Skip to content
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

Rebase Approval for Branches #73

Open
Cemberk opened this issue Nov 18, 2024 · 8 comments
Open

Rebase Approval for Branches #73

Cemberk opened this issue Nov 18, 2024 · 8 comments

Comments

@Cemberk
Copy link
Owner

Cemberk commented Nov 18, 2024

This issue contains the test results for the upstream sync, develop PR, and release testing branches. Comment 'proceed with rebase' to approve. Close when maintenance is complete or there will be problems on the next run.

@Cemberk
Copy link
Owner Author

Cemberk commented Nov 18, 2024

Git diff artifact URL for tmp-main-20241118 vs main:
Compare URL on Github: ROCm/transformers@rocm6.4_testing_rel4.46...ROCm:transformers:tmp-main-20241118

Unit Test Results for develop upstream_sync
```
Aggregated Unit Test Summary:

180 passed, 1 failed, 180 skipped, 180 deselected, 124 warnings
```

Performance Test Results for upstream_sync
<br /> python examples/pytorch/language-modeling/run_mlm.py --model_name_or_path bert-base-uncased --dataset_name wikitext --dataset_config_name wikitext-2-raw-v1 --do_train --do_eval --output_dir /tmp/test-mlm --per_device_train_batch_size 8 --per_device_eval_batch_size 8 --max_steps 500 <br />

@Cemberk
Copy link
Owner Author

Cemberk commented Nov 18, 2024

Unit Test Results for Automated PR tmp-main-20241118

reports/huggingface_unit_tests__run_models_gpu_models/altclip/stats.txt

========= 61 passed, 137 skipped, 123 deselected, 3 warnings in 9.80s ==========
reports/huggingface_unit_tests__run_models_gpu_models/pycache/stats.txt
============================== 1 warning in 0.01s ==============================
reports/huggingface_unit_tests__run_models_gpu_models/albert/stats.txt
========= 39 passed, 105 skipped, 236 deselected, 2 warnings in 14.81s =========
reports/huggingface_unit_tests__run_models_gpu_models/align/stats.txt
========= 56 passed, 143 skipped, 141 deselected, 2 warnings in 15.11s =========
reports/huggingface_unit_tests__run_models_gpu_models/audio_spectrogram_transformer/stats.txt
========== 22 passed, 45 skipped, 117 deselected, 3 warnings in 7.86s ==========


### Performance Test Results for Automated PR tmp-main-20241118

python examples/pytorch/language-modeling/run_mlm.py --model_name_or_path bert-base-uncased --dataset_name wikitext --dataset_config_name wikitext-2-raw-v1 --do_train --do_eval --output_dir /tmp/test-mlm --per_device_train_batch_size 8 --per_device_eval_batch_size 8 --max_steps 500

@Cemberk
Copy link
Owner Author

Cemberk commented Nov 18, 2024

Git diff artifact URL for tmp-main-20241118 vs main:
Compare URL on Github: ROCm/transformers@rocm6.4_testing_rel4.46...ROCm:transformers:tmp-main-20241118

Unit Test Results for develop upstream_sync
```
Aggregated Unit Test Summary:

141 passed, 1 failed, 141 skipped, 141 deselected, 53 warnings
```

Performance Test Results for upstream_sync
<br /> python examples/pytorch/language-modeling/run_mlm.py --model_name_or_path bert-base-uncased --dataset_name wikitext --dataset_config_name wikitext-2-raw-v1 --do_train --do_eval --output_dir /tmp/test-mlm --per_device_train_batch_size 8 --per_device_eval_batch_size 8 --max_steps 500 <br />

@Cemberk
Copy link
Owner Author

Cemberk commented Nov 19, 2024

Git diff artifact URL for tmp-main-20241119 vs main:
Compare URL on Github: ROCm/transformers@rocm6.4_testing_rel4.46...ROCm:transformers:tmp-main-20241119

