New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Transformer{DecoderLayer} : no batch dim #70322
Transformer{DecoderLayer} : no batch dim #70322
Conversation
CI Flow Status⚛️ CI FlowRuleset - Version:
You can add a comment to the PR and tag @pytorchbot with the following commands: # ciflow rerun, "ciflow/default" will always be added automatically
@pytorchbot ciflow rerun
# ciflow rerun with additional labels "-l <ciflow/label_name>", which is equivalent to adding these labels manually and trigger the rerun
@pytorchbot ciflow rerun -l ciflow/scheduled -l ciflow/slow For more information, please take a look at the CI Flow Wiki. |
🔗 Helpful links
💊 CI failures summary and remediationsAs of commit 73fc9cb (more details on the Dr. CI page): 💚 💚 Looks good so far! There are no failures yet. 💚 💚 This comment was automatically generated by Dr. CI (expand for details).Please report bugs/suggestions to the (internal) Dr. CI Users group. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I thought of using the same sample_fn for TransformerEncoder and Decoder (but decoder takes a mask as well), ended up keeping them seperate.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM! thanks :)
@jbschlosser has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
Fixes #60585
TransformerDecoder Test Timings (takes about 30s)
Transformer Test Timings (takes about 1m10s)
test/test_modules.py ss......ss......ss..ssssssssss.................. [100%]
==================================================================================
============================================================================================== slowest 10 durations ==============================================================================================
46.40s call test/test_modules.py::TestModuleCUDA::test_gradgrad_nn_Transformer_cuda_float64
11.09s call test/test_modules.py::TestModuleCPU::test_gradgrad_nn_Transformer_cpu_float64
2.48s call test/test_modules.py::TestModuleCUDA::test_grad_nn_Transformer_cuda_float64
1.03s call test/test_modules.py::TestModuleCPU::test_grad_nn_Transformer_cpu_float64
0.96s call test/test_modules.py::TestModuleCUDA::test_cpu_gpu_parity_nn_Transformer_cuda_float32
0.87s call test/test_modules.py::TestModuleCUDA::test_non_contiguous_tensors_nn_Transformer_cuda_float32
0.85s call test/test_modules.py::TestModuleCUDA::test_non_contiguous_tensors_nn_Transformer_cuda_float64
0.85s call test/test_modules.py::TestModuleCPU::test_cpu_gpu_parity_nn_Transformer_cpu_float32
0.65s call test/test_modules.py::TestModuleCUDA::test_cpu_gpu_parity_nn_Transformer_cuda_float64
0.47s call test/test_modules.py::TestModuleCUDA::test_multiple_device_transfer_nn_Transformer_cuda_float32
============================================================================================ short test summary info =============================================================================================
===================================================================== 32 passed, 16 skipped, 591 deselected, 3 warnings in 70.19s (0:01:10) ======================================================================