnVector.EventHandler) - Method in class ai.rapids.cudf.ColumnVector
为此向量设置事件处理程序。
setEventHandler(HostColumnVector.EventHandler) - Method in class ai.rapids.cudf.HostColumnVector
为此主机向量设置事件处理程序。
setEventHandler(MemoryBuffer.EventHandler) - Method in class ai.rapids.cudf.MemoryBuffer
为此缓冲区设置事件处理程序。
setEventHandler(RmmEventHandler) - Static method in class ai.rapids.cudf.Rmm
设置在 RMM 事件发生时(例如:分配失败)调用的事件处理程序。
setEventHandler(RmmEventHandler, boolean) - Static method in class ai.rapids.cudf.Rmm
设置在 RMM 事件发生时(例如:分配失败)调用的事件处理程序,并可选择启用调试模式(每次分配和释放时进行回调)。
setFlags(EnumSet<RegexFlag>) - Method in class ai.rapids.cudf.RegexProgram
重置此实例的正则表达式标志设置
setFloat(long, float) - Method in class ai.rapids.cudf.HostMemoryBuffer
在该偏移量处设置 Float 值
setFloats(long, float[], long, long) - Method in class ai.rapids.cudf.HostMemoryBuffer
将数组中的一组单精度浮点数复制到缓冲区偏移量处。
setInt(long, int) - Method in class ai.rapids.cudf.HostMemoryBuffer
在该偏移量处设置 Integer 值
setInts(long, int[], long, long) - Method in class ai.rapids.cudf.HostMemoryBuffer
将数组中的一组整数复制到缓冲区偏移量处。
setKernelPinnedCopyThreshold(long) - Static method in class ai.rapids.cudf.Cudf
小于阈值的 cuDF 复制将使用内核进行复制,而不是使用 cudaMemcpyAsync。
setLong(long, long) - Method in class ai.rapids.cudf.HostMemoryBuffer
在该偏移量处设置 Long 值
setLongs(long, long[], long, long) - Method in class ai.rapids.cudf.HostMemoryBuffer
将数组中的一组长整数复制到缓冲区偏移量处。
setMemory(long, long, byte) - Method in class ai.rapids.cudf.HostMemoryBuffer
重复设置此缓冲区中的值
setNullAt(long) - Method in class ai.rapids.cudf.HostColumnVector.Builder
将特定索引设置为 null。
setPinnedAllocationThreshold(long) - Static method in class ai.rapids.cudf.Cudf
小于阈值的 cudf 分配将使用固定(pinned)主机内存资源。
setShort(long, short) - Method in class ai.rapids.cudf.HostMemoryBuffer
在该偏移量处设置 Short 值
setShorts(long, short[], long, long) - Method in class ai.rapids.cudf.HostMemoryBuffer
将数组中的一组短整数复制到缓冲区偏移量处。
shiftLeft(BinaryOperable, DType) - Method in interface ai.rapids.cudf.BinaryOperable
将此向量的值按 shiftBy 进行位左移。
shiftLeft(BinaryOperable) - Method in interface ai.rapids.cudf.BinaryOperable
将此向量的值按 shiftBy 进行位左移。
shiftRight(BinaryOperable, DType) - Method in interface ai.rapids.cudf.BinaryOperable
将此向量按 shiftBy 进行位右移。
shiftRight(BinaryOperable) - Method in interface ai.rapids.cudf.BinaryOperable
将此向量按 shiftBy 进行位右移。
shiftRightUnsigned(BinaryOperable, DType) - Method in interface ai.rapids.cudf.BinaryOperable
此方法将此向量的值按 shiftBy 进行无符号位右移。
shiftRightUnsigned(BinaryOperable) - Method in interface ai.rapids.cudf.BinaryOperable
此方法将此向量的值按 shiftBy 进行无符号位右移。
shouldCudfPruneSchema() - Method in class ai.rapids.cudf.JSONOptions
 
shutdown() - Static method in class ai.rapids.cudf.PinnedMemoryPool
关闭 RMM 的 pool_memory_resource,并将我们的引用设为 null。
shutdown() - Static method in class ai.rapids.cudf.Rmm
关闭任何已初始化的 RMM 实例。
shutdown(long, long, TimeUnit) - Static method in class ai.rapids.cudf.Rmm
关闭任何已初始化的 RMM 实例。
sin() - Method in class ai.rapids.cudf.ColumnView
计算 sin(正弦),输出类型与输入相同。
sinh() - Method in class ai.rapids.cudf.ColumnView
计算双曲 sin(正弦),输出类型与输入相同。
size() - Method in class ai.rapids.cudf.CloseableArray
 
size() - Method in class ai.rapids.cudf.DataSource
获取源的大小(以字节为单位)。
size() - Method in class ai.rapids.cudf.MultiBufferDataSource
 
slice(int...) - Method in class ai.rapids.cudf.ColumnView
根据一组索引将列(包括 null 值)切分为一组列。
slice(long, long) - Method in class ai.rapids.cudf.CudaMemoryBuffer
切分出设备缓冲区的一部分。
slice(long, long) - Method in class ai.rapids.cudf.CuFileBuffer
 
slice(long, long) - Method in class ai.rapids.cudf.DeviceMemoryBuffer
切分出设备缓冲区的一部分。
slice(long, long) - Method in class ai.rapids.cudf.DeviceMemoryBufferView
目前我们不需要切分视图。
slice(long, long) - Method in class ai.rapids.cudf.HostMemoryBuffer
切分出主机缓冲区的一部分。
slice(long, long) - Method in class ai.rapids.cudf.MemoryBuffer
切分出缓冲区的一部分。
sliceWithCopy(long, long) - Method in class ai.rapids.cudf.BaseDeviceMemoryBuffer
切分出设备缓冲区的一部分,并复制它而不是进行引用计数。
sliceWithCopy(long, long) - Method in class ai.rapids.cudf.HostMemoryBuffer
切分出主机缓冲区的一部分,实际上复制数据。
sortOrder(OrderByArg...) - Method in class ai.rapids.cudf.Table
获取可用于对数据进行排序的 gather(收集)映射。
split(int...) - Method in class ai.rapids.cudf.ColumnView
根据一组索引将列(包括 null 值)拆分为一组列。
splitAsViews(int...) - Method in class ai.rapids.cudf.ColumnView
根据一组索引将 ColumnView(包括 null 值)拆分为一组 ColumnView。
sqrt() - Method in class ai.rapids.cudf.ColumnView
计算 sqrt(平方根),输出类型与输入相同。
standardDeviation() - Method in class ai.rapids.cudf.ColumnView
返回列中所有值的样本标准差,返回 FLOAT64 标量,除非列类型是 FLOAT32,则返回 FLOAT32 标量。
standardDeviation(DType) - Method in class ai.rapids.cudf.ColumnView
返回列中所有值的样本标准差,返回指定类型的标量。
standardDeviation() - Static method in class ai.rapids.cudf.GroupByAggregation
标准差聚合,自由度的 delta 为 1。
standardDeviation(int) - Static method in class ai.rapids.cudf.GroupByAggregation
标准差聚合。
standardDeviation() - Static method in class ai.rapids.cudf.ReductionAggregation
标准差聚合,自由度的 delta 为 1。
standardDeviation(int) - Static method in class ai.rapids.cudf.ReductionAggregation
标准差聚合。
standardDeviation() - Static method in class ai.rapids.cudf.RollingAggregation
滚动窗口标准差,自由度的 delta (DDOF) 为 1。
standardDeviation(int) - Static method in class ai.rapids.cudf.RollingAggregation
滚动窗口标准差,可配置自由度的 delta (DDOF)。
startsWith(Scalar) - Method in class ai.rapids.cudf.ColumnView
检查列中的每个字符串是否以指定的比较字符串开头,结果是布尔值的并行列。
Stream(boolean) - Constructor for class ai.rapids.cudf.Cuda.Stream
创建一个新的 CUDA 流
StreamedTableReader - Interface in ai.rapids.cudf
提供一个接口,用于从单个输入源读取多个表。
strictValidation() - Method in class ai.rapids.cudf.JSONOptions
 
