Ecosyste.ms: Issues
An open API service for providing issue and pull request metadata for open source projects.
GitHub / paddlepaddle/benchmark issues and pull requests
#1594 - Add einsum.
Pull Request -
State: closed - Opened by Xreki almost 2 years ago
- 1 comment
#1588 - disable unsupported activation and maths test
Pull Request -
State: closed - Opened by shaojiewang almost 2 years ago
- 1 comment
#1581 - Add t5_for_conditional_generation and gpt_for_sequence_classification benchmark
Pull Request -
State: open - Opened by FrostML almost 2 years ago
- 2 comments
#1556 - Reorgnize the configs.
Pull Request -
State: open - Opened by Xreki about 2 years ago
- 2 comments
#1555 - fix eigh backward test case
Pull Request -
State: open - Opened by Courtesy-Xs about 2 years ago
- 2 comments
#1507 - add fleet for TSM
Pull Request -
State: open - Opened by huangjun12 over 2 years ago
- 3 comments
#1484 - max_iter changes for transformer base model benchmark test
Pull Request -
State: open - Opened by JamesLim-sy over 2 years ago
- 2 comments
#1417 - Support to run only 1 time to get gpu_time only for speed task.
Pull Request -
State: open - Opened by Xreki almost 3 years ago
- 1 comment
#1347 - Add is_empty op benchmark
Pull Request -
State: open - Opened by wjj19950828 almost 3 years ago
- 2 comments
#1285 - Add flops&bytes computation for bilinear_interp script
Pull Request -
State: open - Opened by AshburnLee almost 3 years ago
- 2 comments
#1270 - [GPT] Using CUDA112 for GPT.
Pull Request -
State: closed - Opened by ZHUI about 3 years ago
- 2 comments
#1254 - 【hold 】Test gan 0105
Pull Request -
State: open - Opened by hysunflower about 3 years ago
- 2 comments
#1240 - [DO NOT MERGE] add scheduling evaluation in paddle dygraph mode
Pull Request -
State: open - Opened by pangyoki about 3 years ago
- 3 comments
#1237 - 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.
Pull Request -
State: open - Opened by Xreki about 3 years ago
- 2 comments
#1207 - adjust iters in model_benchmark_ci
Pull Request -
State: open - Opened by xiegegege about 3 years ago
- 1 comment
#1143 - add topk input shape of mmdnn
Pull Request -
State: open - Opened by zkh2016 about 3 years ago
- 2 comments
#1085 - add time caculate
Pull Request -
State: open - Opened by Gzxl over 3 years ago
- 3 comments
#1084 - Add frame and overlap_add op benchmark cases.
Pull Request -
State: open - Opened by KPatr1ck over 3 years ago
- 3 comments
#1058 - support tensorflow batch_norm NCHW format
Pull Request -
State: open - Opened by Zjq9409 over 3 years ago
- 2 comments
#1056 - add unique_consecutive op
Pull Request -
State: open - Opened by firestonelib over 3 years ago
- 2 comments
#1049 - Change absolute-error to relative-error when check outputs
Pull Request -
State: open - Opened by AshburnLee over 3 years ago
- 2 comments
#1011 - 【dont merge】update_senta_models, test=document_fix
Pull Request -
State: open - Opened by hysunflower over 3 years ago
- 2 comments
#1010 - 【dont merge】update_seq2seq_models, test=document_fix
Pull Request -
State: open - Opened by hysunflower over 3 years ago
- 2 comments
#992 - Add op benchmark for dot
Pull Request -
State: open - Opened by tongxin over 3 years ago
- 2 comments
#948 - Add lstm.py and rnn.py in dynamic_tests_v2
Pull Request -
State: open - Opened by AshburnLee almost 4 years ago
- 1 comment
#947 - add config for temporal_shift
Pull Request -
State: open - Opened by zhangting2020 almost 4 years ago
- 1 comment
#830 - update layer_norm for params weight and bias not none
Pull Request -
State: open - Opened by windstamp about 4 years ago
- 2 comments
#793 - Support getting the pure backward time for benchmark on dynamic
Pull Request -
State: open - Opened by Xreki about 4 years ago
- 3 comments
#538 - Add README.md for the OP benchmark system, representing how to use 'tests/run.sh'
Pull Request -
State: open - Opened by AshburnLee over 4 years ago
- 3 comments
#190 - add paadle and mxnet run.sh
Pull Request -
State: open - Opened by gentelyang over 5 years ago
- 1 comment
#100 - Add Megatron-LM submodule
Pull Request -
State: closed - Opened by chengduoZH over 5 years ago
#99 - Enable the use of py_reader in PaddingRNN
Pull Request -
State: closed - Opened by Xreki over 5 years ago
#98 - add dygraph benchmark
Pull Request -
State: closed - Opened by DDDivano over 5 years ago
#97 - update for the val dataset for train(using shorter time to profile)
Pull Request -
State: closed - Opened by wangchaochaohu over 5 years ago
#96 - change multiprocess logs path
Pull Request -
State: closed - Opened by ccmeteorljh over 5 years ago
#95 - mask-rcnn 有时会出现loss 为nan 的问题
Issue -
State: open - Opened by ccmeteorljh over 5 years ago
#94 - Add drop local scope iterations
Pull Request -
State: closed - Opened by chengduoZH over 5 years ago
#93 - Disbale gc flags for PaddingRNN.
