rapidsai/cudf issues and pull requests

#17312 - Add version config

Pull Request - State: open - Opened by vyasr 3 days ago
Labels: feature request, libcudf, CMake, non-breaking

#17311 - [BUG] Chunked parquet reader incorrect results for positive values of `n_rows`

Issue - State: open - Opened by brandon-b-miller 3 days ago
Labels: bug, cuIO

#17310 - [WIP] Migrate ORC Writer to pylibcudf

Pull Request - State: open - Opened by Matt711 4 days ago
Labels: feature request, Python, non-breaking, pylibcudf

#17309 - Clean up misc, unneeded pylibcudf.libcudf in cudf._lib

Pull Request - State: open - Opened by mroeschke 4 days ago
Labels: Python, improvement, non-breaking, pylibcudf

#17308 - Exclude nanoarrow and flatbuffers from installation

Pull Request - State: closed - Opened by vyasr 4 days ago - 4 comments
Labels: libcudf, CMake, improvement, non-breaking

#17307 - Use the new logger generator in rmm

Pull Request - State: open - Opened by vyasr 4 days ago
Labels: libcudf, CMake, improvement, non-breaking

#17306 - Update CI jobs to include Polars in nightlies and improve IWYU

Pull Request - State: open - Opened by vyasr 4 days ago - 1 comment
Labels: libcudf, CMake, improvement, non-breaking

#17305 - Fix reading of single-row unterminated CSV files

Pull Request - State: open - Opened by vuule 4 days ago
Labels: bug, libcudf, Python, non-breaking

#17304 - Move strings repeat benchmarks to nvbench

Pull Request - State: open - Opened by davidwendt 4 days ago
Labels: 2 - In Progress, libcudf, CMake, improvement, non-breaking

#17303 - [BUG] GDS error when LIBCUDF_CUFILE_POLICY is ALWAYS

Issue - State: open - Opened by kingcrimsontianyu 4 days ago - 2 comments
Labels: bug

#17302 - Fix synchronization bug in bool parquet mukernels

Pull Request - State: closed - Opened by pmattione-nvidia 4 days ago - 1 comment
Labels: libcudf, ! - Hotfix, improvement, non-breaking

#17301 - Move strings replace benchmarks to nvbench

Pull Request - State: open - Opened by davidwendt 4 days ago
Labels: 3 - Ready for Review, libcudf, CMake, improvement, non-breaking

#17300 - [FEA] Some level of automation in pylibcudf type stub generation

Issue - State: open - Opened by wence- 4 days ago
Labels: feature request

#17299 - Support polars 1.13

Pull Request - State: closed - Opened by wence- 4 days ago - 1 comment
Labels: Python, improvement, non-breaking, cudf.polars

#17298 - Replace FindcuFile with upstream FindCUDAToolkit support

Pull Request - State: closed - Opened by KyleFromNVIDIA 4 days ago - 1 comment
Labels: libcudf, CMake, improvement, non-breaking

#17297 - [DO NOT MERGE] chunked reader + kvikio fixes

Pull Request - State: open - Opened by brandon-b-miller 4 days ago
Labels: libcudf, Python, 5 - DO NOT MERGE, cudf.polars

#17296 - [FEA] Error message that explains why installing failed

Issue - State: open - Opened by betatim 4 days ago - 5 comments
Labels: feature request

#17295 - transpose.hpp

Issue - State: closed - Opened by shrshi 4 days ago - 2 comments

#17294 - Expose stream-ordering in public transpose API

Pull Request - State: closed - Opened by shrshi 5 days ago - 1 comment
Labels: libcudf, CMake, improvement, non-breaking

#17293 - Replace workaround of JNI build with CUDF_KVIKIO_REMOTE_IO=OFF

Pull Request - State: closed - Opened by pxLi 5 days ago - 3 comments
Labels: Java, Spark, improvement, non-breaking

#17292 - Fixed lifetime issue in ast transform tests

Pull Request - State: closed - Opened by lamarrr 5 days ago - 3 comments
Labels: bug, libcudf, non-breaking