STRING - Static variable in class ai.rapids.cudf.DType
 
stringConcatenate(ColumnView[]) - Static method in class ai.rapids.cudf.ColumnVector
将字符串列连接起来,将每列中相应的行组合成一个新列的单个字符串行,组合的每个字符串之间不插入分隔符字符串,并保留组合行中的 null 值。
stringConcatenate(Scalar, Scalar, ColumnView[]) - Static method in class ai.rapids.cudf.ColumnVector
将字符串列连接起来,将每列中相应的行组合成一个新列的单个字符串行。
stringConcatenate(Scalar, Scalar, ColumnView[], boolean) - Static method in class ai.rapids.cudf.ColumnVector
将字符串列连接起来,将每列中相应的行组合成一个新列的单个字符串行。
stringConcatenate(ColumnView[], ColumnView) - Static method in class ai.rapids.cudf.ColumnVector
使用为每行指定的分隔符将字符串列连接起来,并将结果作为字符串列返回。
stringConcatenate(ColumnView[], ColumnView, Scalar, Scalar, boolean) - Static method in class ai.rapids.cudf.ColumnVector
使用为每行指定的分隔符将字符串列连接起来,并将结果作为字符串列返回。
stringConcatenateListElements(ColumnView) - Method in class ai.rapids.cudf.ColumnView
给定一个字符串列表列(每行是一个字符串列表),连接每行内的字符串并返回单个字符串列结果。
stringConcatenateListElements(ColumnView, Scalar, Scalar, boolean, boolean) - Method in class ai.rapids.cudf.ColumnView
给定一个字符串列表列(每行是一个字符串列表),连接每行内的字符串并返回单个字符串列结果。
stringConcatenateListElements(Scalar, Scalar, boolean, boolean) - Method in class ai.rapids.cudf.ColumnView
给定一个字符串列表列(每行是一个字符串列表),连接每行内的字符串并返回单个字符串列结果。
stringContains(Scalar) - Method in class ai.rapids.cudf.ColumnView
检查列中的每个字符串是否包含指定的比较字符串,结果是布尔值的并行列。
stringContains(ColumnView) - Method in class ai.rapids.cudf.ColumnView
 
stringLocate(Scalar) - Method in class ai.rapids.cudf.ColumnView
查找给定字符串在列的每行中第一次出现的起始索引。
stringLocate(Scalar, int) - Method in class ai.rapids.cudf.ColumnView
查找给定字符串在列的每行中第一次出现的起始索引。
stringLocate(Scalar, int, int) - Method in class ai.rapids.cudf.ColumnView
查找给定字符串在列的每行中第一次出现的起始索引。
stringReplace(Scalar, Scalar) - Method in class ai.rapids.cudf.ColumnView
返回一个新字符串列,其中每个字符串内的目标字符串被指定的替换字符串替换。
stringReplace(ColumnView, ColumnView) - Method in class ai.rapids.cudf.ColumnView
返回一个新字符串列,其中每个字符串中的目标字符串被相应的替换字符串替换。
stringReplaceWithBackrefs(String, String) - Method in class ai.rapids.cudf.ColumnView
已弃用。
stringReplaceWithBackrefs(RegexProgram, String) - Method in class ai.rapids.cudf.ColumnView
对于每个字符串,使用反向引用替换模板替换与给定正则表达式程序模式匹配的任何字符序列。
stringSplit(String, int, boolean) - Method in class ai.rapids.cudf.ColumnView
已弃用。
stringSplit(RegexProgram, int) - Method in class ai.rapids.cudf.ColumnView
使用指定的正则表达式程序模式拆分每个字符串,返回一个列列表。
stringSplit(String, boolean) - Method in class ai.rapids.cudf.ColumnView
已弃用。
stringSplit(String, int) - Method in class ai.rapids.cudf.ColumnView
使用指定的字符串文字分隔符拆分每个字符串,返回一个列列表。
stringSplit(String) - Method in class ai.rapids.cudf.ColumnView
使用指定的字符串文字分隔符拆分每个字符串,返回一个列列表。
stringSplit(RegexProgram) - Method in class ai.rapids.cudf.ColumnView
使用指定的正则表达式程序模式拆分每个字符串,返回一个列列表。
stringSplitRecord(String, int, boolean) - Method in class ai.rapids.cudf.ColumnView
已弃用。
stringSplitRecord(RegexProgram, int) - Method in class ai.rapids.cudf.ColumnView
返回一个列,其中包含字符串列表,每个列表是通过使用指定的正则表达式程序模式拆分相应的输入字符串而生成的。
stringSplitRecord(String, boolean) - Method in class ai.rapids.cudf.ColumnView
已弃用。
stringSplitRecord(String, int) - Method in class ai.rapids.cudf.ColumnView
返回一个列,其中包含字符串列表,每个列表是通过使用指定的字符串文字分隔符拆分相应的输入字符串而生成的。
stringSplitRecord(String) - Method in class ai.rapids.cudf.ColumnView
返回一个列,其中包含字符串列表,每个列表是通过使用指定的字符串文字分隔符拆分相应的输入字符串而生成的。
stringSplitRecord(RegexProgram) - Method in class ai.rapids.cudf.ColumnView
返回一个列,其中包含字符串列表,每个列表是通过使用指定的正则表达式程序模式拆分相应的输入字符串而生成的。
strip() - Method in class ai.rapids.cudf.ColumnView
移除字符串开头和结尾的空白字符。
strip(Scalar) - Method in class ai.rapids.cudf.ColumnView
移除每个字符串开头和结尾指定的字符。
stripQuotesFromSingleStrings(boolean) - Method in class ai.rapids.cudf.GetJsonObjectOptions.Builder
 
STRUCT - Static variable in class ai.rapids.cudf.DType
 
structBuilder(String, boolean) - Static method in class ai.rapids.cudf.ColumnWriterOptions
为名为 'name' 的列创建一个 StructBuilder
structBuilder(String, boolean, int) - Static method in class ai.rapids.cudf.ColumnWriterOptions
为名为 'name' 的列创建一个 StructBuilder
structBuilder(String) - Static method in class ai.rapids.cudf.ColumnWriterOptions
为名为 'name' 的列创建一个 StructBuilder
StructBuilder(String, boolean) - Constructor for class ai.rapids.cudf.ColumnWriterOptions.StructBuilder
 
StructBuilder(String, boolean, int) - Constructor for class ai.rapids.cudf.ColumnWriterOptions.StructBuilder
 
StructColumnWriterOptions(ColumnWriterOptions.AbstractStructBuilder) - Constructor for class ai.rapids.cudf.ColumnWriterOptions.StructColumnWriterOptions
 
StructData(List<Object>) - Constructor for class ai.rapids.cudf.HostColumnVector.StructData
 
StructData(Object...) - Constructor for class ai.rapids.cudf.HostColumnVector.StructData
 
structFromColumnViews(ColumnView...s) - Static method in class ai.rapids.cudf.Scalar
从 ColumnView 创建一个 struct 类型的标量。
structFromNull(HostColumnVector.DataType...s) - Static method in class ai.rapids.cudf.Scalar
创建一个 struct 类型的 null 标量。
StructType(boolean, List<HostColumnVector.DataType>) - Constructor for class ai.rapids.cudf.HostColumnVector.StructType
 
StructType(boolean, HostColumnVector.DataType...) - Constructor for class ai.rapids.cudf.HostColumnVector.StructType
 
sub(BinaryOperable, DType) - Method in interface ai.rapids.cudf.BinaryOperable
将一个向量从另一个向量中减去,并使用给定的输出类型。
sub(BinaryOperable) - Method in interface ai.rapids.cudf.BinaryOperable
将一个向量从另一个向量中减去。
substring(int) - Method in class ai.rapids.cudf.ColumnView
返回一个新字符串列,其中包含提供的列中字符串的子串。
substring(int, int) - Method in class ai.rapids.cudf.ColumnView
返回一个新字符串列,其中包含提供的列中字符串的子串。
substring(ColumnView, ColumnView) - Method in class ai.rapids.cudf.ColumnView
返回一个新字符串列,其中包含提供的列中字符串的子串,对每个字符串使用唯一的范围
subVector(int) - Method in class ai.rapids.cudf.ColumnView
返回从开始索引(包含)到向量末尾的子向量。
subVector(int, int) - Method in class ai.rapids.cudf.ColumnView
返回一个子向量。
sum() - Method in class ai.rapids.cudf.ColumnView
计算列中所有值的总和,返回与此列类型相同的标量。
sum(DType) - Method in class ai.rapids.cudf.ColumnView
计算列中所有值的总和,返回指定类型的标量。
sum() - Static method in class ai.rapids.cudf.GroupByAggregation
总和聚合
sum() - Static method in class ai.rapids.cudf.GroupByScanAggregation
总和聚合
sum() - Static method in class ai.rapids.cudf.ReductionAggregation
总和聚合
sum() - Static method in class ai.rapids.cudf.RollingAggregation
滚动窗口总和
sum() - Static method in class ai.rapids.cudf.ScanAggregation
总和聚合
sum() - Static method in class ai.rapids.cudf.SegmentedReductionAggregation
总和聚合
sumOfSquares() - Method in class ai.rapids.cudf.ColumnView
返回列中所有值的平方和,返回与此列类型相同的标量。
sumOfSquares(DType) - Method in class ai.rapids.cudf.ColumnView
返回列中所有值的平方和,返回指定类型的标量。
sumOfSquares() - Static method in class ai.rapids.cudf.ReductionAggregation
平方和归约。
supportsDeviceRead() - Method in class ai.rapids.cudf.DataSource
如果支持直接读取到设备则返回 true,否则返回 false。
supportsDeviceRead() - Method in class ai.rapids.cudf.MultiBufferDataSource
 
sync() - Method in class ai.rapids.cudf.Cuda.Event
阻塞线程以等待事件完成。
sync() - Method in class ai.rapids.cudf.Cuda.Stream
阻塞线程以等待此流上的所有待处理工作完成。

T

Table - Class in ai.rapids.cudf
表示 ColumnVectors 集合以及可对它们集体执行的操作的类。
Table(ColumnVector...) - Constructor for class ai.rapids.cudf.Table
Table 类会复制传递给它的 ColumnVector 数组。
Table(long[]) - Constructor for class ai.rapids.cudf.Table
从现有的设备上 cudf::column 指针数组创建一个 Table。
Table.DuplicateKeepOption - Enum in ai.rapids.cudf
枚举,用于指定哪些重复行/元素将被复制到输出。
Table.GroupByOperation - Class in ai.rapids.cudf
表示 groupby 操作的类
Table.TableOperation - Class in ai.rapids.cudf
 
Table.TestBuilder - Class in ai.rapids.cudf
在 GPU 上创建包含 CPU 数据的表。
TableAndRowCountPair(int, ContiguousTable) - Constructor for class ai.rapids.cudf.JCudfSerialization.TableAndRowCountPair
 
TableDebug - Class in ai.rapids.cudf
 
TableDebug.Builder - Class in ai.rapids.cudf
 
TableDebug.Output - Enum in ai.rapids.cudf
 
TableReference - Enum in ai.rapids.cudf.ast
可在 AST 中引用的表枚举。
TableWithMeta - Class in ai.rapids.cudf
一个表以及一些关于该表的元数据。
TableWithMeta.NestedChildren - Class in ai.rapids.cudf
 
TableWriter - Class in ai.rapids.cudf
提供一个接口,用于分多步写入 Table 信息。
TableWriter.WriteStatistics - Class in ai.rapids.cudf
 
tan() - Method in class ai.rapids.cudf.ColumnView
计算 tan(正切),输出类型与输入相同。
tanh() - Method in class ai.rapids.cudf.ColumnView
计算双曲 tan(正切),输出类型与输入相同。
TestBuilder() - Constructor for class ai.rapids.cudf.Table.TestBuilder
 
TIMESTAMP_DAYS - Static variable in class ai.rapids.cudf.DType
 
TIMESTAMP_MICROSECONDS - Static variable in class ai.rapids.cudf.DType
 
TIMESTAMP_MILLISECONDS - Static variable in class ai.rapids.cudf.DType
 
TIMESTAMP_NANOSECONDS - 类中的静态变量 ai.rapids.cudf.DType
 
TIMESTAMP_SECONDS - 类中的静态变量 ai.rapids.cudf.DType
 
timestampAscending() - 类中的方法 ai.rapids.cudf.WindowOptions.Builder
已弃用。
使用 orderByAscending()
timestampColumnIndex(int) - 类中的方法 ai.rapids.cudf.WindowOptions.Builder
已弃用。
使用 orderByColumnIndex(int index)
timestampDayColumn(Integer...) - 类中的方法 ai.rapids.cudf.Table.TestBuilder
 
timestampDaysFromBoxedInts(Integer...) - 类中的静态方法 ai.rapids.cudf.ColumnVector
从给定值创建新的向量。
timestampDaysFromBoxedInts(Integer...) - 类中的静态方法 ai.rapids.cudf.HostColumnVector
从给定值创建新的向量。
timestampDaysFromInt(int) - 类中的静态方法 ai.rapids.cudf.Scalar
 
timestampDaysFromInt(Integer) - 类中的静态方法 ai.rapids.cudf.Scalar
 
timestampDescending() - 类中的方法 ai.rapids.cudf.WindowOptions.Builder
已弃用。
使用 orderByDescending()
timestampFromLong(DType, long) - 类中的静态方法 ai.rapids.cudf.Scalar
 
timestampFromLong(DType, Long) - 类中的静态方法 ai.rapids.cudf.Scalar
 
timestampMicrosecondsColumn(Long...) - 类中的方法 ai.rapids.cudf.Table.TestBuilder
 
timestampMicroSecondsFromBoxedLongs(Long...) - 类中的静态方法 ai.rapids.cudf.ColumnVector
从给定值创建新的向量。
timestampMicroSecondsFromBoxedLongs(Long...) - 类中的静态方法 ai.rapids.cudf.HostColumnVector
从给定值创建新的向量。
timestampMicroSecondsFromLongs(long...) - 类中的静态方法 ai.rapids.cudf.ColumnVector
从给定值创建新的向量。
timestampMicroSecondsFromLongs(long...) - 类中的静态方法 ai.rapids.cudf.HostColumnVector
从给定值创建新的向量。
timestampMillisecondsColumn(Long...) - 类中的方法 ai.rapids.cudf.Table.TestBuilder
 
