Training in progress, step 18000

.gitattributes

@@ -28,3 +28,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 wandb/run-20220322_102259-2z8qc0km/run-2z8qc0km.wandb filter=lfs diff=lfs merge=lfs -text
 wandb/run-20220322_102259-2z8qc0km/logs/debug-internal.log filter=lfs diff=lfs merge=lfs -text
 wandb/run-20220322_102259-2z8qc0km/files/output.log filter=lfs diff=lfs merge=lfs -text
+wandb/run-20220323_161301-3mj7aa54/run-3mj7aa54.wandb filter=lfs diff=lfs merge=lfs -text
+wandb/run-20220324_170220-1u8cita3/run-1u8cita3.wandb filter=lfs diff=lfs merge=lfs -text

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c3959973a22398139163804e65a2b46e37d925aca294e5169f93c5443127725f
+oid sha256:37bef9d593dfcaf3fe6c5b2444e15a28574e4992fdeea2809be6692e58d70952
 size 2278381245

run_librispeech.sh

@@ -9,7 +9,6 @@ CUDA_VISIBLE_DEVICES=1 python run_speech_recognition_seq2seq.py \
 	--output_dir="./" \
 	--preprocessing_num_workers="1" \
 	--length_column_name="input_length" \
-	--overwrite_output_dir \
 	--num_train_epochs="25" \
 	--per_device_train_batch_size="8" \
 	--per_device_eval_batch_size="8" \

runs/Mar22_10-22-32_sanchit--v100/events.out.tfevents.1647944579.sanchit--v100.30187.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:43646c0a27b06628d87f08566b8310a17d1fa026ec540b0cbcd8f94659576342
-size 2603566
+oid sha256:7c02428e95c8fa7bb06e0d83901936f7298d7dc99a53c1838cde89a7261d2b2f
+size 2698926

runs/Mar23_16-01-22_sanchit--v100/1648051306.4960403/events.out.tfevents.1648051306.sanchit--v100.1748630.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0d3e13f1f911aaf64c1d14143cc403c81bce759664142d77264f0c4d000ac5fe
+size 4973

runs/Mar23_16-01-22_sanchit--v100/events.out.tfevents.1648051306.sanchit--v100.1748630.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c67de6c471ae52bab7cbd53e59bdb8648f90a1eb96ce591a6ff165d15f0383d0
+size 9592

runs/Mar23_16-12-34_sanchit--v100/1648051981.1882057/events.out.tfevents.1648051981.sanchit--v100.1748915.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8fc7acf27812020fe764f7c29bd03dbee44a6354d7513ca69ca713583ebb0f91
+size 4973

runs/Mar23_16-12-34_sanchit--v100/events.out.tfevents.1648051981.sanchit--v100.1748915.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b40018955a0243f594aefbea3ecd3e683c9ea292663fa67ae26225c0a5bddb6c
+size 55513

runs/Mar24_17-01-37_sanchit--v100/1648141340.1570067/events.out.tfevents.1648141340.sanchit--v100.1754191.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:79f4aa7292e7a8a83dbdf5c768c7899899e75d427c253b7ec822053f16b5ce24
+size 4973

runs/Mar24_17-01-37_sanchit--v100/events.out.tfevents.1648141340.sanchit--v100.1754191.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:08c83bb4dd6d22b9633d2de2452a52e8c2ce422333ffe14809a99e83b459a6f0
+size 249917

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:92ad3209da41567672c357aa38a7d29fdf01ade5a4b62724c505a6969ece7279
+oid sha256:b25e09af537a2ef6c34fdf5b2281371d7b7990eb9de03197e62880e0ff1a819e
 size 3119

wandb/debug-internal.log

@@ -1 +1 @@
-run-20220322_102259-2z8qc0km/logs/debug-internal.log
+run-20220324_170220-1u8cita3/logs/debug-internal.log

wandb/debug.log

@@ -1 +1 @@
-run-20220322_102259-2z8qc0km/logs/debug.log
+run-20220324_170220-1u8cita3/logs/debug.log

wandb/latest-run

@@ -1 +1 @@
-run-20220322_102259-2z8qc0km
+run-20220324_170220-1u8cita3

wandb/run-20220322_102259-2z8qc0km/files/output.log

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:379e12bafb4fd5cd0974ba719f68cfd001d30aa8e88142ffde6ef7b39dd0da71
-size 21664085
+oid sha256:c5e19b6f9fd78378d00cd5f1ee6a06789f82a10ec2ea5bde0197a91c8943d4a8
+size 22762332

wandb/run-20220322_102259-2z8qc0km/logs/debug-internal.log

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bbfe467120a8a2f343c8f856556b2c7a92a6f1c9f1845f3636d6540677e6fea2
-size 24128747
+oid sha256:310380712affbcbd4b0659534f36ac0d6a67251248a9ab50131051a6f9c099f6
+size 24958798

wandb/run-20220322_102259-2z8qc0km/run-2z8qc0km.wandb

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:867a8987ba4407e2ed316ff20b90836d28031527afb275c34658677356ee6d9e
-size 833177901
+oid sha256:67c9de9e798335a4dd158007f1df840fe39202ae663a415013e8195faebbc408
+size 862665856

wandb/run-20220323_160146-1mg2wnhf/files/config.yaml

@@ -0,0 +1,24 @@
+wandb_version: 1
+
+_wandb:
+  desc: null
+  value:
+    cli_version: 0.12.10
+    framework: huggingface
+    huggingface_version: 4.17.0.dev0
+    is_jupyter_run: false
+    is_kaggle_kernel: false
+    python_version: 3.9.5
+    start_time: 1648051306
+    t:
+      1:
+      - 1
+      - 5
+      - 11
+      3:
+      - 13
+      4: 3.9.5
+      5: 0.12.10
+      6: 4.17.0.dev0
+      8:
+      - 5

