All pull requests

Pull requests

• [PaddleNLP]Fix llm for LLama-Factory

  #1817·Liujie0926 opened Mar 17, 2025·

  ·

• Udpatename

  #1743·wjm202 opened Oct 12, 2023·

  ·
  2

• Develop

  #1742·wjm202 opened Oct 12, 2023·

  ·
  3

• Fix the proble when test op benmark ,can't use nsight system

  contributor
  #1741·bapijun opened Sep 25, 2023·

  ·
  2

• Update the oneDNN file name for ernie inference.

  contributorstatus: proposed
  #1719·LifengWang opened Jul 21, 2023·

  ·
  2

• delete embedding mapping in op_benchmark.config

  #1700·LUZY0726 opened Jun 16, 2023·

  ·
  4

• [Benchmark] add pointpillars for paddle3D benchmark

  #1670·Stinky-Tofu opened Apr 4, 2023·

  ·
  2

• add N4C32 shell for PETRV2

  #1662·xysheng-baidu opened Mar 29, 2023·

  ·
  2

• Add TSM compile args

  #1635·huangjun12 opened Mar 20, 2023·

  ·
  1

• add PETR repo

  #1634·wangna11BD opened Mar 20, 2023·

  ·
  1

• Add t5_for_conditional_generation and gpt_for_sequence_classification benchmark

  #1581·FrostML opened Feb 10, 2023·

  ·
  6

• Submit add_n categorical test script and json

  #1577·liuruyan opened Feb 1, 2023·

  ·
  4

• nodify_benchmar_ci

  #1568·risemeup1 opened Jan 7, 2023·

  ·
  1

• Reorgnize the configs.

  #1556·Xreki opened Dec 8, 2022·

  ·
  2

• fix eigh backward test case

  #1555·Courtesy-Xs opened Dec 8, 2022·

  ·
  2

• add fleet for TSM

  #1507·huangjun12 opened Aug 30, 2022·

  ·
  3

• max_iter changes for transformer base model benchmark test

  #1484·JamesLim-sy opened Aug 11, 2022·

  ·
  2

• Support to run only 1 time to get gpu_time only for speed task.

  #1417·Xreki opened Mar 23, 2022·

  ·
  1

• Add is_empty op benchmark

  #1347·wjj19950828 opened Mar 4, 2022·

  ·
  2

• Add flops&bytes computation for bilinear_interp script

  #1285·AshburnLee opened Feb 8, 2022·

  ·
  2

• 【hold 】Test gan 0105

  #1254·hysunflower opened Jan 12, 2022·

  ·
  2

• [DO NOT MERGE] add scheduling evaluation in paddle dygraph mode

  #1240·pangyoki opened Jan 5, 2022·

  ·
  3

• modify is_ampere_gpu in api/common/launch.py

  #1238·Courtesy-Xs opened Jan 5, 2022·

  ·
  3

• Enable to get the precise gpu_time for tf by using ctypes package to load libcudart.so and add add nchw support for batch_norm.

  #1237·Xreki opened Jan 4, 2022·

  ·
  2

• adjust iters in model_benchmark_ci

  #1207·xiegegege opened Dec 23, 2021·

  ·
  1