timestampMilliSecondsFromBoxedLongs(Long...) - 类中的静态方法 ai.rapids.cudf.ColumnVector
从给定值创建新的向量。
timestampMilliSecondsFromBoxedLongs(Long...) - 类中的静态方法 ai.rapids.cudf.HostColumnVector
从给定值创建新的向量。
timestampMilliSecondsFromLongs(long...) - 类中的静态方法 ai.rapids.cudf.ColumnVector
从给定值创建新的向量。
timestampMilliSecondsFromLongs(long...) - 类中的静态方法 ai.rapids.cudf.HostColumnVector
从给定值创建新的向量。
timestampNanosecondsColumn(Long...) - 类中的方法 ai.rapids.cudf.Table.TestBuilder
 
timestampNanoSecondsFromBoxedLongs(Long...) - 类中的静态方法 ai.rapids.cudf.ColumnVector
从给定值创建新的向量。
timestampNanoSecondsFromBoxedLongs(Long...) - 类中的静态方法 ai.rapids.cudf.HostColumnVector
从给定值创建新的向量。
timestampNanoSecondsFromLongs(long...) - 类中的静态方法 ai.rapids.cudf.ColumnVector
从给定值创建新的向量。
timestampNanoSecondsFromLongs(long...) - 类中的静态方法 ai.rapids.cudf.HostColumnVector
从给定值创建新的向量。
timestampSecondsColumn(Long...) - 类中的方法 ai.rapids.cudf.Table.TestBuilder
 
timestampSecondsFromBoxedLongs(Long...) - 类中的静态方法 ai.rapids.cudf.ColumnVector
从给定值创建新的向量。
timestampSecondsFromBoxedLongs(Long...) - 类中的静态方法 ai.rapids.cudf.HostColumnVector
从给定值创建新的向量。
timestampSecondsFromLongs(long...) - 类中的静态方法 ai.rapids.cudf.ColumnVector
从给定值创建新的向量。
timestampSecondsFromLongs(long...) - 类中的静态方法 ai.rapids.cudf.HostColumnVector
从给定值创建新的向量。
title(long) - 类中的静态方法 ai.rapids.cudf.ColumnView
 
toColumnView(long, int) - 类中的方法 ai.rapids.cudf.GatherMap
创建一个列视图,可用于执行聚集操作。
toContiguousTable() - 类中的方法 ai.rapids.cudf.JCudfSerialization.HostConcatResult
从这个主机连接的结果在设备内存中构建一个连续的表
toHex() - 类中的方法 ai.rapids.cudf.ColumnView
将此整数列转换为十六进制列并返回一个新的字符串列。任何空条目将在输出列中产生相应的空条目。
toNativeId() - 枚举 ai.rapids.cudf.nvcomp.CompressionType 中的方法
获取类型的本地代码标识符
toString() - 类中的方法 ai.rapids.cudf.ArrowColumnBuilder
 
toString() - 类中的方法 ai.rapids.cudf.ColumnVector.OffHeapState
 
toString() - 类中的方法 ai.rapids.cudf.ColumnVector
 
toString() - 类中的方法 ai.rapids.cudf.ColumnView
 
toString() - 类中的方法 ai.rapids.cudf.Cuda.Event
 
toString() - 类中的方法 ai.rapids.cudf.Cuda.Stream
 
toString() - 类中的方法 ai.rapids.cudf.DType
 
toString() - 类中的方法 ai.rapids.cudf.HostColumnVector.Builder
 
toString() - 类中的方法 ai.rapids.cudf.HostColumnVector.ColumnBuilder
 
toString() - 类中的方法 ai.rapids.cudf.HostColumnVector
 
toString() - 类中的方法 ai.rapids.cudf.HostColumnVectorCore.OffHeapState
 
toString() - 类中的方法 ai.rapids.cudf.HostColumnVectorCore
 
toString() - 类中的方法 ai.rapids.cudf.MemoryBuffer
 
toString() - 类中的方法 ai.rapids.cudf.OrderByArg
 
toString() - 类中的方法 ai.rapids.cudf.RmmArenaMemoryResource
 
toString() - 类中的方法 ai.rapids.cudf.RmmCudaAsyncMemoryResource
 
toString() - 类中的方法 ai.rapids.cudf.RmmCudaMemoryResource
 
toString() - 类中的方法 ai.rapids.cudf.RmmEventHandlerResourceAdaptor
 
toString() - 类中的方法 ai.rapids.cudf.RmmLimitingResourceAdaptor
 
toString() - 类中的方法 ai.rapids.cudf.RmmLoggingResourceAdaptor
 
toString() - 类中的方法 ai.rapids.cudf.RmmManagedMemoryResource
 
toString() - 类中的方法 ai.rapids.cudf.RmmPoolMemoryResource
 
toString() - 类中的方法 ai.rapids.cudf.RmmTrackingResourceAdaptor
 
toString() - 类中的方法 ai.rapids.cudf.Scalar
 
toString() - 类中的方法 ai.rapids.cudf.Schema
 
toString() - 类中的方法 ai.rapids.cudf.Table
 
toString() - 类中的方法 ai.rapids.cudf.TableWithMeta.NestedChildren
 
total - 类中的变量 ai.rapids.cudf.CudaMemInfo
总内存大小(字节)
toTitle() - 类中的方法 ai.rapids.cudf.ColumnView
返回一个字符串列,其中,对于输入中的每个字符串行,空格后的第一个字符转换为大写,而单词中所有剩余字符转换为小写。
transform(String, boolean) - 类中的方法 ai.rapids.cudf.ColumnView
使用自定义函数转换向量。
trueDiv(BinaryOperable, DType) - 接口 ai.rapids.cudf.BinaryOperable 中的方法
将一个向量除以另一个向量,并在两者之间转换为 FLOAT64,使用给定的输出类型。
trueDiv(BinaryOperable) - 接口 ai.rapids.cudf.BinaryOperable 中的方法
将一个向量除以另一个向量,并在两者之间转换为 FLOAT64。
tryAllocate(long) - 类中的静态方法 ai.rapids.cudf.PinnedMemoryPool
创建固定主机内存缓冲区的工厂方法。
type - 类中的变量 ai.rapids.cudf.ColumnView
 
type - 类中的变量 ai.rapids.cudf.HostColumnVectorCore
 

U

UINT16 - 类中的静态变量 ai.rapids.cudf.DType
 
UINT32 - 类中的静态变量 ai.rapids.cudf.DType
 
UINT64 - 类中的静态变量 ai.rapids.cudf.DType
 
UINT8 - 类中的静态变量 ai.rapids.cudf.DType
 
unaryOp(UnaryOp) - 类中的方法 ai.rapids.cudf.ColumnView
多个不同的一元操作。
UnaryOp - 枚举 ai.rapids.cudf.ai.rapids.cudf
数学一元操作。
UnaryOperation - 类 ai.rapids.cudf.ast.ai.rapids.cudf.ast
由运算符和操作数组成的一元操作。
UnaryOperation(UnaryOperator, AstExpression) - 类 ai.rapids.cudf.ast.UnaryOperation 的构造函数
 