wandb/run-20220323_160146-1mg2wnhf/files/output.log

@@ -0,0 +1,2 @@
+
+

wandb/run-20220323_160146-1mg2wnhf/files/requirements.txt

@@ -0,0 +1,186 @@
+absl-py==1.0.0
+aiohttp==3.8.1
+aiosignal==1.2.0
+anyio==3.5.0
+appdirs==1.4.4
+argon2-cffi-bindings==21.2.0
+argon2-cffi==21.3.0
+asttokens==2.0.5
+async-timeout==4.0.2
+attrs==21.4.0
+audioread==2.1.9
+babel==2.9.1
+backcall==0.2.0
+bitsandbytes-cuda113==0.26.0
+black==22.1.0
+bleach==4.1.0
+cachetools==5.0.0
+certifi==2021.10.8
+cffi==1.15.0
+charset-normalizer==2.0.11
+chex==0.1.0
+click==8.0.3
+clldutils==3.10.1
+colorlog==6.6.0
+csvw==1.11.0
+cycler==0.11.0
+datasets==1.18.3
+debugpy==1.5.1
+decorator==5.1.1
+defusedxml==0.7.1
+dill==0.3.4
+dlinfo==1.2.1
+dm-tree==0.1.6
+docker-pycreds==0.4.0
+entrypoints==0.4
+executing==0.8.2
+filelock==3.4.2
+flatbuffers==2.0
+flax==0.4.0
+fonttools==4.29.1
+frozenlist==1.3.0
+fsspec==2022.1.0
+gitdb==4.0.9
+gitpython==3.1.27
+google-auth-oauthlib==0.4.6
+google-auth==2.6.0
+grpcio==1.43.0
+huggingface-hub==0.4.0
+hypothesis==6.36.1
+idna==3.3
+importlib-metadata==4.10.1
+ipdb==0.13.9
+ipykernel==6.8.0
+ipython-genutils==0.2.0
+ipython==8.0.1
+ipywidgets==7.6.5
+isodate==0.6.1
+jax==0.2.28
+jaxlib==0.1.76+cuda11.cudnn82
+jedi==0.18.1
+jinja2==3.0.3
+jiwer==2.3.0
+joblib==1.1.0
+json5==0.9.6
+jsonschema==4.4.0
+jupyter-client==7.1.2
+jupyter-console==6.4.0
+jupyter-core==4.9.1
+jupyter-server==1.13.5
+jupyter==1.0.0
+jupyterlab-pygments==0.1.2
+jupyterlab-server==2.10.3
+jupyterlab-widgets==1.0.2
+jupyterlab==3.2.9
+kiwisolver==1.3.2
+librosa==0.8.1
+llvmlite==0.38.0
+markdown==3.3.6
+markupsafe==2.0.1
+matplotlib-inline==0.1.3
+matplotlib==3.5.1
+mistune==0.8.4
+msgpack==1.0.3
+multidict==6.0.2
+multiprocess==0.70.12.2
+mypy-extensions==0.4.3
+nbclassic==0.3.5
+nbclient==0.5.10
+nbconvert==6.4.1
+nbformat==5.1.3
+nest-asyncio==1.5.4
+notebook==6.4.8
+numba==0.55.1
+numpy==1.21.5
+oauthlib==3.2.0
+opt-einsum==3.3.0
+optax==0.1.0
+packaging==21.3
+pandas==1.4.0
+pandocfilters==1.5.0
+parso==0.8.3
+pathspec==0.9.0
+pathtools==0.1.2
+pexpect==4.8.0
+phonemizer==3.0.1
+pickleshare==0.7.5
+pillow==9.0.0
+pip==22.0.2
+pkg-resources==0.0.0
+platformdirs==2.4.1
+pooch==1.6.0
+prometheus-client==0.13.1
+promise==2.3
+prompt-toolkit==3.0.26
+protobuf==3.19.4
+psutil==5.9.0
+ptyprocess==0.7.0
+pure-eval==0.2.2
+pyarrow==6.0.1
+pyasn1-modules==0.2.8
+pyasn1==0.4.8
+pycparser==2.21
+pyctcdecode==0.3.0
+pygments==2.11.2
+pygtrie==2.4.2
+pyparsing==3.0.7
+pyrsistent==0.18.1
+python-dateutil==2.8.2
+python-levenshtein==0.12.2
+pytz==2021.3
+pyyaml==6.0
+pyzmq==22.3.0
+qtconsole==5.2.2
+qtpy==2.0.1
+regex==2022.1.18
+requests-oauthlib==1.3.1
+requests==2.27.1
+resampy==0.2.2
+rfc3986==2.0.0
+rsa==4.8
+sacremoses==0.0.47
+scikit-learn==1.0.2
+scipy==1.7.3
+segments==2.2.0
+send2trash==1.8.0
+sentry-sdk==1.5.6
+setuptools==44.1.1
+shortuuid==1.0.8
+six==1.16.0
+smmap==5.0.0
+sniffio==1.2.0
+sortedcontainers==2.4.0
+soundfile==0.10.3.post1
+stack-data==0.1.4
+tabulate==0.8.9
+tensorboard-data-server==0.6.1
+tensorboard-plugin-wit==1.8.1
+tensorboard==2.8.0
+termcolor==1.1.0
+terminado==0.13.1
+testpath==0.5.0
+threadpoolctl==3.1.0
+tokenizers==0.11.4
+toml==0.10.2
+tomli==2.0.0
+toolz==0.11.2
+torch==1.10.2+cu113
+torchaudio==0.10.2+cu113
+tornado==6.1
+tqdm==4.62.3
+traitlets==5.1.1
+transformers==4.17.0.dev0
+typing-extensions==3.10.0.2
+uritemplate==4.1.1
+urllib3==1.26.8
+wandb==0.12.10
+wcwidth==0.2.5
+webencodings==0.5.1
+websocket-client==1.2.3
+werkzeug==2.0.2
+wheel==0.37.1
+widgetsnbextension==3.5.2
+xxhash==2.0.2
+yarl==1.7.2
+yaspin==2.1.0
+zipp==3.7.0

wandb/run-20220323_160146-1mg2wnhf/files/wandb-metadata.json

@@ -0,0 +1,59 @@
+{
+    "os": "Linux-5.11.0-1028-gcp-x86_64-with-glibc2.33",
+    "python": "3.9.5",
+    "heartbeatAt": "2022-03-23T16:01:47.894818",
+    "startedAt": "2022-03-23T16:01:46.813621",
+    "docker": null,
+    "gpu": "Tesla V100-SXM2-16GB",
+    "gpu_count": 2,
+    "cpu_count": 16,
+    "cuda": null,
+    "args": [
+        "--dataset_name=librispeech_asr",
+        "--model_name_or_path=./",
+        "--tokenizer_name=./",
+        "--dataset_config_name=clean",
+        "--train_split_name=train.100",
+        "--eval_split_name=validation",
+        "--output_dir=./",
+        "--preprocessing_num_workers=1",
+        "--length_column_name=input_length",
+        "--overwrite_output_dir",
+        "--num_train_epochs=25",
+        "--per_device_train_batch_size=8",
+        "--per_device_eval_batch_size=8",
+        "--gradient_accumulation_steps=4",
+        "--generation_max_length=40",
+        "--generation_num_beams=1",
+        "--learning_rate=3e-5",
+        "--warmup_steps=1000",
+        "--evaluation_strategy=steps",
+        "--text_column_name=text",
+        "--save_steps=1500",
+        "--eval_steps=1500",
+        "--logging_steps=1",
+        "--save_total_limit=1",
+        "--freeze_feature_encoder",
+        "--gradient_checkpointing",
+        "--fp16",
+        "--group_by_length",
+        "--predict_with_generate",
+        "--do_lower_case",
+        "--do_eval",
+        "--do_train",
+        "--push_to_hub",
+        "--use_auth_token"
+    ],
+    "state": "running",
+    "program": "/home/sanchit_huggingface_co/wav2vec2-2-rnd-no-adapter-regularisation/run_speech_recognition_seq2seq.py",
+    "codePath": "run_speech_recognition_seq2seq.py",
+    "git": {
+        "remote": "https://huggingface.co/sanchit-gandhi/wav2vec2-2-rnd-no-adapter-regularisation",
+        "commit": "156c71831efc84fc27b6a9a52f81872c0138b990"
+    },
+    "email": "sanchit@huggingface.co",
+    "root": "/home/sanchit_huggingface_co/wav2vec2-2-rnd-no-adapter-regularisation",
+    "host": "sanchit--v100",
+    "username": "sanchit_huggingface_co",
+    "executable": "/home/sanchit_huggingface_co/gcp/bin/python"
+}

