diff --git a/logs/.gitkeep b/logs/.gitkeep deleted file mode 100644 index e69de29..0000000 diff --git a/logs/r36test_adapter.pt.part00 b/logs/r36test_adapter.pt.part00 deleted file mode 100644 index 5538e0d..0000000 Binary files a/logs/r36test_adapter.pt.part00 and /dev/null differ diff --git a/logs/r36test_adapter_manifest.json b/logs/r36test_adapter_manifest.json deleted file mode 100644 index 2657440..0000000 --- a/logs/r36test_adapter_manifest.json +++ /dev/null @@ -1 +0,0 @@ -{"tag": "r36test", "parts": 1, "bytes": 2149, "params": 80, "format": "torch.save(DecisionModel trainable state_dict fp32)"} \ No newline at end of file diff --git a/logs/train-final-nordk.log b/logs/train-final-nordk.log deleted file mode 100644 index 224d807..0000000 --- a/logs/train-final-nordk.log +++ /dev/null @@ -1,42 +0,0 @@ -Traceback (most recent call last): - File "/usr/local/lib/python3.10/dist-packages/transformers/utils/hub.py", line 479, in cached_files - hf_hub_download( - File "/usr/local/lib/python3.10/dist-packages/huggingface_hub/utils/_validators.py", line 106, in _inner_fn - validate_repo_id(arg_value) - File "/usr/local/lib/python3.10/dist-packages/huggingface_hub/utils/_validators.py", line 154, in validate_repo_id - raise HFValidationError( -huggingface_hub.errors.HFValidationError: Repo id must be in the form 'repo_name' or 'namespace/repo_name': '/tmp/base'. Use `repo_type` argument if needed. - -During handling of the above exception, another exception occurred: - -Traceback (most recent call last): - File "/usr/local/lib/python3.10/dist-packages/transformers/configuration_utils.py", line 721, in _get_config_dict - resolved_config_file = cached_file( - File "/usr/local/lib/python3.10/dist-packages/transformers/utils/hub.py", line 322, in cached_file - file = cached_files(path_or_repo_id=path_or_repo_id, filenames=[filename], **kwargs) - File "/usr/local/lib/python3.10/dist-packages/transformers/utils/hub.py", line 531, in cached_files - resolved_files = [ - File "/usr/local/lib/python3.10/dist-packages/transformers/utils/hub.py", line 532, in - _get_cache_file_to_return(path_or_repo_id, filename, cache_dir, revision, repo_type) - File "/usr/local/lib/python3.10/dist-packages/transformers/utils/hub.py", line 143, in _get_cache_file_to_return - resolved_file = try_to_load_from_cache( - File "/usr/local/lib/python3.10/dist-packages/huggingface_hub/utils/_validators.py", line 106, in _inner_fn - validate_repo_id(arg_value) - File "/usr/local/lib/python3.10/dist-packages/huggingface_hub/utils/_validators.py", line 154, in validate_repo_id - raise HFValidationError( -huggingface_hub.errors.HFValidationError: Repo id must be in the form 'repo_name' or 'namespace/repo_name': '/tmp/base'. Use `repo_type` argument if needed. - -During handling of the above exception, another exception occurred: - -Traceback (most recent call last): - File "/tmp/jd/train_laya_noul.py", line 184, in - main() - File "/tmp/jd/train_laya_noul.py", line 97, in main - enc_cfg = AutoConfig.from_pretrained(args.base) - File "/usr/local/lib/python3.10/dist-packages/transformers/models/auto/configuration_auto.py", line 1332, in from_pretrained - config_dict, unused_kwargs = PretrainedConfig.get_config_dict(pretrained_model_name_or_path, **kwargs) - File "/usr/local/lib/python3.10/dist-packages/transformers/configuration_utils.py", line 662, in get_config_dict - config_dict, kwargs = cls._get_config_dict(pretrained_model_name_or_path, **kwargs) - File "/usr/local/lib/python3.10/dist-packages/transformers/configuration_utils.py", line 744, in _get_config_dict - raise OSError( -OSError: Can't load the configuration of '/tmp/base'. If you were trying to load it from 'https://huggingface.co/models', make sure you don't have a local directory with the same name. Otherwise, make sure '/tmp/base' is the correct path to a directory containing a config.json file diff --git a/logs/train-final-v03.log b/logs/train-final-v03.log deleted file mode 100644 index 1eafaac..0000000 --- a/logs/train-final-v03.log +++ /dev/null @@ -1,32 +0,0 @@ -[jev] device=cuda base=/tmp/base -Some weights of XLMRobertaForSequenceClassification were not initialized from the model checkpoint at /tmp/base and are newly initialized: ['classifier.dense.bias', 'classifier.dense.weight', 'classifier.out_proj.bias', 'classifier.out_proj.weight'] -You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference. -trainable params: 1,476,097 || all params: 279,520,514 || trainable%: 0.5281 -[jev] train=36378 calib=5291 -[jev] ep0 step100/2272 loss=0.4184 -[jev] ep0 step200/2272 loss=0.2023 -[jev] ep0 step300/2272 loss=0.1250 -[jev] ep0 step400/2272 loss=0.1653 -[jev] ep0 step500/2272 loss=0.1658 -[jev] ep0 step600/2272 loss=0.2868 -[jev] ep0 step700/2272 loss=0.3366 -[jev] ep0 step800/2272 loss=0.1671 -[jev] ep0 step900/2272 loss=0.2182 -[jev] ep0 step1000/2272 loss=0.1411 -[jev] ep0 step1100/2272 loss=0.2944 -[jev] epoch 0 calibration: {'auc': 0.9934, 'ece': 0.0208, 'n': 5291} -[jev] live-checkpoint 已写 /tmp/jd/model_live -[jev] ep1 step1200/2272 loss=0.1838 -[jev] ep1 step1300/2272 loss=0.1062 -[jev] ep1 step1400/2272 loss=0.2059 -[jev] ep1 step1500/2272 loss=0.4619 -[jev] ep1 step1600/2272 loss=0.1851 -[jev] ep1 step1700/2272 loss=0.2368 -[jev] ep1 step1800/2272 loss=0.2459 -[jev] ep1 step1900/2272 loss=0.5413 -[jev] ep1 step2000/2272 loss=0.0789 -[jev] ep1 step2100/2272 loss=0.1193 -[jev] ep1 step2200/2272 loss=0.1051 -[jev] epoch 1 calibration: {'auc': 0.9937, 'ece': 0.0111, 'n': 5291} -[jev] live-checkpoint 已写 /tmp/jd/model_live -[jev] DONE best={'auc': 0.9937, 'ece': 0.0111, 'epoch': 1} diff --git a/logs/train-final-v04.log b/logs/train-final-v04.log deleted file mode 100644 index 48728ea..0000000 --- a/logs/train-final-v04.log +++ /dev/null @@ -1,33 +0,0 @@ -[jev] device=cuda base=/tmp/base -Some weights of XLMRobertaForSequenceClassification were not initialized from the model checkpoint at /tmp/base and are newly initialized: ['classifier.dense.bias', 'classifier.dense.weight', 'classifier.out_proj.bias', 'classifier.out_proj.weight'] -You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference. -trainable params: 1,476,097 || all params: 279,520,514 || trainable%: 0.5281 -[jev] train=37182 calib=5291 -[jev] ep0 step100/2322 loss=0.2847 -[jev] ep0 step200/2322 loss=0.1989 -[jev] ep0 step300/2322 loss=0.2504 -[jev] ep0 step400/2322 loss=0.3098 -[jev] ep0 step500/2322 loss=0.1724 -[jev] ep0 step600/2322 loss=0.2330 -[jev] ep0 step700/2322 loss=0.2886 -[jev] ep0 step800/2322 loss=0.2370 -[jev] ep0 step900/2322 loss=0.3038 -[jev] ep0 step1000/2322 loss=0.2497 -[jev] ep0 step1100/2322 loss=0.1760 -[jev] epoch 0 calibration: {'auc': 0.9926, 'ece': 0.0072, 'n': 5291} -[jev] live-checkpoint 已写 /tmp/jd/model_live -[jev] ep1 step1200/2322 loss=0.2885 -[jev] ep1 step1300/2322 loss=0.3465 -[jev] ep1 step1400/2322 loss=0.2921 -[jev] ep1 step1500/2322 loss=0.1571 -[jev] ep1 step1600/2322 loss=0.1510 -[jev] ep1 step1700/2322 loss=0.1976 -[jev] ep1 step1800/2322 loss=0.0938 -[jev] ep1 step1900/2322 loss=0.1129 -[jev] ep1 step2000/2322 loss=0.1629 -[jev] ep1 step2100/2322 loss=0.4172 -[jev] ep1 step2200/2322 loss=0.2923 -[jev] ep1 step2300/2322 loss=0.2418 -[jev] epoch 1 calibration: {'auc': 0.9935, 'ece': 0.0083, 'n': 5291} -[jev] live-checkpoint 已写 /tmp/jd/model_live -[jev] DONE best={'auc': 0.9935, 'ece': 0.0083, 'epoch': 1} diff --git a/logs/train-final-v05.log b/logs/train-final-v05.log deleted file mode 100644 index e5bf5bf..0000000 --- a/logs/train-final-v05.log +++ /dev/null @@ -1,34 +0,0 @@ -[jev] device=cuda base=/tmp/base tokenizers=0.22.2 -Some weights of ModernBertForSequenceClassification were not initialized from the model checkpoint at /tmp/base and are newly initialized: ['classifier.bias', 'classifier.weight'] -You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference. -[jev] LoRA targets: ['Wqkv', 'Wo'] -trainable params: 2,298,625 || all params: 309,829,634 || trainable%: 0.7419 -[jev] train=27954 calib=5291 -[jev] canaries: ['en', 'zh'] -[jev] ep0 step100/1746 loss=0.2554 -[jev] ep0 step200/1746 loss=0.4286 -[jev] ep0 step300/1746 loss=0.4042 -[jev] ep0 step400/1746 loss=0.3410 -[jev] ep0 step500/1746 loss=0.3232 -[jev] ep0 step600/1746 loss=0.2050 -[jev] ep0 step700/1746 loss=0.2220 -[jev] ep0 step800/1746 loss=0.2669 -/usr/local/lib/python3.10/dist-packages/torch/_inductor/compile_fx.py:282: UserWarning: TensorFloat32 tensor cores for float32 matrix multiplication available but not enabled. Consider setting `torch.set_float32_matmul_precision('high')` for better performance. - warnings.warn( -[jev] epoch 0 calibration: {'auc': 0.9936, 'ece': 0.0129, 'n': 5291} -[jev] epoch 0 canary_en: {'auc': 0.8138, 'ece': 0.075, 'n': 300} -[jev] epoch 0 canary_zh: {'auc': 1.0, 'ece': 0.0038, 'n': 300} -[jev] live-checkpoint 已写 /tmp/jd/model_live -[jev] ep1 step900/1746 loss=0.2454 -[jev] ep1 step1000/1746 loss=0.2127 -[jev] ep1 step1100/1746 loss=0.1612 -[jev] ep1 step1200/1746 loss=0.1135 -[jev] ep1 step1300/1746 loss=0.1401 -[jev] ep1 step1400/1746 loss=0.1381 -[jev] ep1 step1500/1746 loss=0.3087 -[jev] ep1 step1600/1746 loss=0.1147 -[jev] ep1 step1700/1746 loss=0.1655 -[jev] epoch 1 calibration: {'auc': 0.9949, 'ece': 0.0105, 'n': 5291} -[jev] epoch 1 canary_en: {'auc': 0.8466, 'ece': 0.066, 'n': 300} -[jev] epoch 1 canary_zh: {'auc': 1.0, 'ece': 0.0032, 'n': 300} -[jev] DONE best={'auc': 0.9936, 'ece': 0.0129, 'epoch': 0} diff --git a/logs/train-final.log b/logs/train-final.log deleted file mode 100644 index d7079ac..0000000 --- a/logs/train-final.log +++ /dev/null @@ -1,32 +0,0 @@ -[jev] device=cuda base=/tmp/base -Some weights of XLMRobertaForSequenceClassification were not initialized