UnaryOperator - 枚举 ai.rapids.cudf.ast.ai.rapids.cudf.ast
可出现在一元操作中的 AST 运算符枚举。
unboundedFollowing() - 类中的方法 ai.rapids.cudf.WindowOptions.Builder
 
unboundedPreceding() - 类中的方法 ai.rapids.cudf.WindowOptions.Builder
 
UNKNOWN_NULL_COUNT - 类中的静态变量 ai.rapids.cudf.ColumnVector.OffHeapState
 
UNKNOWN_NULL_COUNT - 类中的静态变量 ai.rapids.cudf.ColumnView
 
UNKNOWN_PRECISION - 类中的静态变量 ai.rapids.cudf.ColumnWriterOptions
 
unpackHostColumnVectors(JCudfSerialization.SerializedTableHeader, HostMemoryBuffer) - 类中的静态方法 ai.rapids.cudf.JCudfSerialization
将序列化的连续表反序列化为主机列数组。
unquotedControlChars() - 类中的方法 ai.rapids.cudf.JSONOptions
 
upper() - 类中的方法 ai.rapids.cudf.ColumnView
将字符串转换为大写。
upperBound(boolean[], Table, boolean[]) - 类中的方法 ai.rapids.cudf.Table
在排序表中找到要插入值以保持顺序的最大索引。
upperBound(Table, OrderByArg...) - 类中的方法 ai.rapids.cudf.Table
在排序表中找到要插入值以保持顺序的最大索引。
urlDecode() - 类中的方法 ai.rapids.cudf.ColumnView
将所有以'%'开头的字符序列转换为字符代码点,并将后面的2个字符解释为十六进制值以创建代码点。
urlEncode() - 类中的方法 ai.rapids.cudf.ColumnView
将大多数非 ascii 字符和控制字符转换为以'%'为前缀的 UTF-8 十六进制代码点。

V

valid - 类中的变量 ai.rapids.cudf.HostColumnVectorCore.OffHeapState
 
valueOf(String) - 枚举 ai.rapids.cudf.ast.AstExpression.ExpressionType 中的静态方法
返回具有指定名称的此枚举类型的常数。
valueOf(String) - 枚举 ai.rapids.cudf.ast.BinaryOperator 中的静态方法
返回具有指定名称的此枚举类型的常数。
valueOf(String) - 枚举 ai.rapids.cudf.ast.TableReference 中的静态方法
返回具有指定名称的此枚举类型的常数。
valueOf(String) - 枚举 ai.rapids.cudf.ast.UnaryOperator 中的静态方法
返回具有指定名称的此枚举类型的常数。
valueOf(String) - 枚举 ai.rapids.cudf.BinaryOp 中的静态方法
返回具有指定名称的此枚举类型的常数。
valueOf(String) - 枚举 ai.rapids.cudf.BufferType 中的静态方法
返回具有指定名称的此枚举类型的常数。
valueOf(String) - 枚举 ai.rapids.cudf.CaptureGroups 中的静态方法
返回具有指定名称的此枚举类型的常数。
valueOf(String) - 枚举 ai.rapids.cudf.ColumnView.FindOptions 中的静态方法
返回具有指定名称的此枚举类型的常数。
valueOf(String) - 枚举 ai.rapids.cudf.CompressionType 中的静态方法
返回具有指定名称的此枚举类型的常数。
valueOf(String) - 枚举 ai.rapids.cudf.CudaComputeMode 中的静态方法
返回具有指定名称的此枚举类型的常数。
valueOf(String) - 枚举 ai.rapids.cudf.CudaException.CudaError 中的静态方法
返回具有指定名称的此枚举类型的常数。
valueOf(String) - 枚举 ai.rapids.cudf.DateTimeComponent 中的静态方法
返回具有指定名称的此枚举类型的常数。
valueOf(String) - 枚举 ai.rapids.cudf.DateTimeRoundingFrequency 中的静态方法
返回具有指定名称的此枚举类型的常数。
valueOf(String) - 枚举 ai.rapids.cudf.DType.DTypeEnum 中的静态方法
返回具有指定名称的此枚举类型的常数。
valueOf(String) - 枚举 ai.rapids.cudf.HashType 中的静态方法
返回具有指定名称的此枚举类型的常数。
valueOf(String) - 枚举 ai.rapids.cudf.NaNEquality 中的静态方法
返回具有指定名称的此枚举类型的常数。
valueOf(String) - 枚举 ai.rapids.cudf.NullEquality 中的静态方法
返回具有指定名称的此枚举类型的常数。
valueOf(String) - 枚举 ai.rapids.cudf.NullPolicy 中的静态方法
返回具有指定名称的此枚举类型的常数。
valueOf(String) - 枚举 ai.rapids.cudf.nvcomp.CompressionType 中的静态方法
返回具有指定名称的此枚举类型的常数。
valueOf(String) - 枚举 ai.rapids.cudf.NvtxColor 中的静态方法
返回具有指定名称的此枚举类型的常数。
valueOf(String) - 枚举 ai.rapids.cudf.OutOfBoundsPolicy 中的静态方法
返回具有指定名称的此枚举类型的常数。
valueOf(String) - 枚举 ai.rapids.cudf.PadSide 中的静态方法
返回具有指定名称的此枚举类型的常数。
valueOf(String) - 枚举 ai.rapids.cudf.ParquetWriterOptions.StatisticsFrequency 中的静态方法
返回具有指定名称的此枚举类型的常数。
valueOf(String) - 枚举 ai.rapids.cudf.QuantileMethod 中的静态方法
返回具有指定名称的此枚举类型的常数。
valueOf(String) - 枚举 ai.rapids.cudf.QuoteStyle 中的静态方法
返回具有指定名称的此枚举类型的常数。
valueOf(String) - 枚举 ai.rapids.cudf.RegexFlag 中的静态方法
返回具有指定名称的此枚举类型的常数。
valueOf(String) - 枚举 ai.rapids.cudf.ReplacePolicy 中的静态方法
返回具有指定名称的此枚举类型的常数。
valueOf(String) - 枚举 ai.rapids.cudf.RoundMode 中的静态方法
返回具有指定名称的此枚举类型的常数。
valueOf(String) - 枚举 ai.rapids.cudf.ScanType 中的静态方法
返回具有指定名称的此枚举类型的常数。
valueOf(String) - 枚举 ai.rapids.cudf.Table.DuplicateKeepOption 中的静态方法
返回具有指定名称的此枚举类型的常数。
valueOf(String) - 枚举 ai.rapids.cudf.TableDebug.Output 中的静态方法
返回具有指定名称的此枚举类型的常数。
valueOf(String) - 枚举 ai.rapids.cudf.UnaryOp 中的静态方法
返回具有指定名称的此枚举类型的常数。
values() - 枚举 ai.rapids.cudf.ast.AstExpression.ExpressionType 中的静态方法
返回一个包含此枚举类型的常数的数组,按照它们声明的顺序。
values() - 枚举 ai.rapids.cudf.ast.BinaryOperator 中的静态方法
返回一个包含此枚举类型的常数的数组,按照它们声明的顺序。
values() - 枚举 ai.rapids.cudf.ast.TableReference 中的静态方法
返回一个包含此枚举类型的常数的数组,按照它们声明的顺序。
values() - 枚举 ai.rapids.cudf.ast.UnaryOperator 中的静态方法
返回一个包含此枚举类型的常数的数组,按照它们声明的顺序。
values() - 枚举 ai.rapids.cudf.BinaryOp 中的静态方法
返回一个包含此枚举类型的常数的数组,按照它们声明的顺序。
values() - 枚举 ai.rapids.cudf.BufferType 中的静态方法
返回一个包含此枚举类型的常数的数组,按照它们声明的顺序。
values() - 枚举 ai.rapids.cudf.CaptureGroups 中的静态方法
返回一个包含此枚举类型的常数的数组,按照它们声明的顺序。
values() - 枚举 ai.rapids.cudf.ColumnView.FindOptions 中的静态方法
返回一个包含此枚举类型的常数的数组,按照它们声明的顺序。
values() - 枚举 ai.rapids.cudf.CompressionType 中的静态方法
返回一个包含此枚举类型的常数的数组,按照它们声明的顺序。
values() - 枚举 ai.rapids.cudf.CudaComputeMode 中的静态方法
返回一个包含此枚举类型的常数的数组,按照它们声明的顺序。
values() - 枚举 ai.rapids.cudf.CudaException.CudaError 中的静态方法
返回一个包含此枚举类型的常数的数组,按照它们声明的顺序。
values() - 枚举 ai.rapids.cudf.DateTimeComponent 中的静态方法
返回一个包含此枚举类型的常数的数组,按照它们声明的顺序。
values() - 枚举 ai.rapids.cudf.DateTimeRoundingFrequency 中的静态方法
返回一个包含此枚举类型的常数的数组,按照它们声明的顺序。
values() - 枚举 ai.rapids.cudf.DType.DTypeEnum 中的静态方法
返回一个包含此枚举类型的常数的数组,按照它们声明的顺序。
values() - 枚举 ai.rapids.cudf.HashType 中的静态方法
返回一个包含此枚举类型的常数的数组,按照它们声明的顺序。
values() - 枚举 ai.rapids.cudf.NaNEquality 中的静态方法
返回一个包含此枚举类型的常数的数组,按照它们声明的顺序。
values() - 枚举 ai.rapids.cudf.NullEquality 中的静态方法
返回一个包含此枚举类型的常数的数组,按照它们声明的顺序。
values() - 枚举 ai.rapids.cudf.NullPolicy 中的静态方法
返回一个包含此枚举类型的常数的数组,按照它们声明的顺序。
values() - 枚举 ai.rapids.cudf.nvcomp.CompressionType 中的静态方法
返回一个包含此枚举类型的常数的数组,按照它们声明的顺序。
values() - 枚举 ai.rapids.cudf.NvtxColor 中的静态方法
返回一个包含此枚举类型的常数的数组,按照它们声明的顺序。
values() - 枚举 ai.rapids.cudf.OutOfBoundsPolicy 中的静态方法
返回一个包含此枚举类型的常数的数组,按照它们声明的顺序。
values() - 枚举 ai.rapids.cudf.PadSide 中的静态方法
返回一个包含此枚举类型的常数的数组,按照它们声明的顺序。
values() - 枚举 ai.rapids.cudf.ParquetWriterOptions.StatisticsFrequency 中的静态方法
返回一个包含此枚举类型的常数的数组,按照它们声明的顺序。
values() - 枚举 ai.rapids.cudf.QuantileMethod 中的静态方法
返回一个包含此枚举类型的常数的数组,按照它们声明的顺序。
values() - 枚举 ai.rapids.cudf.QuoteStyle 中的静态方法
返回一个包含此枚举类型的常数的数组,按照它们声明的顺序。
values() - 枚举 ai.rapids.cudf.RegexFlag 中的静态方法
返回一个包含此枚举类型的常数的数组,按照它们声明的顺序。
values() - 枚举 ai.rapids.cudf.ReplacePolicy 中的静态方法
返回一个包含此枚举类型的常数的数组,按照它们声明的顺序。
values() - 枚举 ai.rapids.cudf.RoundMode 中的静态方法
返回一个包含此枚举类型的常数的数组,按照它们声明的顺序。
values() - 枚举 ai.rapids.cudf.ScanType 中的静态方法
返回一个包含此枚举类型的常数的数组,按照它们声明的顺序。
values() - 枚举 ai.rapids.cudf.Table.DuplicateKeepOption 中的静态方法
返回一个包含此枚举类型的常数的数组,按照它们声明的顺序。
values() - 枚举 ai.rapids.cudf.TableDebug.Output 中的静态方法
返回一个包含此枚举类型的常数的数组,按照它们声明的顺序。
values() - 枚举 ai.rapids.cudf.UnaryOp 中的静态方法
返回一个包含此枚举类型的常数的数组,按照它们声明的顺序。
variance() - 类中的方法 ai.rapids.cudf.ColumnView
返回列中所有值的方差,返回 FLOAT64 标量,除非列类型为 FLOAT32,则返回 FLOAT32 标量。
variance(DType) - 类中的方法 ai.rapids.cudf.ColumnView
返回列中所有值的方差,返回指定类型的标量。
variance() - 类中的静态方法 ai.rapids.cudf.GroupByAggregation
方差聚集,delta degrees of freedom 为 1。
variance(int) - 类中的静态方法 ai.rapids.cudf.GroupByAggregation
方差聚集。
variance() - 类中的静态方法 ai.rapids.cudf.ReductionAggregation
方差聚集,delta degrees of freedom 为 1。
variance(int) - 类中的静态方法 ai.rapids.cudf.ReductionAggregation
方差聚集。
viewHandle - 类中的变量 ai.rapids.cudf.ColumnView
 