wandb/run-20220323_160146-1mg2wnhf/files/wandb-summary.json

@@ -0,0 +1 @@
+{}

wandb/run-20220323_160146-1mg2wnhf/logs/debug-internal.log

@@ -0,0 +1,54 @@
+2022-03-23 16:01:47,672 INFO    MainThread:1748732 [internal.py:wandb_internal():89] W&B internal server running at pid: 1748732, started at: 2022-03-23 16:01:47.672124
+2022-03-23 16:01:47,674 INFO    WriterThread:1748732 [datastore.py:open_for_write():77] open: /home/sanchit_huggingface_co/wav2vec2-2-rnd-no-adapter-regularisation/wandb/run-20220323_160146-1mg2wnhf/run-1mg2wnhf.wandb
+2022-03-23 16:01:47,676 DEBUG   SenderThread:1748732 [sender.py:send():235] send: header
+2022-03-23 16:01:47,676 DEBUG   HandlerThread:1748732 [handler.py:handle_request():131] handle_request: check_version
+2022-03-23 16:01:47,676 DEBUG   SenderThread:1748732 [sender.py:send_request():249] send_request: check_version
+2022-03-23 16:01:47,744 DEBUG   SenderThread:1748732 [sender.py:send():235] send: run
+2022-03-23 16:01:47,889 INFO    SenderThread:1748732 [dir_watcher.py:__init__():169] watching files in: /home/sanchit_huggingface_co/wav2vec2-2-rnd-no-adapter-regularisation/wandb/run-20220323_160146-1mg2wnhf/files
+2022-03-23 16:01:47,889 INFO    SenderThread:1748732 [sender.py:_start_run_threads():809] run started: 1mg2wnhf with start time 1648051306
+2022-03-23 16:01:47,890 DEBUG   SenderThread:1748732 [sender.py:send():235] send: summary
+2022-03-23 16:01:47,890 INFO    SenderThread:1748732 [sender.py:_save_file():944] saving file wandb-summary.json with policy end
+2022-03-23 16:01:47,890 DEBUG   HandlerThread:1748732 [handler.py:handle_request():131] handle_request: run_start
+2022-03-23 16:01:47,894 DEBUG   HandlerThread:1748732 [meta.py:__init__():36] meta init
+2022-03-23 16:01:47,894 DEBUG   HandlerThread:1748732 [meta.py:__init__():50] meta init done
+2022-03-23 16:01:47,894 DEBUG   HandlerThread:1748732 [meta.py:probe():210] probe
+2022-03-23 16:01:47,901 DEBUG   HandlerThread:1748732 [meta.py:_setup_git():200] setup git
+2022-03-23 16:01:47,915 DEBUG   HandlerThread:1748732 [meta.py:_setup_git():207] setup git done
+2022-03-23 16:01:47,915 DEBUG   HandlerThread:1748732 [meta.py:_save_pip():54] save pip
+2022-03-23 16:01:47,916 DEBUG   HandlerThread:1748732 [meta.py:_save_pip():68] save pip done
+2022-03-23 16:01:47,916 DEBUG   HandlerThread:1748732 [meta.py:probe():248] probe done
+2022-03-23 16:01:47,992 DEBUG   SenderThread:1748732 [sender.py:send():235] send: files
+2022-03-23 16:01:47,992 INFO    SenderThread:1748732 [sender.py:_save_file():944] saving file wandb-metadata.json with policy now
+2022-03-23 16:01:47,996 DEBUG   HandlerThread:1748732 [handler.py:handle_request():131] handle_request: stop_status
+2022-03-23 16:01:47,997 DEBUG   SenderThread:1748732 [sender.py:send_request():249] send_request: stop_status
+2022-03-23 16:01:48,035 DEBUG   SenderThread:1748732 [sender.py:send():235] send: config
+2022-03-23 16:01:48,035 DEBUG   SenderThread:1748732 [sender.py:send():235] send: metric
+2022-03-23 16:01:48,036 DEBUG   SenderThread:1748732 [sender.py:send():235] send: metric
+2022-03-23 16:01:48,036 WARNING SenderThread:1748732 [sender.py:send_metric():902] Seen metric with glob (shouldnt happen)
+2022-03-23 16:01:48,267 INFO    Thread-11 :1748732 [upload_job.py:push():137] Uploaded file /tmp/tmpxct45m19wandb/18lpj5zq-wandb-metadata.json
+2022-03-23 16:01:48,891 INFO    Thread-8  :1748732 [dir_watcher.py:_on_file_created():217] file/dir created: /home/sanchit_huggingface_co/wav2vec2-2-rnd-no-adapter-regularisation/wandb/run-20220323_160146-1mg2wnhf/files/output.log
+2022-03-23 16:01:48,892 INFO    Thread-8  :1748732 [dir_watcher.py:_on_file_created():217] file/dir created: /home/sanchit_huggingface_co/wav2vec2-2-rnd-no-adapter-regularisation/wandb/run-20220323_160146-1mg2wnhf/files/wandb-summary.json
+2022-03-23 16:01:48,892 INFO    Thread-8  :1748732 [dir_watcher.py:_on_file_created():217] file/dir created: /home/sanchit_huggingface_co/wav2vec2-2-rnd-no-adapter-regularisation/wandb/run-20220323_160146-1mg2wnhf/files/wandb-metadata.json
+2022-03-23 16:01:48,892 INFO    Thread-8  :1748732 [dir_watcher.py:_on_file_created():217] file/dir created: /home/sanchit_huggingface_co/wav2vec2-2-rnd-no-adapter-regularisation/wandb/run-20220323_160146-1mg2wnhf/files/requirements.txt
+2022-03-23 16:01:50,891 INFO    Thread-8  :1748732 [dir_watcher.py:_on_file_modified():230] file/dir modified: /home/sanchit_huggingface_co/wav2vec2-2-rnd-no-adapter-regularisation/wandb/run-20220323_160146-1mg2wnhf/files/output.log
+2022-03-23 16:01:52,740 WARNING MainThread:1748732 [internal.py:wandb_internal():152] Internal process interrupt: 1
+2022-03-23 16:01:52,891 INFO    Thread-8  :1748732 [dir_watcher.py:_on_file_modified():230] file/dir modified: /home/sanchit_huggingface_co/wav2vec2-2-rnd-no-adapter-regularisation/wandb/run-20220323_160146-1mg2wnhf/files/output.log
+2022-03-23 16:01:53,007 WARNING MainThread:1748732 [internal.py:wandb_internal():152] Internal process interrupt: 2
+2022-03-23 16:01:53,007 ERROR   MainThread:1748732 [internal.py:wandb_internal():155] Internal process interrupted.
+2022-03-23 16:01:53,007 INFO    HandlerThread:1748732 [handler.py:finish():739] shutting down handler
+2022-03-23 16:01:54,006 INFO    SenderThread:1748732 [sender.py:finish():1075] shutting down sender
+2022-03-23 16:01:54,007 INFO    WriterThread:1748732 [datastore.py:close():281] close: /home/sanchit_huggingface_co/wav2vec2-2-rnd-no-adapter-regularisation/wandb/run-20220323_160146-1mg2wnhf/run-1mg2wnhf.wandb
+2022-03-23 16:01:54,007 INFO    SenderThread:1748732 [dir_watcher.py:finish():283] shutting down directory watcher
+2022-03-23 16:01:54,892 INFO    SenderThread:1748732 [dir_watcher.py:finish():313] scan: /home/sanchit_huggingface_co/wav2vec2-2-rnd-no-adapter-regularisation/wandb/run-20220323_160146-1mg2wnhf/files
+2022-03-23 16:01:54,892 INFO    SenderThread:1748732 [dir_watcher.py:finish():327] scan save: /home/sanchit_huggingface_co/wav2vec2-2-rnd-no-adapter-regularisation/wandb/run-20220323_160146-1mg2wnhf/files/wandb-metadata.json wandb-metadata.json
+2022-03-23 16:01:54,892 INFO    SenderThread:1748732 [dir_watcher.py:finish():327] scan save: /home/sanchit_huggingface_co/wav2vec2-2-rnd-no-adapter-regularisation/wandb/run-20220323_160146-1mg2wnhf/files/output.log output.log
+2022-03-23 16:01:54,893 INFO    SenderThread:1748732 [dir_watcher.py:finish():327] scan save: /home/sanchit_huggingface_co/wav2vec2-2-rnd-no-adapter-regularisation/wandb/run-20220323_160146-1mg2wnhf/files/wandb-summary.json wandb-summary.json
+2022-03-23 16:01:54,893 INFO    SenderThread:1748732 [dir_watcher.py:finish():327] scan save: /home/sanchit_huggingface_co/wav2vec2-2-rnd-no-adapter-regularisation/wandb/run-20220323_160146-1mg2wnhf/files/requirements.txt requirements.txt
+2022-03-23 16:01:54,893 INFO    SenderThread:1748732 [dir_watcher.py:finish():327] scan save: /home/sanchit_huggingface_co/wav2vec2-2-rnd-no-adapter-regularisation/wandb/run-20220323_160146-1mg2wnhf/files/config.yaml config.yaml
+2022-03-23 16:01:54,893 INFO    SenderThread:1748732 [file_pusher.py:finish():177] shutting down file pusher
+2022-03-23 16:01:54,893 INFO    SenderThread:1748732 [file_pusher.py:join():182] waiting for file pusher
+2022-03-23 16:01:55,208 INFO    Thread-15 :1748732 [upload_job.py:push():137] Uploaded file /home/sanchit_huggingface_co/wav2vec2-2-rnd-no-adapter-regularisation/wandb/run-20220323_160146-1mg2wnhf/files/config.yaml
+2022-03-23 16:01:55,216 INFO    Thread-14 :1748732 [upload_job.py:push():137] Uploaded file /home/sanchit_huggingface_co/wav2vec2-2-rnd-no-adapter-regularisation/wandb/run-20220323_160146-1mg2wnhf/files/requirements.txt
+2022-03-23 16:01:55,218 INFO    Thread-12 :1748732 [upload_job.py:push():137] Uploaded file /home/sanchit_huggingface_co/wav2vec2-2-rnd-no-adapter-regularisation/wandb/run-20220323_160146-1mg2wnhf/files/output.log
+2022-03-23 16:01:55,219 INFO    Thread-13 :1748732 [upload_job.py:push():137] Uploaded file /home/sanchit_huggingface_co/wav2vec2-2-rnd-no-adapter-regularisation/wandb/run-20220323_160146-1mg2wnhf/files/wandb-summary.json
+2022-03-23 16:01:55,951 INFO    MainThread:1748732 [internal.py:handle_exit():79] Internal process exited