from the model checkpoint at /tmp/base and are newly initialized: ['classifier.dense.bias', 'classifier.dense.weight', 'classifier.out_proj.bias', 'classifier.out_proj.weight'] -You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference. -trainable params: 1,476,097 || all params: 279,520,514 || trainable%: 0.5281 -[jev] train=35848 calib=5291 -[jev] ep0 step100/2240 loss=0.3442 -[jev] ep0 step200/2240 loss=0.1200 -[jev] ep0 step300/2240 loss=0.3088 -[jev] ep0 step400/2240 loss=0.3102 -[jev] ep0 step500/2240 loss=0.0746 -[jev] ep0 step600/2240 loss=0.1947 -[jev] ep0 step700/2240 loss=0.2139 -[jev] ep0 step800/2240 loss=0.2837 -[jev] ep0 step900/2240 loss=0.1628 -[jev] ep0 step1000/2240 loss=0.1508 -[jev] ep0 step1100/2240 loss=0.2874 -[jev] epoch 0 calibration: {'auc': 0.9936, 'ece': 0.0118, 'n': 5291} -[jev] live-checkpoint 已写 /tmp/jd/model_live -[jev] ep1 step1200/2240 loss=0.5534 -[jev] ep1 step1300/2240 loss=0.2997 -[jev] ep1 step1400/2240 loss=0.2127 -[jev] ep1 step1500/2240 loss=0.1171 -[jev] ep1 step1600/2240 loss=0.2262 -[jev] ep1 step1700/2240 loss=0.0793 -[jev] ep1 step1800/2240 loss=0.1416 -[jev] ep1 step1900/2240 loss=0.2552 -[jev] ep1 step2000/2240 loss=0.0808 -[jev] ep1 step2100/2240 loss=0.1460 -[jev] ep1 step2200/2240 loss=0.0818 -[jev] epoch 1 calibration: {'auc': 0.9939, 'ece': 0.0138, 'n': 5291} -[jev] live-checkpoint 已写 /tmp/jd/model_live -[jev] DONE best={'auc': 0.9939, 'ece': 0.0138, 'epoch': 1} diff --git a/logs/train-progress.log b/logs/train-progress.log deleted file mode 100644 index 87401c4..0000000 --- a/logs/train-progress.log +++ /dev/null @@ -1,30 +0,0 @@ -[jev] device=cuda base=/tmp/base tokenizers=0.22.2 -Some weights of ModernBertForSequenceClassification were not initialized from the model checkpoint at /tmp/base and are newly initialized: ['classifier.bias', 'classifier.weight'] -You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference. -[jev] LoRA targets: ['Wqkv', 'Wo'] -trainable params: 2,298,625 || all params: 309,829,634 || trainable%: 0.7419 -[jev] train=27954 calib=5291 -[jev] canaries: ['en', 'zh'] -[jev] ep0 step100/1746 loss=0.2554 -[jev] ep0 step200/1746 loss=0.4286 -[jev] ep0 step300/1746 loss=0.4042 -[jev] ep0 step400/1746 loss=0.3410 -[jev] ep0 step500/1746 loss=0.3232 -[jev] ep0 step600/1746 loss=0.2050 -[jev] ep0 step700/1746 loss=0.2220 -[jev] ep0 step800/1746 loss=0.2669 -/usr/local/lib/python3.10/dist-packages/torch/_inductor/compile_fx.py:282: UserWarning: TensorFloat32 tensor cores for float32 matrix multiplication available but not enabled. Consider setting `torch.set_float32_matmul_precision('high')` for better performance. - warnings.warn( -[jev] epoch 0 calibration: {'auc': 0.9936, 'ece': 0.0129, 'n': 5291} -[jev] epoch 0 canary_en: {'auc': 0.8138, 'ece': 0.075, 'n': 300} -[jev] epoch 0 canary_zh: {'auc': 1.0, 'ece': 0.0038, 'n': 300} -[jev] live-checkpoint 已写 /tmp/jd/model_live -[jev] ep1 step900/1746 loss=0.2454 -[jev] ep1 step1000/1746 loss=0.2127 -[jev] ep1 step1100/1746 loss=0.1612 -[jev] ep1 step1200/1746 loss=0.1135 -[jev] ep1 step1300/1746 loss=0.1401 -[jev] ep1 step1400/1746 loss=0.1381 -[jev] ep1 step1500/1746 loss=0.3087 -[jev] ep1 step1600/1746 loss=0.1147 -[jev] ep1 step1700/1746 loss=0.1655