W

waitOn(Cuda.Event) - 类中的方法 ai.rapids.cudf.Cuda.Stream
使此流在事件中记录的工作完成之前不执行新工作。
wasDataRead() - 类中的方法 ai.rapids.cudf.JCudfSerialization.SerializedTableHeader
成功从流中读取数据后,此值设置为 true。
wasInitialized() - 类中的方法 ai.rapids.cudf.JCudfSerialization.SerializedTableHeader
如果已读取此表的元数据,则返回 true,否则返回 false,表示遇到了 EOF。
weekDay() - 类中的方法 ai.rapids.cudf.ColumnView
从时间戳中获取星期几。
window(ColumnVector, ColumnVector) - 类中的方法 ai.rapids.cudf.WindowOptions.Builder
设置窗口大小,每行一个条目。
window(Scalar, Scalar) - 类中的方法 ai.rapids.cudf.WindowOptions.Builder
设置范围窗口的大小。
windowOptions - 类中的变量 ai.rapids.cudf.AggregationOverWindow
 
WindowOptions - 类 ai.rapids.cudf.ai.rapids.cudf
滚动窗口的选项。
WindowOptions.Builder - 类 ai.rapids.cudf.ai.rapids.cudf
 
withBinary(String, boolean) - 类中的方法 ai.rapids.cudf.ColumnWriterOptions.NestedBuilder
 
withBinary(String, boolean, int) - 类中的方法 ai.rapids.cudf.ColumnWriterOptions.NestedBuilder
 
withBinaryColumn(String, boolean, int) - 类中的方法 ai.rapids.cudf.ColumnWriterOptions.NestedBuilder
设置二进制子元数据
withBinaryColumn(String, boolean) - 类中的方法 ai.rapids.cudf.ColumnWriterOptions.NestedBuilder
设置二进制子元数据
withCallback(ArrowIPCOptions.NeedGpu) - 类中的方法 ai.rapids.cudf.ArrowIPCOptions.Builder
 
withCallback(ArrowIPCWriterOptions.DoneOnGpu) - 类中的方法 ai.rapids.cudf.ArrowIPCWriterOptions.Builder
 
withColumn(String, boolean) - 类中的方法 ai.rapids.cudf.ColumnWriterOptions.NestedBuilder
 
withColumn(String, boolean, int) - 类中的方法 ai.rapids.cudf.ColumnWriterOptions.NestedBuilder
 