Pull Request -
State: closed - Opened by Xreki over 5 years ago
- 1 comment
#92 - 运行maskrcnn-from-fb有报错
Issue -
State: closed - Opened by xiegegege over 5 years ago
- 1 comment
#91 - Polish Mask-RCNN train.py
Pull Request -
State: closed - Opened by chengduoZH over 5 years ago
#90 - Revert "Revert multi process in mask rcnn"
Pull Request -
State: closed - Opened by chengduoZH over 5 years ago
#89 - Revert multi process in mask rcnn
Pull Request -
State: closed - Opened by chengduoZH over 5 years ago
#88 - Update mask_rcnn
Pull Request -
State: closed - Opened by chengduoZH over 5 years ago
#87 - Fix bug of mask_rcnn
Pull Request -
State: closed - Opened by chengduoZH over 5 years ago
#86 - add multi_process cases
Pull Request -
State: closed - Opened by ccmeteorljh over 5 years ago
#85 - modify run.sh to support multi_process for transformer, maskrcnn, yol…
Pull Request -
State: closed - Opened by ccmeteorljh over 5 years ago
#84 - Which is used for BERT training benchmark
Issue -
State: open - Opened by LeoZhao-Intel over 5 years ago
- 10 comments
#83 - Mask rcnn跑多卡多进程时,只有一个卡在跑,其它卡未启动。
Issue -
State: closed - Opened by gentelyang over 5 years ago
#82 - Mask rcnn跑多卡多进程时,只有一个卡在跑,其它卡未启动。
Issue -
State: closed - Opened by gentelyang over 5 years ago
#81 - transformer 多进程单卡下报错
Issue -
State: open - Opened by ccmeteorljh over 5 years ago
- 4 comments
#80 - Fix run_with_multi_process of yolov3
Pull Request -
State: closed - Opened by chengduoZH over 5 years ago
#79 - Refine the pe options for padding-rnn and mask-rcnn.
Pull Request -
State: closed - Opened by Xreki over 5 years ago
#78 - se-resnext 8个进程下hang住,4进程下没问题
Issue -
State: open - Opened by ccmeteorljh over 5 years ago
#77 - Fix bug of PaddingRnn/lm_model.py
Pull Request -
State: closed - Opened by chengduoZH over 5 years ago
#76 - Add multi process implementation for DeepLab_V3
Pull Request -
State: closed - Opened by chengduoZH over 5 years ago
#75 - [multi process] Add multi process implementation for yolov3
Pull Request -
State: closed - Opened by chengduoZH over 5 years ago
#74 - [multi process] Add multi process implementation For Bert
Pull Request -
State: closed - Opened by chengduoZH over 5 years ago
#73 - [multi process] Update se_resnext
Pull Request -
State: closed - Opened by chengduoZH over 5 years ago
#72 - [multi-process] Use multi-process run Transformer
Pull Request -
State: closed - Opened by chengduoZH over 5 years ago
#71 - [multi-process] Update learning rate of se-resnext
Pull Request -
State: closed - Opened by chengduoZH over 5 years ago
#70 - modify paddingRnn run.sh
Pull Request -
State: closed - Opened by ccmeteorljh over 5 years ago
#69 - Optimize PaddingRNN using fused_elementwise_activation op.
Pull Request -
State: closed - Opened by Xreki over 5 years ago
#68 - Update se_resnext
Pull Request -
State: closed - Opened by chengduoZH over 5 years ago
#67 - Set learning_rate for mask-rcnn.
Pull Request -
State: closed - Opened by Xreki over 5 years ago
#66 - 基于CUDA10在配置竞品环境时碰到的问题
Issue -
State: open - Opened by gentelyang over 5 years ago
#65 - Enable the use of tensor-core in cublas for V100.