#17291 - cmake option: `CUDF_KVIKIO_REMOTE_IO`

Pull Request - State: closed - Opened by madsbk 7 days ago - 5 comments
Labels: libcudf, CMake, improvement, non-breaking

#17290 - Occupancy improvement for distinct hash join with specialized dispatch

Pull Request - State: open - Opened by tgujar 8 days ago - 2 comments
Labels: 3 - Ready for Review, libcudf, Performance, improvement, breaking

#17289 - 🚧 Support for reading bloom filters from Parquet files

Pull Request - State: open - Opened by mhaseeb123 8 days ago
Labels: feature request, 2 - In Progress, libcudf, cuIO, non-breaking, cuco

#17288 - Use more pylibcudf Python enums in cudf._lib

Pull Request - State: closed - Opened by mroeschke 8 days ago - 1 comment
Labels: Python, improvement, non-breaking

#17287 - Use pylibcudf enums in cudf Python quantile

Pull Request - State: closed - Opened by mroeschke 8 days ago - 1 comment
Labels: Python, improvement, non-breaking

#17284 - enforce wheel size limits, README formatting in CI

Pull Request - State: closed - Opened by jameslamb 8 days ago - 1 comment
Labels: Python, improvement, non-breaking, cudf.polars, pylibcudf

#17281 - Java JNI for Multiple contains

Pull Request - State: open - Opened by res-life 8 days ago - 2 comments
Labels: feature request, Java, non-breaking

#17280 - Use numba-cuda<0.0.18

Pull Request - State: closed - Opened by gmarkall 8 days ago - 1 comment
Labels: Python, improvement, non-breaking

#17279 - Add compute_column_expression to pylibcudf for transform.compute_column

Pull Request - State: open - Opened by mroeschke 9 days ago - 1 comment
Labels: Python, improvement, non-breaking, pylibcudf

#17278 - Optimize distinct inner join to use set `find` instead of `retrieve`

Pull Request - State: open - Opened by PointKernel 9 days ago - 3 comments
Labels: 3 - Ready for Review, libcudf, Performance, improvement, non-breaking

#17274 - [BUG] AST_TEST TransformTest.DeeplyNestedArithmeticLogicalExpression fails when run in a debug build

Issue - State: closed - Opened by davidwendt 9 days ago - 4 comments
Labels: bug, libcudf

#17269 - Move strings filter benchmarks to nvbench

Pull Request - State: open - Opened by davidwendt 9 days ago
Labels: 3 - Ready for Review, libcudf, CMake, improvement, non-breaking

#17268 - Raise errors on specific types of fallback in `cudf.pandas`

Pull Request - State: closed - Opened by Matt711 9 days ago - 1 comment
Labels: feature request, Python, non-breaking, cudf.pandas

#17267 - Add `catboost` to the third-party integration tests

Pull Request - State: closed - Opened by Matt711 9 days ago - 4 comments
Labels: feature request, Python, non-breaking, cudf.pandas

#17266 - Expose delimiter character in JSON reader options to JSON reader APIs

Pull Request - State: closed - Opened by shrshi 9 days ago - 1 comment
Labels: bug, libcudf, 5 - Ready to Merge, cuIO, non-breaking

#17263 - Add write_parquet to pylibcudf

Pull Request - State: open - Opened by mroeschke 10 days ago
Labels: Python, improvement, non-breaking, pylibcudf

#17262 - Single-partition Dask executor for cuDF-Polars

Pull Request - State: open - Opened by rjzamora 10 days ago - 2 comments
Labels: 2 - In Progress, Python, improvement, non-breaking, cudf.polars

#17261 - [BUG] `cudf::io::json::detail::normalize_single_quotes` outputs incorrect result when the input has `\n` character

Issue - State: closed - Opened by ttnghia 10 days ago
Labels: bug, cuIO

#17260 - Always prefer `device_read`s and `device_write`s when kvikIO is enabled

Pull Request - State: closed - Opened by vuule 10 days ago - 2 comments
Labels: libcudf, 5 - Ready to Merge, cuIO, Performance, improvement, non-breaking