Unit Test Results for develop upstream_sync
```
Aggregated Unit Test Summary:

141 passed, 1 failed, 141 skipped, 141 deselected, 53 warnings
```

Performance Test Results for upstream_sync
<br /> python examples/pytorch/language-modeling/run_mlm.py --model_name_or_path bert-base-uncased --dataset_name wikitext --dataset_config_name wikitext-2-raw-v1 --do_train --do_eval --output_dir /tmp/test-mlm --per_device_train_batch_size 8 --per_device_eval_batch_size 8 --max_steps 500 <br />

2 similar comments
@Cemberk
Copy link
Owner Author

Cemberk commented Nov 19, 2024

Git diff artifact URL for tmp-main-20241119 vs main:
Compare URL on Github: ROCm/transformers@rocm6.4_testing_rel4.46...ROCm:transformers:tmp-main-20241119

Unit Test Results for develop upstream_sync
```
Aggregated Unit Test Summary:

141 passed, 1 failed, 141 skipped, 141 deselected, 53 warnings
```

Performance Test Results for upstream_sync
<br /> python examples/pytorch/language-modeling/run_mlm.py --model_name_or_path bert-base-uncased --dataset_name wikitext --dataset_config_name wikitext-2-raw-v1 --do_train --do_eval --output_dir /tmp/test-mlm --per_device_train_batch_size 8 --per_device_eval_batch_size 8 --max_steps 500 <br />

@Cemberk
Copy link
Owner Author

Cemberk commented Nov 19, 2024

Git diff artifact URL for tmp-main-20241119 vs main:
Compare URL on Github: ROCm/transformers@rocm6.4_testing_rel4.46...ROCm:transformers:tmp-main-20241119

Unit Test Results for develop upstream_sync
```
Aggregated Unit Test Summary:

141 passed, 1 failed, 141 skipped, 141 deselected, 53 warnings
```

Performance Test Results for upstream_sync
<br /> python examples/pytorch/language-modeling/run_mlm.py --model_name_or_path bert-base-uncased --dataset_name wikitext --dataset_config_name wikitext-2-raw-v1 --do_train --do_eval --output_dir /tmp/test-mlm --per_device_train_batch_size 8 --per_device_eval_batch_size 8 --max_steps 500 <br />

@Cemberk
Copy link
Owner Author

Cemberk commented Nov 19, 2024

Git diff artifact URL for tmp-main-20241119 vs main:
Compare URL on Github: ROCm/transformers@rocm6.4_testing_rel4.46...ROCm:transformers:tmp-main-20241119

Unit Test Results for develop tmp-main-20241119
```
Aggregated Unit Test Summary:

139 passed, failed, 139 skipped, 139 deselected, 139 warnings
```

Performance Test Results for tmp-main-20241119
<br /> python examples/pytorch/language-modeling/run_mlm.py --model_name_or_path bert-base-uncased --dataset_name wikitext --dataset_config_name wikitext-2-raw-v1 --do_train --do_eval --output_dir /tmp/test-mlm --per_device_train_batch_size 8 --per_device_eval_batch_size 8 --max_steps 500 <br />

@Cemberk
Copy link
Owner Author

Cemberk commented Nov 19, 2024

Git diff artifact URL for Upstream Testing vs Downstream Testing:
Compare URL on Github: ROCm/transformers@rocm6.4_testing_rel4.46...huggingface:transformers:v4.46-release

Unit Test Results for Testing Branch rocm6.4_testing_rel4.46
```
Aggregated Unit Test Summary:

145 passed, failed, 145 skipped, 145 deselected, 34 warnings
```

Performance Test Results for rocm6.4_testing_rel4.46
<br /> python examples/pytorch/language-modeling/run_mlm.py --model_name_or_path bert-base-uncased --dataset_name wikitext --dataset_config_name wikitext-2-raw-v1 --do_train --do_eval --output_dir /tmp/test-mlm --per_device_train_batch_size 8 --per_device_eval_batch_size 8 --max_steps 500 <br />

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant