ryujaehun/pytorch-gpu-benchmark issues and pull requests

#31 - Added 7900XTX results with lamikr's support for AMD cards

Pull Request - State: open - Opened by eitch 3 months ago

#30 - plot.ipynb does not work for parsing new results with pytorch 2.30

Issue - State: open - Opened by lamikr 4 months ago

#29 - Test.sh needs changes to allow running tests also for AMD gpus

Issue - State: open - Opened by lamikr 4 months ago

#28 - AMD support to test.sh and AMD RX 6800 results

Pull Request - State: open - Opened by lamikr 4 months ago - 1 comment

#27 - When running benchmark_models.py, show error messages"No module named 'pandas'"

Issue - State: open - Opened by Oracle-Chen 5 months ago

#26 - Added RTX 5000 Ada Laptop performances

Pull Request - State: open - Opened by dot-Eagle96 5 months ago

#25 - Add 6800XT results

Pull Request - State: open - Opened by ToughStyle 7 months ago

#24 - Added 4060 Ti 16GB results

Pull Request - State: closed - Opened by adorobis 7 months ago

#23 - why did this benchmark blow away my linux partition?

Issue - State: closed - Opened by achillez about 1 year ago - 1 comment

#22 - bugfix , support for torch 2.0

Pull Request - State: closed - Opened by hellojixian over 1 year ago - 1 comment

#21 - TypeError: call() got an unexpected keyword argument 'pretrained'

Issue - State: open - Opened by joshhu about 2 years ago - 4 comments

#20 - adding rtx 3060 results

Pull Request - State: closed - Opened by zachcoleman over 2 years ago

#19 - RuntimeError: miopenStatusUnknownError

Issue - State: open - Opened by Bengt almost 3 years ago

#18 - error in type mobilenet_v2

Issue - State: closed - Opened by jjziets over 3 years ago - 5 comments

#17 - GTX 2080 Ti Performance [Windows 10]

Issue - State: closed - Opened by arstropica over 3 years ago - 1 comment

#16 - cannot support ROCM

Issue - State: closed - Opened by znsoftm over 3 years ago - 3 comments

#15 - titan RTX

Issue - State: closed - Opened by ryujaehun over 3 years ago

#14 - results for nvidia A100

Pull Request - State: closed - Opened by kirk86 almost 4 years ago - 2 comments

#13 - poly.ipynb ValueError

Issue - State: closed - Opened by sevaroy almost 4 years ago - 1 comment

#12 - Incorrect help information for arguments: folder

Pull Request - State: closed - Opened by yqtianust about 4 years ago - 3 comments

#11 - test rtx2080ti on windows10

Pull Request - State: closed - Opened by olixu about 4 years ago

#10 - resnet with batchsize=12?

Issue - State: closed - Opened by twmht about 4 years ago - 1 comment

#9 - The performance of cuda on windows10 VS Linux

Issue - State: closed - Opened by olixu about 4 years ago - 7 comments

#8 - If the batch size becomes smaller, it would occurs because the other part is overloaded rather than the operation.

Issue - State: closed - Opened by olixu about 4 years ago

#7 - have you any interest to try larger batch size, such as 64 or 128?

Issue - State: closed - Opened by qixiang109 over 4 years ago - 1 comment

#6 - Have you tried batch_size 1 for fp32 and fp16?

Issue - State: closed - Opened by Edwardmark over 4 years ago - 1 comment

#5 - RuntimeError: CUDA out of memory. Tried to allocate 32.00 MiB (GPU 0; 7.79 GiB total capacity; 6.64 GiB already allocated; 21.00 MiB free; 150.86 MiB cached) .

Issue - State: closed - Opened by M-Abdallah almost 5 years ago - 4 comments

#4 - Fix bar plot layout, add a new figure

Pull Request - State: closed - Opened by elombardi2 about 5 years ago

#3 - My 1080 GTX Ti is 50% slower than your benchmark

Issue - State: closed - Opened by NikEyX over 5 years ago - 5 comments

#2 - Update typo

Pull Request - State: closed - Opened by johmathe over 5 years ago

#1 - Summary BenchMark

Issue - State: closed - Opened by ryujaehun almost 6 years ago

Ecosyste.ms: Issues

GitHub / ryujaehun/pytorch-gpu-benchmark issues and pull requests