Pull Request -
State: closed - Opened by Xreki over 5 years ago
#64 - add max_bs for run.sh
Pull Request -
State: closed - Opened by ccmeteorljh over 5 years ago
#63 - auto_run_paddle.sh执行的时候,paddingRnn总是会会在large model出现错误
Issue -
State: closed - Opened by ccmeteorljh over 5 years ago
- 1 comment
#62 - add dyanmic_rnn(like staticRNN op); test=develop
Pull Request -
State: closed - Opened by phlrain over 5 years ago
#61 - deeplabv3+ 有时出现异常结束的问题,导致run.sh 退出无法获取结果
Issue -
State: open - Opened by ccmeteorljh over 5 years ago
#60 - Unite the log path
Pull Request -
State: closed - Opened by ccmeteorljh over 5 years ago
#59 - Assign last_cell/hidden to init_cell/hidden in program.
Pull Request -
State: closed - Opened by Xreki over 5 years ago
#58 - Refine padding rnn mem
Pull Request -
State: closed - Opened by sneaxiy over 5 years ago
#57 - modify auto run
Pull Request -
State: closed - Opened by ccmeteorljh over 5 years ago
#56 - add docker_run.sh
Pull Request -
State: closed - Opened by ccmeteorljh over 5 years ago
#55 - Use multi-process to train se_resnext
Pull Request -
State: closed - Opened by chengduoZH over 5 years ago
- 2 comments
#54 - add gc for deeplabv3+
Pull Request -
State: closed - Opened by ccmeteorljh over 5 years ago
#53 - refine logs name
Pull Request -
State: closed - Opened by ccmeteorljh over 5 years ago
#52 - [multi process]Use multi process run mask rcnn
Pull Request -
State: closed - Opened by chengduoZH over 5 years ago
#51 - Modify padding rnn model
Pull Request -
State: closed - Opened by sneaxiy over 5 years ago
#50 - Modify the way BERT and Transformer speed statistics.
Pull Request -
State: closed - Opened by chengduoZH over 5 years ago
#49 - modify run.sh of mask-rcnn
Pull Request -
State: closed - Opened by ccmeteorljh almost 6 years ago
#48 - modify se-resnext FLAGS_fraction_of_gpu_memory_to_use
Pull Request -
State: closed - Opened by ccmeteorljh almost 6 years ago
#47 - modify auto_run_paddle.sh
Pull Request -
State: closed - Opened by ccmeteorljh almost 6 years ago
#46 - modify run.sh of yolov3
Pull Request -
State: closed - Opened by ccmeteorljh almost 6 years ago
#45 - Add LARK_Paddle_BERT2
Pull Request -
State: closed - Opened by chengduoZH almost 6 years ago
#44 - Add LARK_Paddle_BERT
Pull Request -
State: closed - Opened by chengduoZH almost 6 years ago
#43 - open gc in run.sh of cyclegan, mask-rcnn, se-resnext
Pull Request -
State: closed - Opened by ccmeteorljh almost 6 years ago
#42 - modify run.sh of Mask-RCNN, PaddingRNN, CycleGan
Pull Request -
State: closed - Opened by ccmeteorljh almost 6 years ago
#41 - modify PaddingRNN scripts
Pull Request -
State: closed - Opened by ccmeteorljh almost 6 years ago
#40 - Mask-rcnn paddle需要加入一个环境变量才可以最大bs=5的问题
Issue -
State: closed - Opened by ccmeteorljh almost 6 years ago
- 3 comments
#39 - pr #31 会导致transformer出现loss异常现象
Issue -
State: closed - Opened by ccmeteorljh almost 6 years ago
- 1 comment
#38 - paddingRnn 出现了loss异常
Issue -
State: closed - Opened by ccmeteorljh almost 6 years ago
#37 - unified models unit
Pull Request -
State: closed - Opened by ccmeteorljh almost 6 years ago
#36 - YOLOv3
Pull Request -
State: closed - Opened by wangchaochaohu almost 6 years ago
#35 - Update se_resnext_50
Pull Request -
State: closed - Opened by chengduoZH almost 6 years ago
#34 - remove yolov3 paddle script, add link and commit info
Pull Request -
State: closed - Opened by liupluswei almost 6 years ago
#33 - add benchmark scripts for yolov3: MxNet vs Padddle
Pull Request -
State: closed - Opened by liupluswei almost 6 years ago
#32 - Set padding rnn thread number
Pull Request -
State: closed - Opened by sneaxiy almost 6 years ago
#31 - Update transformer model
Pull Request -
State: closed - Opened by chengduoZH almost 6 years ago