wandb/run-20220323_160146-1mg2wnhf/logs/debug.log

@@ -0,0 +1,29 @@
+2022-03-23 16:01:46,815 INFO    MainThread:1748630 [wandb_setup.py:_flush():75] Loading settings from /home/sanchit_huggingface_co/.config/wandb/settings
+2022-03-23 16:01:46,815 INFO    MainThread:1748630 [wandb_setup.py:_flush():75] Loading settings from /home/sanchit_huggingface_co/wav2vec2-2-rnd-no-adapter-regularisation/wandb/settings
+2022-03-23 16:01:46,815 INFO    MainThread:1748630 [wandb_setup.py:_flush():75] Loading settings from environment variables: {}
+2022-03-23 16:01:46,815 INFO    MainThread:1748630 [wandb_setup.py:_flush():75] Inferring run settings from compute environment: {'program_relpath': 'run_speech_recognition_seq2seq.py', 'program': '/home/sanchit_huggingface_co/wav2vec2-2-rnd-no-adapter-regularisation/run_speech_recognition_seq2seq.py'}
+2022-03-23 16:01:46,815 INFO    MainThread:1748630 [wandb_init.py:_log_setup():386] Logging user logs to /home/sanchit_huggingface_co/wav2vec2-2-rnd-no-adapter-regularisation/wandb/run-20220323_160146-1mg2wnhf/logs/debug.log
+2022-03-23 16:01:46,815 INFO    MainThread:1748630 [wandb_init.py:_log_setup():387] Logging internal logs to /home/sanchit_huggingface_co/wav2vec2-2-rnd-no-adapter-regularisation/wandb/run-20220323_160146-1mg2wnhf/logs/debug-internal.log
+2022-03-23 16:01:46,815 INFO    MainThread:1748630 [wandb_init.py:init():420] calling init triggers
+2022-03-23 16:01:46,815 INFO    MainThread:1748630 [wandb_init.py:init():425] wandb.init called with sweep_config: {}
+config: {}
+2022-03-23 16:01:46,815 INFO    MainThread:1748630 [wandb_init.py:init():471] starting backend
+2022-03-23 16:01:46,815 INFO    MainThread:1748630 [backend.py:_multiprocessing_setup():99] multiprocessing start_methods=fork,spawn,forkserver, using: spawn
+2022-03-23 16:01:46,868 INFO    MainThread:1748630 [backend.py:ensure_launched():219] starting backend process...
+2022-03-23 16:01:46,921 INFO    MainThread:1748630 [backend.py:ensure_launched():224] started backend process with pid: 1748732
+2022-03-23 16:01:46,923 INFO    MainThread:1748630 [wandb_init.py:init():480] backend started and connected
+2022-03-23 16:01:46,934 INFO    MainThread:1748630 [wandb_init.py:init():550] updated telemetry
+2022-03-23 16:01:47,062 INFO    MainThread:1748630 [wandb_init.py:init():581] communicating current version
+2022-03-23 16:01:47,742 INFO    MainThread:1748630 [wandb_init.py:init():586] got version response upgrade_message: "wandb version 0.12.11 is available!  To upgrade, please run:\n $ pip install wandb --upgrade"
+
+2022-03-23 16:01:47,743 INFO    MainThread:1748630 [wandb_init.py:init():596] communicating run to backend with 30 second timeout
+2022-03-23 16:01:47,890 INFO    MainThread:1748630 [wandb_init.py:init():624] starting run threads in backend
+2022-03-23 16:01:47,996 INFO    MainThread:1748630 [wandb_run.py:_console_start():1827] atexit reg
+2022-03-23 16:01:47,996 INFO    MainThread:1748630 [wandb_run.py:_redirect():1701] redirect: SettingsConsole.REDIRECT
+2022-03-23 16:01:47,997 INFO    MainThread:1748630 [wandb_run.py:_redirect():1706] Redirecting console.
+2022-03-23 16:01:47,998 INFO    MainThread:1748630 [wandb_run.py:_redirect():1762] Redirects installed.
+2022-03-23 16:01:47,999 INFO    MainThread:1748630 [wandb_init.py:init():651] run started, returning control to user process
+2022-03-23 16:01:48,001 INFO    MainThread:1748630 [wandb_run.py:_config_callback():966] config_cb None None {'return_dict': True, 'output_hidden_states': False, 'output_attentions': False, 'torchscript': False, 'torch_dtype': 'torch.float32', 'use_bfloat16': False, 'pruned_heads': {}, 'tie_word_embeddings': False, 'is_encoder_decoder': True, 'is_decoder': False, 'cross_attention_hidden_size': None, 'add_cross_attention': False, 'tie_encoder_decoder': False, 'max_length': 50, 'min_length': 0, 'do_sample': False, 'early_stopping': False, 'num_beams': 1, 'num_beam_groups': 1, 'diversity_penalty': 0.0, 'temperature': 1.0, 'top_k': 50, 'top_p': 1.0, 'repetition_penalty': 1.0, 'length_penalty': 1.0, 'no_repeat_ngram_size': 0, 'encoder_no_repeat_ngram_size': 0, 'bad_words_ids': None, 'num_return_sequences': 1, 'chunk_size_feed_forward': 0, 'output_scores': False, 'return_dict_in_generate': False, 'forced_bos_token_id': None, 'forced_eos_token_id': None, 'remove_invalid_values': False, 'architectures': ['SpeechEncoderDecoderModel'], 'finetuning_task': None, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'tokenizer_class': None, 'prefix': None, 'bos_token_id': None, 'pad_token_id': 1, 'eos_token_id': 2, 'sep_token_id': None, 'decoder_start_token_id': 0, 'task_specific_params': None, 'problem_type': None, '_name_or_path': './', 'transformers_version': None, 'decoder': {'vocab_size': 50265, 'max_position_embeddings': 1024, 'd_model': 1024, 'encoder_ffn_dim': 4096, 'encoder_layers': 12, 'encoder_attention_heads': 16, 'decoder_ffn_dim': 4096, 'decoder_layers': 12, 'decoder_attention_heads': 16, 'dropout': 0.1, 'attention_dropout': 0.1, 'activation_dropout': 0.1, 'activation_function': 'gelu', 'init_std': 0.02, 'encoder_layerdrop': 0.0, 'decoder_layerdrop': 0.0, 'classifier_dropout': 0.0, 'use_cache': False, 'num_hidden_layers': 12, 'scale_embedding': False, 'return_dict': True, 'output_hidden_states': False, 'output_attentions': False, 'torchscript': False, 'torch_dtype': 'float32', 'use_bfloat16': False, 'pruned_heads': {}, 'tie_word_embeddings': True, 'is_encoder_decoder': False, 'is_decoder': True, 'cross_attention_hidden_size': None, 'add_cross_attention': True, 'tie_encoder_decoder': False, 'max_length': 20, 'min_length': 0, 'do_sample': False, 'early_stopping': True, 'num_beams': 4, 'num_beam_groups': 1, 'diversity_penalty': 0.0, 'temperature': 1.0, 'top_k': 50, 'top_p': 1.0, 'repetition_penalty': 1.0, 'length_penalty': 1.0, 'no_repeat_ngram_size': 3, 'encoder_no_repeat_ngram_size': 0, 'bad_words_ids': None, 'num_return_sequences': 1, 'chunk_size_feed_forward': 0, 'output_scores': False, 'return_dict_in_generate': False, 'forced_bos_token_id': 0, 'forced_eos_token_id': 2, 'remove_invalid_values': False, 'architectures': ['BartForCausalLM'], 'finetuning_task': None, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1', 2: 'LABEL_2'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1, 'LABEL_2': 2}, 'tokenizer_class': None, 