#17259 - [FEA] Adjust libcudf to use kvikIO for small host reads

Issue - State: closed - Opened by GregoryKimball 10 days ago - 4 comments
Labels: feature request, libcudf, cuIO

#17258 - Add type stubs for pylibcudf

Pull Request - State: closed - Opened by wence- 10 days ago - 3 comments
Labels: feature request, Python, CMake, non-breaking, cudf.polars, pylibcudf

#17257 - Expose streams in public quantile APIs

Pull Request - State: closed - Opened by shrshi 10 days ago - 1 comment
Labels: libcudf, CMake, improvement, non-breaking

#17255 - Move strings/numeric convert benchmarks to nvbench

Pull Request - State: open - Opened by davidwendt 10 days ago
Labels: 3 - Ready for Review, libcudf, CMake, improvement, non-breaking

#17254 - Fix extract-datetime deprecation warning in ndsh benchmark

Pull Request - State: open - Opened by davidwendt 10 days ago
Labels: bug, 3 - Ready for Review, libcudf, non-breaking

#17253 - Disallow cuda-python 12.6.1 and 11.8.4

Pull Request - State: open - Opened by bdice 10 days ago - 1 comment
Labels: bug, Python, non-breaking, pylibcudf

#17252 - Add write_parquet to pylibcudf

Pull Request - State: open - Opened by mroeschke 11 days ago
Labels: Python, improvement, non-breaking, pylibcudf

#17251 - Wrap custom iterator result

Pull Request - State: open - Opened by galipremsagar 11 days ago - 2 comments
Labels: bug, Python, non-breaking, cudf.pandas

#17250 - Add new ``dask_cudf.read_parquet`` API

Pull Request - State: open - Opened by rjzamora 11 days ago
Labels: 2 - In Progress, Python, improvement, non-breaking

#17249 - POC: Implement `HOST_UDF` aggregations

Pull Request - State: open - Opened by ttnghia 11 days ago - 1 comment
Labels: feature request, 2 - In Progress, libcudf, CMake, Spark

#17248 - Add breaking change workflow trigger

Pull Request - State: open - Opened by AyodeAwe 11 days ago - 2 comments
Labels: improvement, non-breaking

#17247 - [FEA] Compare performance of decompression engine (HW) versus decompression kernels (SW) on Blackwell

Issue - State: open - Opened by GregoryKimball 11 days ago
Labels: feature request, libcudf, cuIO

#17246 - Use pylibcudf contiguous split APIs in cudf python

Pull Request - State: open - Opened by Matt711 11 days ago
Labels: feature request, Python, non-breaking, pylibcudf

#17245 - Add read_parquet_metadata to pylibcudf

Pull Request - State: open - Opened by mroeschke 12 days ago
Labels: Python, CMake, improvement, non-breaking, pylibcudf

#17244 - [FEA] Support `strict=False` casting in `cudf-polars`

Issue - State: open - Opened by brandon-b-miller 12 days ago
Labels: feature request, cudf.polars

#17243 - Search for kvikio with lowercase

Pull Request - State: closed - Opened by vyasr 12 days ago
Labels: libcudf, CMake, improvement, non-breaking

#17242 - Fix the example in documentation for `get_dremel_data()`

Pull Request - State: open - Opened by mhaseeb123 12 days ago
Labels: doc, libcudf, 4 - Needs Review, non-breaking

#17241 - Support Hash-based group by aggregations for min/max with nesting

Issue - State: open - Opened by revans2 12 days ago
Labels: feature request, Performance, Spark

#17240 - [BUG] JSON parser still returns outdated schema structure for strings column

Issue - State: open - Opened by ttnghia 12 days ago - 3 comments
Labels: bug, cuIO

#17239 - KvikIO shared library

Pull Request - State: closed - Opened by madsbk 12 days ago - 2 comments
Labels: Python, improvement, non-breaking

#17238 - [FEA] read_csv optimizations with streaming multiprocessors and MGPU

Issue - State: open - Opened by Hamed-AlMehdhar 14 days ago - 1 comment
Labels: feature request, pylibcudf