withColumn(boolean, String, int) - 类中的方法 ai.rapids.cudf.ColumnWriterOptions.NestedBuilder
设置简单的子元数据
withColumnNames(String...) - 类中的方法 ai.rapids.cudf.ArrowIPCWriterOptions.Builder
为可空列添加名称。
withColumnNames(List<String>) - 类中的方法 ai.rapids.cudf.CSVWriterOptions.Builder
 
withColumnNames(String...) - 类中的方法 ai.rapids.cudf.CSVWriterOptions.Builder
 
withColumns(boolean, String...) - 类中的方法 ai.rapids.cudf.ColumnWriterOptions.NestedBuilder
设置简单的子元数据
withComment(char) - 类中的方法 ai.rapids.cudf.CSVOptions.Builder
设置标记注释行开头的字符。
withCompressionType(CompressionType) - 类中的方法 ai.rapids.cudf.CompressionMetadataWriterOptions.Builder
设置写入时使用的压缩类型
withCudfPruneSchema(boolean) - 类中的方法 ai.rapids.cudf.JSONOptions.Builder
 
withDayFirst(boolean) - 类中的方法 ai.rapids.cudf.JSONOptions.Builder
是否将日期解析为 DD/MM 而不是 MM/DD
withDecimal(String, int, boolean) - 类中的方法 ai.rapids.cudf.ColumnWriterOptions.NestedBuilder
 
withDecimal(String, int, boolean, int) - 类中的方法 ai.rapids.cudf.ColumnWriterOptions.NestedBuilder
 
withDecimalColumn(String, int, boolean) - 类中的方法 ai.rapids.cudf.ColumnWriterOptions.NestedBuilder
设置 Decimal 子元数据
withDecimalColumn(String, int, boolean, int) - 类中的方法 ai.rapids.cudf.ColumnWriterOptions.NestedBuilder
设置 Decimal 子元数据
withDecimalColumn(String, int) - 类中的方法 ai.rapids.cudf.ColumnWriterOptions.NestedBuilder
设置 Decimal 子元数据
withDelim(char) - 类中的方法 ai.rapids.cudf.CSVOptions.Builder
设置条目分隔符。
withExperimental(boolean) - 类中的方法 ai.rapids.cudf.JSONOptions.Builder
是否启用实验性功能
withFalseValue(String...) - 类中的方法 ai.rapids.cudf.CSVOptions.Builder
 
withFalseValue(String) - 类中的方法 ai.rapids.cudf.CSVWriterOptions.Builder
 
withFieldDelimiter(byte) - 类中的方法 ai.rapids.cudf.CSVWriterOptions.Builder
 
withHeaderAtRow(int) - 类中的方法 ai.rapids.cudf.CSVOptions.Builder
标头数据的行号(从 0 开始计数)。
withIgnoreNullKeys(boolean) - 类中的方法 ai.rapids.cudf.GroupByOptions.Builder
如果为 true,则 cudf 分组将忽略为空的分组键。
withIncludeHeader(Boolean) - 类中的方法 ai.rapids.cudf.CSVWriterOptions.Builder
 
withKeepQuotes(boolean) - 类中的方法 ai.rapids.cudf.JSONOptions.Builder
设置读取器是否应保留字符串值的引号。
withKeysDescending(boolean...) - 类中的方法 ai.rapids.cudf.GroupByOptions.Builder
如果 `keysSorted == true`,指示每列是升序还是降序。
withKeysNullSmallest(boolean...) - 类中的方法 ai.rapids.cudf.GroupByOptions.Builder
如果 `keysSorted == true`,指示每列中 null 值的排序。
withKeysSorted(boolean) - 类中的方法 ai.rapids.cudf.GroupByOptions.Builder
指示 `keys` 中的行是否已排序。
withLeadingZeros(boolean) - 类中的方法 ai.rapids.cudf.JSONOptions.Builder
是否允许数字有前导零。
withLineDelimiter(char) - 类中的方法 ai.rapids.cudf.JSONOptions.Builder
 
withLines(boolean) - 类中的方法 ai.rapids.cudf.JSONOptions.Builder
是否将文件按行作为 json 对象读取
withListColumn(ColumnWriterOptions.ListColumnWriterOptions) - 类中的方法 ai.rapids.cudf.ColumnWriterOptions.NestedBuilder
设置列表列元数据。
withMapColumn(ColumnWriterOptions) - 类中的方法 ai.rapids.cudf.ColumnWriterOptions.NestedBuilder
设置 map 列元数据。
withMaxChunkSize(long) - 类中的方法 ai.rapids.cudf.ArrowIPCWriterOptions.Builder
 
withMetadata(String, String) - 类中的方法 ai.rapids.cudf.CompressionMetadataWriterOptions.Builder
添加元数据键和值
withMetadata(Map<String, String>) - 类中的方法 ai.rapids.cudf.CompressionMetadataWriterOptions.Builder
添加元数据键值对映射
withMixedTypesAsStrings(boolean) - 类中的方法 ai.rapids.cudf.JSONOptions.Builder
指定如何处理包含混合类型的列。
withNonNullableColumns(String...) - 类中的方法 ai.rapids.cudf.ColumnWriterOptions.NestedBuilder
设置列名
withNonNumericNumbers(boolean) - 类中的方法 ai.rapids.cudf.JSONOptions.Builder
是否允许非数字数字。
withNormalizeSingleQuotes(boolean) - 类中的方法 ai.rapids.cudf.JSONOptions.Builder
是否应规范化单引号。
withNormalizeWhitespace(boolean) - 类中的方法 ai.rapids.cudf.JSONOptions.Builder
是否应删除未加引号的空白。
withNotNullableColumnNames(String...) - 类中的方法 ai.rapids.cudf.ArrowIPCWriterOptions.Builder
为不可空列添加名称。
withNullableColumns(String...) - 类中的方法 ai.rapids.cudf.ColumnWriterOptions.NestedBuilder
设置可空列元数据
withNullableDecimalColumn(String, int) - 类中的方法 ai.rapids.cudf.ColumnWriterOptions.NestedBuilder
设置 Decimal 子元数据
withNullableTimestampColumn(String, boolean) - 类中的方法 ai.rapids.cudf.ColumnWriterOptions.NestedBuilder
设置时间戳子元数据
withNullValue(String...) - 类中的方法 ai.rapids.cudf.CSVOptions.Builder
 
withNullValue(String) - 类中的方法 ai.rapids.cudf.CSVWriterOptions.Builder
 
withNumPyTypes(boolean) - 类中的方法 ai.rapids.cudf.ORCOptions.Builder
指定解析器是否应隐含地将 TIMESTAMP_DAYS 列提升为 TIMESTAMP_MILLISECONDS,以实现与 NumPy 的兼容性。
withoutComments() - 类中的方法 ai.rapids.cudf.CSVOptions.Builder
 
withOutput(TableDebug.Output) - 类中的方法 ai.rapids.cudf.TableDebug.Builder
 
withQuote(char) - 类中的方法 ai.rapids.cudf.CSVOptions.Builder
设置引号字符。
withQuoteStyle(QuoteStyle) - 类中的方法 ai.rapids.cudf.CSVOptions.Builder
输入 CSV 数据中预期的引号样式。
withQuoteStyle(QuoteStyle) - 类中的方法 ai.rapids.cudf.CSVWriterOptions.Builder
设置写入 CSV 时使用的引号样式。
withRecoverWithNull(boolean) - 类中的方法 ai.rapids.cudf.JSONOptions.Builder
指定解析 json 时如何处理无效行。
withRowDelimiter(String) - 类中的方法 ai.rapids.cudf.CSVWriterOptions.Builder
 