'prefix': None, 'bos_token_id': 0, 'pad_token_id': 1, 'eos_token_id': 2, 'sep_token_id': None, 'decoder_start_token_id': 2, 'task_specific_params': {'summarization': {'length_penalty': 1.0, 'max_length': 128, 'min_length': 12, 'num_beams': 4}, 'summarization_cnn': {'length_penalty': 2.0, 'max_length': 142, 'min_length': 56, 'num_beams': 4}, 'summarization_xsum': {'length_penalty': 1.0, 'max_length': 62, 'min_length': 11, 'num_beams': 6}}, 'problem_type': None, '_name_or_path': 'decoder', 'transformers_version': '4.17.0.dev0', 'add_bias_logits': False, 'add_final_layer_norm': False, 'classif_dropout': 0.1, 'gradient_checkpointing': False, 'normalize_before': False, 'model_type': 'bart'}, 'encoder': {'return_dict': True, 'output_hidden_states': False, 'output_attentions': False, 'torchscript': False, 'torch_dtype': None, 'use_bfloat16': False, 'pruned_heads': {}, 'tie_word_embeddings': True, 'is_encoder_decoder': False, 'is_decoder': False, 'cross_attention_hidden_size': None, 'add_cross_attention': False, 'tie_encoder_decoder': False, 'max_length': 20, 'min_length': 0, 'do_sample': False, 'early_stopping': False, 'num_beams': 1, 'num_beam_groups': 1, 'diversity_penalty': 0.0, 'temperature': 1.0, 'top_k': 50, 'top_p': 1.0, 'repetition_penalty': 1.0, 'length_penalty': 1.0, 'no_repeat_ngram_size': 0, 'encoder_no_repeat_ngram_size': 0, 'bad_words_ids': None, 'num_return_sequences': 1, 'chunk_size_feed_forward': 0, 'output_scores': False, 'return_dict_in_generate': False, 'forced_bos_token_id': None, 'forced_eos_token_id': None, 'remove_invalid_values': False, 'architectures': ['Wav2Vec2ForPreTraining'], 'finetuning_task': None, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'tokenizer_class': None, 'prefix': None, 'bos_token_id': 1, 'pad_token_id': 0, 'eos_token_id': 2, 'sep_token_id': None, 'decoder_start_token_id': None, 'task_specific_params': None, 'problem_type': None, '_name_or_path': 'facebook/wav2vec2-large-lv60', 'transformers_version': '4.17.0.dev0', 'feat_extract_dropout': 0.0, 'gradient_checkpointing': False, 'hidden_dropout_prob': 0.1, 'num_feat_extract_layers': 7, 'hidden_size': 1024, 'feat_extract_norm': 'layer', 'feat_extract_activation': 'gelu', 'conv_dim': [512, 512, 512, 512, 512, 512, 512], 'conv_stride': [5, 2, 2, 2, 2, 2, 2], 'conv_kernel': [10, 3, 3, 3, 3, 2, 2], 'conv_bias': True, 'num_conv_pos_embeddings': 128, 'num_conv_pos_embedding_groups': 16, 'num_hidden_layers': 24, 'intermediate_size': 4096, 'hidden_act': 'gelu', 'num_attention_heads': 16, 'hidden_dropout': 0.1, 'attention_dropout': 0.1, 'activation_dropout': 0.1, 'feat_proj_dropout': 0.0, 'final_dropout': 0.0, 'layerdrop': 0.0, 'layer_norm_eps': 1e-05, 'initializer_range': 0.02, 'vocab_size': 32, 'do_stable_layer_norm': True, 'use_weighted_layer_sum': False, 'apply_spec_augment': True, 'mask_time_prob': 0.1, 'mask_time_length': 10, 'mask_time_min_masks': 2, 'mask_feature_prob': 0.0, 'mask_feature_length': 10, 'mask_feature_min_masks': 0, 'num_codevectors_per_group': 320, 'num_codevector_groups': 2, 'contrastive_logits_temperature': 0.1, 'feat_quantizer_dropout': 0.0, 'num_negatives': 100, 'codevector_dim': 768, 'proj_codevector_dim': 768, 'diversity_loss_weight': 0.1, 'ctc_loss_reduction': 'sum', 'ctc_zero_infinity': False, 'add_adapter': False, 'adapter_kernel_size': 3, 'adapter_stride': 2, 'num_adapter_layers': 3, 'output_hidden_size': 1024, 'classifier_proj_size': 256, 'tdnn_dim': [512, 512, 512, 512, 1500], 'tdnn_kernel': [5, 3, 3, 1, 1], 'tdnn_dilation': [1, 2, 3, 1, 1], 'xvector_output_dim': 512, 'model_type': 'wav2vec2'}, 'model_type': 'speech-encoder-decoder', 'processor_class': 'Wav2Vec2Processor', 'use_cache': False, 'output_dir': './', 'overwrite_output_dir': True, 'do_train': True, 'do_eval': True, 'do_predict': False, 'evaluation_strategy': 'steps', 'prediction_loss_only': False, 'per_device_train_batch_size': 8, 'per_device_eval_batch_size': 8, 'per_gpu_train_batch_size': 'None', 'per_gpu_eval_batch_size': 'None', 'gradient_accumulation_steps': 4, 'eval_accumulation_steps': 'None', 'learning_rate': 3e-05, 'weight_decay': 0.0, 'adam_beta1': 0.9, 'adam_beta2': 0.999, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 25.0, 'max_steps': -1, 'lr_scheduler_type': 'linear', 'warmup_ratio': 0.0, 'warmup_steps': 1000, 'log_level': -1, 'log_level_replica': -1, 'log_on_each_node': True, 'logging_dir': './runs/Mar23_16-01-22_sanchit--v100', 'logging_strategy': 'steps', 'logging_first_step': False, 'logging_steps': 1, 'logging_nan_inf_filter': True, 'save_strategy': 'steps', 'save_steps': 1500, 'save_total_limit': 1, 'save_on_each_node': False, 'no_cuda': False, 'seed': 42, 'bf16': False, 'fp16': True, 'fp16_opt_level': 'O1', 'half_precision_backend': 'amp', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': 'None', 'local_rank': -1, 'xpu_backend': 'None', 'tpu_num_cores': 'None', 'tpu_metrics_debug': False, 'debug': '[]', 'dataloader_drop_last': False, 'eval_steps': 1500, 'dataloader_num_workers': 0, 'past_index': -1, 'run_name': './', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': 'None', 'load_best_model_at_end': False, 'metric_for_best_model': 'None', 'greater_is_better': 'None', 'ignore_data_skip': False, 'sharded_ddp': '[]', 'deepspeed': 'None', 'label_smoothing_factor': 0.0, 'optim': 'adamw_hf', 'adafactor': False, 'group_by_length': True, 'length_column_name': 'input_length', 'report_to': "['tensorboard', 'wandb']", 'ddp_find_unused_parameters': 'None', 'ddp_bucket_cap_mb': 'None', 'dataloader_pin_memory': True, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': True, 'resume_from_checkpoint': 'None', 'hub_model_id': 'None', 'hub_strategy': 'every_save', 'hub_token': '<HUB_TOKEN>', 'gradient_checkpointing': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': 'None', 'push_to_hub_organization': 'None', 'push_to_hub_token': '<PUSH_TO_HUB_TOKEN>', '_n_gpu': 1, 'mp_parameters': '', 'sortish_sampler': False, 'predict_with_generate': True, 'generation_max_length': 40, 'generation_num_beams': 1, 'train_batch_size': 8, 'eval_batch_size': 8}
+2022-03-23 16:01:48,003 INFO    MainThread:1748630 [wandb_watch.py:watch():43] Watching
+2022-03-23 16:01:52,781 INFO    MainThread:1748630 [wandb_run.py:_atexit_cleanup():1797] got exitcode: 255
+2022-03-23 16:01:52,782 INFO    MainThread:1748630 [wandb_run.py:_restore():1769] restore