#17237 - Use more pylibcudf.io.types enums in cudf._libs

Pull Request - State: closed - Opened by mroeschke 15 days ago - 1 comment
Labels: Python, improvement, non-breaking, pylibcudf

#17236 - [FEA] Add function for "deduplicate map" to libcudf

Issue - State: open - Opened by GregoryKimball 15 days ago
Labels: feature request, libcudf

#17235 - Expose mixed and conditional joins in pylibcudf

Pull Request - State: closed - Opened by wence- 15 days ago - 1 comment
Labels: libcudf, Python, improvement, non-breaking, pylibcudf

#17234 - Precompute AST arity

Pull Request - State: open - Opened by bdice 15 days ago - 4 comments
Labels: libcudf, CMake

#17233 - Update to CCCL 2.7.0-rc2.

Pull Request - State: open - Opened by bdice 15 days ago
Labels: libcudf, CMake, improvement, non-breaking

#17232 - Add io.text APIs to pylibcudf

Pull Request - State: open - Opened by mroeschke 16 days ago
Labels: Python, CMake, improvement, non-breaking, pylibcudf

#17231 - Add `num_iterations` axis to the multi-threaded Parquet benchmarks

Pull Request - State: closed - Opened by vuule 16 days ago - 2 comments
Labels: tests, libcudf, cuIO, improvement, non-breaking

#17230 - [BUG] `cudf::io::json::detail::get_token_stream` does not respect `normalize_single_quotes` option

Issue - State: closed - Opened by ttnghia 16 days ago - 1 comment
Labels: bug

#17229 - Move strings to date/time types benchmarks to nvbench

Pull Request - State: open - Opened by davidwendt 16 days ago
Labels: 3 - Ready for Review, libcudf, CMake, improvement, non-breaking

#17228 - [FEA] Use kvikIO for `host_read`s and `host_write`s

Issue - State: open - Opened by vuule 16 days ago
Labels: feature request, cuIO, Performance, improvement

#17227 - Support for polars 1.12 in cudf-polars

Pull Request - State: closed - Opened by wence- 16 days ago - 4 comments
Labels: Python, improvement, non-breaking, cudf.polars

#17226 - Fix binop with LHS numpy datetimelike scalar

Pull Request - State: closed - Opened by mroeschke 16 days ago - 1 comment
Labels: bug, Python, non-breaking

#17224 - Allow generating large strings in benchmarks

Pull Request - State: open - Opened by davidwendt 16 days ago
Labels: 3 - Ready for Review, libcudf, improvement, non-breaking

#17223 - Refactor gather/scatter benchmarks for strings

Pull Request - State: open - Opened by davidwendt 16 days ago
Labels: 3 - Ready for Review, libcudf, CMake, strings, improvement, non-breaking

#17222 - Fix `Dataframe.setitem` slow-downs

Pull Request - State: closed - Opened by galipremsagar 17 days ago - 3 comments
Labels: bug, Python, non-breaking, cudf.pandas

#17221 - Deprecate single component extraction methods in libcudf

Pull Request - State: closed - Opened by Matt711 17 days ago - 4 comments
Labels: libcudf, improvement, breaking

#17219 - Benchmarking JSON reader for compressed inputs

Pull Request - State: open - Opened by shrshi 17 days ago
Labels: libcudf, CMake, Performance, improvement, non-breaking

#17216 - Fix groupby.get_group with length-1 tuple with list-like grouper

Pull Request - State: closed - Opened by mroeschke 17 days ago - 1 comment
Labels: bug, Python, non-breaking

#17215 - Fix discoverability of submodules inside `pd.util`

Pull Request - State: closed - Opened by galipremsagar 17 days ago - 4 comments
Labels: bug, Python, 5 - Ready to Merge, non-breaking, cudf.pandas

#17207 - Mark column chunks in a PQ reader `pass` as large strings when the cumulative `offsets` exceeds the large strings threshold.

Pull Request - State: open - Opened by mhaseeb123 18 days ago - 1 comment
Labels: bug, libcudf, 5 - Ready to Merge, non-breaking

GitHub / rapidsai/cudf issues and pull requests