withRowGroupSizeBytes(long) - 类中的方法 ai.rapids.cudf.ParquetWriterOptions.Builder
 
withRowGroupSizeRows(int) - 类中的方法 ai.rapids.cudf.ParquetWriterOptions.Builder
 
withStatisticsFrequency(ParquetWriterOptions.StatisticsFrequency) - 类中的方法 ai.rapids.cudf.ParquetWriterOptions.Builder
 
withStrictValidation(boolean) - 类中的方法 ai.rapids.cudf.JSONOptions.Builder
json 验证是否严格
withStripeSizeRows(int) - 类中的方法 ai.rapids.cudf.ORCWriterOptions.Builder
 
withStructColumn(ColumnWriterOptions.StructColumnWriterOptions) - 类中的方法 ai.rapids.cudf.ColumnWriterOptions.NestedBuilder
设置子结构体元数据
withTimestamp(String, boolean, boolean) - 类中的方法 ai.rapids.cudf.ColumnWriterOptions.NestedBuilder
 
withTimestamp(String, boolean, boolean, int) - 类中的方法 ai.rapids.cudf.ColumnWriterOptions.NestedBuilder
 
withTimestampColumn(String, boolean, boolean, int) - 类中的方法 ai.rapids.cudf.ColumnWriterOptions.NestedBuilder
设置时间戳子元数据
withTimestampColumn(String, boolean, boolean) - 类中的方法 ai.rapids.cudf.ColumnWriterOptions.NestedBuilder
设置时间戳子元数据
withTimestampColumn(String, boolean) - 类中的方法 ai.rapids.cudf.ColumnWriterOptions.NestedBuilder
设置时间戳子元数据
withTimeUnit(DType) - 类中的方法 ai.rapids.cudf.ORCOptions.Builder
指定返回时间戳时使用的时间单位。
withTimeUnit(DType) - 类中的方法 ai.rapids.cudf.ParquetOptions.Builder
指定返回时间戳时使用的时间单位。
withTrueValue(String...) - 类中的方法 ai.rapids.cudf.CSVOptions.Builder
 
withTrueValue(String) - 类中的方法 ai.rapids.cudf.CSVWriterOptions.Builder
 
withUnquotedControlChars(boolean) - 类中的方法 ai.rapids.cudf.JSONOptions.Builder
字符串中是否允许未加引号的控制字符。
wrap(T[]) - 类中的静态方法 ai.rapids.cudf.CloseableArray
 
wrapped - 类中的变量 ai.rapids.cudf.GroupByAggregationOnColumn
 
wrapped - 类中的变量 ai.rapids.cudf.GroupByScanAggregationOnColumn
 
wrapped - 类中的变量 ai.rapids.cudf.RmmWrappingDeviceMemoryResource
 
wrapped - 类中的变量 ai.rapids.cudf.RollingAggregationOnColumn
 
write(CuFileBuffer, long, long) - 类中的方法 ai.rapids.cudf.CuFileWriteHandle
将指定的 cuFile 缓冲区写入文件。
write(Table) - 类中的方法 ai.rapids.cudf.TableWriter
写出表。
writeArrowIPCChunked(ArrowIPCWriterOptions, File) - 类中的静态方法 ai.rapids.cudf.Table
获取表写入器以将 arrow IPC 数据写入文件。
writeArrowIPCChunked(ArrowIPCWriterOptions, HostBufferConsumer, HostMemoryAllocator) - 类中的静态方法 ai.rapids.cudf.Table
获取表写入器以写入 arrow IPC 数据并使用回调处理每个块。
writeArrowIPCChunked(ArrowIPCWriterOptions, HostBufferConsumer) - 类中的静态方法 ai.rapids.cudf.Table
 
writeColumnViewsToParquet(ParquetWriterOptions, HostBufferConsumer, HostMemoryAllocator, ColumnView...) - 类中的静态方法 ai.rapids.cudf.Table
这是一个正在发展中的 API,很可能在未来版本中移除。
writeColumnViewsToParquet(ParquetWriterOptions, HostBufferConsumer, ColumnView...) - 类中的静态方法 ai.rapids.cudf.Table
 
writeConcatedStream(JCudfSerialization.SerializedTableHeader[], HostMemoryBuffer[], OutputStream) - 类中的静态方法 ai.rapids.cudf.JCudfSerialization
从解析的标头和 dataBuffer 中存储的多个批次获取数据,并将其写入到 out 中,就像它是一个单一缓冲区一样。
writeCSVToFile(CSVWriterOptions, String) - 类中的方法 ai.rapids.cudf.Table
 
writeDeviceBufferToFile(File, long, BaseDeviceMemoryBuffer) - 类中的静态方法 ai.rapids.cudf.CuFile
将设备缓冲区同步写入给定文件路径。
writeDeviceMemoryToFile(File, long, long, long) - 类中的静态方法 ai.rapids.cudf.CuFile
将设备内存同步写入给定文件路径。
writeORCChunked(ORCWriterOptions, File) - 类中的静态方法 ai.rapids.cudf.Table
获取表写入器以将 ORC 数据写入文件。
writeORCChunked(ORCWriterOptions, HostBufferConsumer, HostMemoryAllocator) - 类中的静态方法 ai.rapids.cudf.Table
获取表写入器以写入 ORC 数据并使用回调处理每个块。
writeORCChunked(ORCWriterOptions, HostBufferConsumer) - 类中的静态方法 ai.rapids.cudf.Table
 
writeParquetChunked(ParquetWriterOptions, File) - 类中的静态方法 ai.rapids.cudf.Table
获取表写入器以将 parquet 数据写入文件。
writeParquetChunked(ParquetWriterOptions, HostBufferConsumer, HostMemoryAllocator) - 类中的静态方法 ai.rapids.cudf.Table
获取表写入器以写入 parquet 数据并使用回调处理每个块。
writeParquetChunked(ParquetWriterOptions, HostBufferConsumer) - 类中的静态方法 ai.rapids.cudf.Table
 
writerHandle - 类中的变量 ai.rapids.cudf.TableWriter
 
writeRowsToStream(OutputStream, long) - 类中的静态方法 ai.rapids.cudf.JCudfSerialization
在接收到没有列但行数非零的列式批次时,将仅包含行数的标头写入输出流。
WriteStatistics(long, long, long, double) - 类 ai.rapids.cudf.TableWriter.WriteStatistics 的构造函数
 
writeTo(JCudfSerialization.DataWriter) - 类中的方法 ai.rapids.cudf.JCudfSerialization.SerializedColumnHeader
将此列标头写入指定的写入器
writeTo(JCudfSerialization.DataWriter) - 类中的方法 ai.rapids.cudf.JCudfSerialization.SerializedTableHeader
 
writeToStream(Table, OutputStream, long, long) - Static method in class ai.rapids.cudf.JCudfSerialization
以内部格式写入表的部分或全部内容。
writeToStream(ColumnVector[], OutputStream, long, long) - Static method in class ai.rapids.cudf.JCudfSerialization
以内部格式写入一组列的部分或全部内容。
writeToStream(HostColumnVector[], OutputStream, long, long) - Static method in class ai.rapids.cudf.JCudfSerialization
以内部格式写入一组列的部分或全部内容。

Y

year() - Method in class ai.rapids.cudf.ColumnView
从时间戳获取年份。

Z

zfill(int) - Method in class ai.rapids.cudf.ColumnView
在每个字符串的左侧添加 '0' 进行填充。
A B C D E F G H I J K L M N O P Q R S T U V W Y Z 
跳过导航链接

Copyright © 2025. 保留所有权利。