wandb/run-20220323_161301-3mj7aa54/files/requirements.txt

@@ -0,0 +1,186 @@
+absl-py==1.0.0
+aiohttp==3.8.1
+aiosignal==1.2.0
+anyio==3.5.0
+appdirs==1.4.4
+argon2-cffi-bindings==21.2.0
+argon2-cffi==21.3.0
+asttokens==2.0.5
+async-timeout==4.0.2
+attrs==21.4.0
+audioread==2.1.9
+babel==2.9.1
+backcall==0.2.0
+bitsandbytes-cuda113==0.26.0
+black==22.1.0
+bleach==4.1.0
+cachetools==5.0.0
+certifi==2021.10.8
+cffi==1.15.0
+charset-normalizer==2.0.11
+chex==0.1.0
+click==8.0.3
+clldutils==3.10.1
+colorlog==6.6.0
+csvw==1.11.0
+cycler==0.11.0
+datasets==1.18.3
+debugpy==1.5.1
+decorator==5.1.1
+defusedxml==0.7.1
+dill==0.3.4
+dlinfo==1.2.1
+dm-tree==0.1.6
+docker-pycreds==0.4.0
+entrypoints==0.4
+executing==0.8.2
+filelock==3.4.2
+flatbuffers==2.0
+flax==0.4.0
+fonttools==4.29.1
+frozenlist==1.3.0
+fsspec==2022.1.0
+gitdb==4.0.9
+gitpython==3.1.27
+google-auth-oauthlib==0.4.6
+google-auth==2.6.0
+grpcio==1.43.0
+huggingface-hub==0.4.0
+hypothesis==6.36.1
+idna==3.3
+importlib-metadata==4.10.1
+ipdb==0.13.9
+ipykernel==6.8.0
+ipython-genutils==0.2.0
+ipython==8.0.1
+ipywidgets==7.6.5
+isodate==0.6.1
+jax==0.2.28
+jaxlib==0.1.76+cuda11.cudnn82
+jedi==0.18.1
+jinja2==3.0.3
+jiwer==2.3.0
+joblib==1.1.0
+json5==0.9.6
+jsonschema==4.4.0
+jupyter-client==7.1.2
+jupyter-console==6.4.0
+jupyter-core==4.9.1
+jupyter-server==1.13.5
+jupyter==1.0.0
+jupyterlab-pygments==0.1.2
+jupyterlab-server==2.10.3
+jupyterlab-widgets==1.0.2
+jupyterlab==3.2.9
+kiwisolver==1.3.2
+librosa==0.8.1
+llvmlite==0.38.0
+markdown==3.3.6
+markupsafe==2.0.1
+matplotlib-inline==0.1.3
+matplotlib==3.5.1
+mistune==0.8.4
+msgpack==1.0.3
+multidict==6.0.2
+multiprocess==0.70.12.2
+mypy-extensions==0.4.3
+nbclassic==0.3.5
+nbclient==0.5.10
+nbconvert==6.4.1
+nbformat==5.1.3
+nest-asyncio==1.5.4
+notebook==6.4.8
+numba==0.55.1
+numpy==1.21.5
+oauthlib==3.2.0
+opt-einsum==3.3.0
+optax==0.1.0
+packaging==21.3
+pandas==1.4.0
+pandocfilters==1.5.0
+parso==0.8.3
+pathspec==0.9.0
+pathtools==0.1.2
+pexpect==4.8.0
+phonemizer==3.0.1
+pickleshare==0.7.5
+pillow==9.0.0
+pip==22.0.2
+pkg-resources==0.0.0
+platformdirs==2.4.1
+pooch==1.6.0
+prometheus-client==0.13.1
+promise==2.3
+prompt-toolkit==3.0.26
+protobuf==3.19.4
+psutil==5.9.0
+ptyprocess==0.7.0
+pure-eval==0.2.2
+pyarrow==6.0.1
+pyasn1-modules==0.2.8
+pyasn1==0.4.8
+pycparser==2.21
+pyctcdecode==0.3.0
+pygments==2.11.2
+pygtrie==2.4.2
+pyparsing==3.0.7
+pyrsistent==0.18.1
+python-dateutil==2.8.2
+python-levenshtein==0.12.2
+pytz==2021.3
+pyyaml==6.0
+pyzmq==22.3.0
+qtconsole==5.2.2
+qtpy==2.0.1
+regex==2022.1.18
+requests-oauthlib==1.3.1
+requests==2.27.1
+resampy==0.2.2
+rfc3986==2.0.0
+rsa==4.8
+sacremoses==0.0.47
+scikit-learn==1.0.2
+scipy==1.7.3
+segments==2.2.0
+send2trash==1.8.0
+sentry-sdk==1.5.6
+setuptools==44.1.1
+shortuuid==1.0.8
+six==1.16.0
+smmap==5.0.0
+sniffio==1.2.0
+sortedcontainers==2.4.0
+soundfile==0.10.3.post1
+stack-data==0.1.4
+tabulate==0.8.9
+tensorboard-data-server==0.6.1
+tensorboard-plugin-wit==1.8.1
+tensorboard==2.8.0
+termcolor==1.1.0
+terminado==0.13.1
+testpath==0.5.0
+threadpoolctl==3.1.0
+tokenizers==0.11.4
+toml==0.10.2
+tomli==2.0.0
+toolz==0.11.2
+torch==1.10.2+cu113
+torchaudio==0.10.2+cu113
+tornado==6.1
+tqdm==4.62.3
+traitlets==5.1.1
+transformers==4.17.0.dev0
+typing-extensions==3.10.0.2
+uritemplate==4.1.1
+urllib3==1.26.8
+wandb==0.12.10
+wcwidth==0.2.5
+webencodings==0.5.1
+websocket-client==1.2.3
+werkzeug==2.0.2
+wheel==0.37.1
+widgetsnbextension==3.5.2
+xxhash==2.0.2
+yarl==1.7.2
+yaspin==2.1.0
+zipp==3.7.0

wandb/run-20220323_161301-3mj7aa54/files/wandb-metadata.json

@@ -0,0 +1,58 @@
+{
+    "os": "Linux-5.11.0-1028-gcp-x86_64-with-glibc2.33",
+    "python": "3.9.5",
+    "heartbeatAt": "2022-03-23T16:13:03.264023",
+    "startedAt": "2022-03-23T16:13:01.531554",
+    "docker": null,
+    "gpu": "Tesla V100-SXM2-16GB",
+    "gpu_count": 2,
+    "cpu_count": 16,
+    "cuda": null,
+    "args": [
+        "--dataset_name=librispeech_asr",
+        "--model_name_or_path=./",
+        "--tokenizer_name=./",
+        "--dataset_config_name=clean",
+        "--train_split_name=train.100",
+        "--eval_split_name=validation",
+        "--output_dir=./",
+        "--preprocessing_num_workers=1",
+        "--length_column_name=input_length",
+        "--num_train_epochs=25",
+        "--per_device_train_batch_size=8",
+        "--per_device_eval_batch_size=8",
+        "--gradient_accumulation_steps=4",
+        "--generation_max_length=40",
+        "--generation_num_beams=1",
+        "--learning_rate=3e-5",
+        "--warmup_steps=1000",
+        "--evaluation_strategy=steps",
+        "--text_column_name=text",
+        "--save_steps=1500",
+        "--eval_steps=1500",
+        "--logging_steps=1",
+        "--save_total_limit=1",
+        "--freeze_feature_encoder",
+        "--gradient_checkpointing",
+        "--fp16",
+        "--group_by_length",
+        "--predict_with_generate",
+        "--do_lower_case",
+        "--do_eval",
+        "--do_train",
+        "--push_to_hub",
+        "--use_auth_token"
+    ],
+    "state": "running",
+    "program": "/home/sanchit_huggingface_co/wav2vec2-2-rnd-no-adapter-regularisation/run_speech_recognition_seq2seq.py",
+    "codePath": "run_speech_recognition_seq2seq.py",
+    "git": {
+        "remote": "https://huggingface.co/sanchit-gandhi/wav2vec2-2-rnd-no-adapter-regularisation",
+        "commit": "156c71831efc84fc27b6a9a52f81872c0138b990"
+    },
+    "email": "sanchit@huggingface.co",
+    "root": "/home/sanchit_huggingface_co/wav2vec2-2-rnd-no-adapter-regularisation",
+    "host": "sanchit--v100",
+    "username": "sanchit_huggingface_co",
+    "executable": "/home/sanchit_huggingface_co/gcp/bin/python"
+}

wandb/run-20220323_161301-3mj7aa54/logs/debug.log

@@ -0,0 +1,29 @@
+2022-03-23 16:13:01,533 INFO    MainThread:1748915 [wandb_setup.py:_flush():75] Loading settings from /home/sanchit_huggingface_co/.config/wandb/settings
+2022-03-23 16:13:01,533 INFO    MainThread:1748915 [wandb_setup.py:_flush():75] Loading settings from /home/sanchit_huggingface_co/wav2vec2-2-rnd-no-adapter-regularisation/wandb/settings
+2022-03-23 16:13:01,533 INFO    MainThread:1748915 [wandb_setup.py:_flush():75] Loading settings from environment variables: {}
+2022-03-23 16:13:01,533 INFO    MainThread:1748915 [wandb_setup.py:_flush():75] Inferring run settings from compute environment: {'program_relpath': 'run_speech_recognition_seq2seq.py', 'program': '/home/sanchit_huggingface_co/wav2vec2-2-rnd-no-adapter-regularisation/run_speech_recognition_seq2seq.py'}
+2022-03-23 16:13:01,533 INFO    MainThread:1748915 [wandb_init.py:_log_setup():386] Logging user logs to /home/sanchit_huggingface_co/wav2vec2-2-rnd-no-adapter-regularisation/wandb/run-20220323_161301-3mj7aa54/logs/debug.log
+2022-03-23 16:13:01,533 INFO    MainThread:1748915 [wandb_init.py:_log_setup():387] Logging internal logs to /home/sanchit_huggingface_co/wav2vec2-2-rnd-no-adapter-regularisation/wandb/run-20220323_161301-3mj7aa54/logs/debug-internal.log
+2022-03-23 16:13:01,533 INFO    MainThread:1748915 [wandb_init.py:init():420] calling init triggers
+2022-03-23 16:13:01,533 INFO    MainThread:1748915 [wandb_init.py:init():425] wandb.init called with sweep_config: {}
+config: {}
+2022-03-23 16:13:01,533 INFO    MainThread:1748915 [wandb_init.py:init():471] starting backend
+2022-03-23 16:13:01,533 INFO    MainThread:1748915 [backend.py:_multiprocessing_setup():99] multiprocessing start_methods=fork,spawn,forkserver, using: spawn
+2022-03-23 16:13:01,587 INFO    MainThread:1748915 [backend.py:ensure_launched():219] starting backend process...
+2022-03-23 16:13:01,641 INFO    MainThread:1748915 [backend.py:ensure_launched():224] started backend process with pid: 1749018
+2022-03-23 16:13:01,644 INFO    MainThread:1748915 [wandb_init.py:init():480] backend started and connected
+2022-03-23 16:13:01,655 INFO    MainThread:1748915 [wandb_init.py:init():550] updated telemetry
+2022-03-23 16:13:01,784 INFO    MainThread:1748915 [wandb_init.py:init():581] communicating current version
+2022-03-23 16:13:02,463 INFO    MainThread:1748915 [wandb_init.py:init():586] got version response upgrade_message: "wandb version 0.12.11 is available!  To upgrade, please run:\n $ pip install wandb --upgrade"
+
+2022-03-23 16:13:02,463 INFO    MainThread:1748915 [wandb_init.py:init():596] communicating run to backend with 30 second timeout
+2022-03-23 16:13:03,259 INFO    MainThread:1748915 [wandb_init.py:init():624] starting run threads in backend
+2022-03-23 16:13:03,361 INFO    MainThread:1748915 [wandb_run.py:_console_start():1827] atexit reg
+2022-03-23 16:13:03,361 INFO    MainThread:1748915 [wandb_run.py:_redirect():1701] redirect: SettingsConsole.REDIRECT
+2022-03-23 16:13:03,362 INFO    MainThread:1748915 [wandb_run.py:_redirect():1706] Redirecting console.
+2022-03-23 16:13:03,363 INFO    MainThread:1748915 [wandb_run.py:_redirect():1762] Redirects installed.
+2022-03-23 16:13:03,363 INFO    MainThread:1748915 [wandb_init.py:init():651] run started, returning control to user process
+2022-03-23 16:13:03,366 INFO    MainThread:1748915 [wandb_run.py:_config_callback():966] config_cb None None {'return_dict': True, 'output_hidden_states': False, 'output_attentions': False, 'torchscript': False, 'torch_dtype': 'torch.float32', 'use_bfloat16': False, 'pruned_heads': {}, 'tie_word_embeddings': False, 'is_encoder_decoder': True, 'is_decoder': False, 'cross_attention_hidden_size': None, 'add_cross_attention': False, 'tie_encoder_decoder': False, 'max_length': 50, 'min_length': 0, 'do_sample': False, 'early_stopping': False, 'num_beams': 1, 'num_beam_groups': 1, 'diversity_penalty': 0.0, 'temperature': 1.0, 'top_k': 50, 'top_p': 1.0, 'repetition_penalty': 1.0, 'length_penalty': 1.0, 'no_repeat_ngram_size': 0, 'encoder_no_repeat_ngram_size': 0, 'bad_words_ids': None, 'num_return_sequences': 1, 'chunk_size_feed_forward': 0, 'output_scores': False, 'return_dict_in_generate': False, 'forced_bos_token_id': None, 'forced_eos_token_id': None, 'remove_invalid_values': False, 'architectures': ['SpeechEncoderDecoderModel'], 'finetuning_task': None, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'tokenizer_class': None, 'prefix': None, 'bos_token_id': None, 'pad_token_id': 1, 'eos_token_id': 2, 'sep_token_id': None, 'decoder_start_token_id': 0, 'task_specific_params': None, 'problem_type': None, '_name_or_path': './', 'transformers_version': None, 'decoder': {'vocab_size': 50265, 'max_position_embeddings': 1024, 'd_model': 1024, 'encoder_ffn_dim': 4096, 'encoder_layers': 12, 'encoder_attention_heads': 16, 'decoder_ffn_dim': 4096, 'decoder_layers': 12, 'decoder_attention_heads': 16, 'dropout': 0.1, 'attention_dropout': 0.1, 'activation_dropout': 0.1, 'activation_function': 'gelu', 'init_std': 0.02, 'encoder_layerdrop': 0.0, 'decoder_layerdrop': 0.0, 'classifier_dropout': 0.0, 'use_cache': False, 'num_hidden_layers': 12, 'scale_embedding': False, 'return_dict': True, 'output_hidden_states': False, 'output_attentions': False, 'torchscript': False, 'torch_dtype': 'float32', 'use_bfloat16': False, 'pruned_heads': {}, 'tie_word_embeddings': True, 'is_encoder_decoder': False, 'is_decoder': True, 'cross_attention_hidden_size': None, 'add_cross_attention': True, 'tie_encoder_decoder': False, 'max_length': 20, 'min_length': 0, 'do_sample': False, 'early_stopping': True, 'num_beams': 4, 'num_beam_groups': 1, 'diversity_penalty': 0.0, 'temperature': 1.0, 'top_k': 50, 'top_p': 1.0, 'repetition_penalty': 1.0, 'length_penalty': 1.0, 'no_repeat_ngram_size': 3, 'encoder_no_repeat_ngram_size': 0, 'bad_words_ids': None, 'num_return_sequences': 1, 'chunk_size_feed_forward': 0, 'output_scores': False, 'return_dict_in_generate': False, 'forced_bos_token_id': 0, 'forced_eos_token_id': 2, 'remove_invalid_values': False, 'architectures': ['BartForCausalLM'], 'finetuning_task': None, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1', 2: 'LABEL_2'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1, 'LABEL_2': 2}, 'tokenizer_class': None, 'prefix': None, 'bos_token_id': 0, 'pad_token_id': 1, 'eos_token_id': 2, 'sep_token_id': None, 'decoder_start_token_id': 2, 'task_specific_params': {'summarization': {'length_penalty': 1.0, 'max_length': 128, 'min_length': 12, 'num_beams': 4}, 'summarization_cnn': {'length_penalty': 2.0, 'max_length': 142, 'min_length': 56, 'num_beams': 4}, 'summarization_xsum': {'length_penalty': 1.0, 'max_length': 62, 'min_length': 11, 'num_beams': 6}}, 'problem_type': None, '_name_or_path': 'decoder', 'transformers_version': '4.17.0.dev0', 'add_bias_logits': False, 'add_final_layer_norm': False, 'classif_dropout': 0.1, 'gradient_checkpointing': False, 'normalize_before': False, 'model_type': 'bart'}, 'encoder': {'return_dict': True, 'output_hidden_states': False, 'output_attentions': False, 'torchscript': False, 'torch_dtype': None, 'use_bfloat16': False, 'pruned_heads': {}, 'tie_word_embeddings': True, 'is_encoder_decoder': False, 'is_decoder': False, 'cross_attention_hidden_size': None, 'add_cross_attention': False, 'tie_encoder_decoder': False, 'max_length': 20, 'min_length': 0, 'do_sample': False, 'early_stopping': False, 'num_beams': 1, 'num_beam_groups': 1, 'diversity_penalty': 0.0, 'temperature': 1.0, 'top_k': 50, 'top_p': 1.0, 'repetition_penalty': 1.0, 'length_penalty': 1.0, 'no_repeat_ngram_size': 0, 'encoder_no_repeat_ngram_size': 0, 'bad_words_ids': None, 'num_return_sequences': 1, 'chunk_size_feed_forward': 0, 'output_scores': False, 'return_dict_in_generate': False, 'forced_bos_token_id': None, 'forced_eos_token_id': None, 'remove_invalid_values': False, 'architectures': ['Wav2Vec2ForPreTraining'], 'finetuning_task': None, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'tokenizer_class': None, 'prefix': None, 'bos_token_id': 1, 'pad_token_id': 0, 'eos_token_id': 2, 'sep_token_id': None, 'decoder_start_token_id': None, 'task_specific_params': None, 'problem_type': None, '_name_or_path': 'facebook/wav2vec2-large-lv60', 'transformers_version': '4.17.0.dev0', 'feat_extract_dropout': 0.0, 'gradient_checkpointing': False, 'hidden_dropout_prob': 0.1, 'num_feat_extract_layers': 7, 'hidden_size': 1024, 'feat_extract_norm': 'layer', 'feat_extract_activation': 'gelu', 'conv_dim': [512, 512, 512, 512, 512, 512, 512], 'conv_stride': [5, 2, 2, 2, 2, 2, 2], 'conv_kernel': [10, 3, 3, 3, 3, 2, 2], 'conv_bias': True, 'num_conv_pos_embeddings': 128, 'num_conv_pos_embedding_groups': 16, 'num_hidden_layers': 24, 'intermediate_size': 4096, 'hidden_act': 'gelu', 'num_attention_heads': 16, 'hidden_dropout': 0.1, 'attention_dropout': 0.1, 'activation_dropout': 0.1, 'feat_proj_dropout': 0.0, 'final_dropout': 0.0, 'layerdrop': 0.0, 'layer_norm_eps': 1e-05, 'initializer_range': 0.02, 'vocab_size': 32, 'do_stable_layer_norm': True, 'use_weighted_layer_sum': False, 'apply_spec_augment': True, 'mask_time_prob': 0.1, 'mask_time_length': 10, 'mask_time_min_masks': 2, 'mask_feature_prob': 0.0, 'mask_feature_length': 10, 'mask_feature_min_masks': 0, 'num_codevectors_per_group': 320, 'num_codevector_groups': 2, 'contrastive_logits_temperature': 0.1, 'feat_quantizer_dropout': 0.0, 'num_negatives': 100, 'codevector_dim': 768, 'proj_codevector_dim': 768, 'diversity_loss_weight': 0.1, 'ctc_loss_reduction': 'sum', 'ctc_zero_infinity': False, 'add_adapter': False, 'adapter_kernel_size': 3, 'adapter_stride': 2, 'num_adapter_layers': 3, 'output_hidden_size': 1024, 'classifier_proj_size': 256, 'tdnn_dim': [512, 512, 512, 512, 1500], 'tdnn_kernel': [5, 3, 3, 1, 1], 'tdnn_dilation': [1, 2, 3, 1, 1], 'xvector_output_dim': 512, 'model_type': 'wav2vec2'}, 'model_type': 'speech-encoder-decoder', 'processor_class': 'Wav2Vec2Processor', 'use_cache': False, 'output_dir': './', 'overwrite_output_dir': False, 'do_train': True, 'do_eval': True, 'do_predict': False, 'evaluation_strategy': 'steps', 'prediction_loss_only': False, 'per_device_train_batch_size': 8, 'per_device_eval_batch_size': 8, 'per_gpu_train_batch_size': 'None', 'per_gpu_eval_batch_size': 'None', 'gradient_accumulation_steps': 4, 'eval_accumulation_steps': 'None', 'learning_rate': 3e-05, 'weight_decay': 0.0, 'adam_beta1': 0.9, 'adam_beta2': 0.999, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 25.0, 'max_steps': -1, 'lr_scheduler_type': 'linear', 'warmup_ratio': 0.0, 'warmup_steps': 1000, 'log_level': -1, 'log_level_replica': -1, 'log_on_each_node': True, 'logging_dir': './runs/Mar23_16-12-34_sanchit--v100', 'logging_strategy': 'steps', 'logging_first_step': False, 'logging_steps': 1, 'logging_nan_inf_filter': True, 'save_strategy': 'steps', 'save_steps': 1500, 'save_total_limit': 1, 'save_on_each_node': False, 'no_cuda': False, 'seed': 42, 'bf16': False, 'fp16': True, 'fp16_opt_level': 'O1', 'half_precision_backend': 'amp', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': 'None', 'local_rank': -1, 'xpu_backend': 'None', 'tpu_num_cores': 'None', 'tpu_metrics_debug': False, 'debug': '[]', 'dataloader_drop_last': False, 'eval_steps': 1500, 'dataloader_num_workers': 0, 'past_index': -1, 'run_name': './', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': 'None', 'load_best_model_at_end': False, 'metric_for_best_model': 'None', 'greater_is_better': 'None', 'ignore_data_skip': False, 'sharded_ddp': '[]', 'deepspeed': 'None', 'label_smoothing_factor': 0.0, 'optim': 'adamw_hf', 'adafactor': False, 'group_by_length': True, 'length_column_name': 'input_length', 'report_to': "['tensorboard', 'wandb']", 'ddp_find_unused_parameters': 'None', 'ddp_bucket_cap_mb': 'None', 'dataloader_pin_memory': True, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': True, 'resume_from_checkpoint': 'None', 'hub_model_id': 'None', 'hub_strategy': 'every_save', 'hub_token': '<HUB_TOKEN>', 'gradient_checkpointing': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': 'None', 'push_to_hub_organization': 'None', 'push_to_hub_token': '<PUSH_TO_HUB_TOKEN>', '_n_gpu': 1, 'mp_parameters': '', 'sortish_sampler': False, 'predict_with_generate': True, 'generation_max_length': 40, 'generation_num_beams': 1, 'train_batch_size': 8, 'eval_batch_size': 8}
+2022-03-23 16:13:03,371 INFO    MainThread:1748915 [wandb_watch.py:watch():43] Watching
+2022-03-23 16:57:48,093 INFO    MainThread:1748915 [wandb_run.py:_atexit_cleanup():1797] got exitcode: 255
+2022-03-23 16:57:48,095 INFO    MainThread:1748915 [wandb_run.py:_restore():1769] restore

wandb/run-20220323_161301-3mj7aa54/run-3mj7aa54.wandb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eb4b833be4b55ca0a3b1187c6c6e147da974572275ca09505fa1912cf7ad446a
+size 13841952

wandb/run-20220324_170220-1u8cita3/files/requirements.txt

@@ -0,0 +1,186 @@
+absl-py==1.0.0
+aiohttp==3.8.1
+aiosignal==1.2.0
+anyio==3.5.0
+appdirs==1.4.4
+argon2-cffi-bindings==21.2.0
+argon2-cffi==21.3.0
+asttokens==2.0.5
+async-timeout==4.0.2
+attrs==21.4.0
+audioread==2.1.9
+babel==2.9.1
+backcall==0.2.0
+bitsandbytes-cuda113==0.26.0
+black==22.1.0
+bleach==4.1.0
+cachetools==5.0.0
+certifi==2021.10.8
+cffi==1.15.0
+charset-normalizer==2.0.11
+chex==0.1.0
+click==8.0.3
+clldutils==3.10.1
+colorlog==6.6.0
+csvw==1.11.0
+cycler==0.11.0
+datasets==1.18.3
+debugpy==1.5.1
+decorator==5.1.1
+defusedxml==0.7.1
+dill==0.3.4
+dlinfo==1.2.1
+dm-tree==0.1.6
+docker-pycreds==0.4.0
+entrypoints==0.4
+executing==0.8.2
+filelock==3.4.2
+flatbuffers==2.0
+flax==0.4.0
+fonttools==4.29.1
+frozenlist==1.3.0
+fsspec==2022.1.0
+gitdb==4.0.9
+gitpython==3.1.27
+google-auth-oauthlib==0.4.6
+google-auth==2.6.0
+grpcio==1.43.0
+huggingface-hub==0.4.0
+hypothesis==6.36.1
+idna==3.3
+importlib-metadata==4.10.1
+ipdb==0.13.9
+ipykernel==6.8.0
+ipython-genutils==0.2.0
+ipython==8.0.1
+ipywidgets==7.6.5
+isodate==0.6.1
+jax==0.2.28
+jaxlib==0.1.76+cuda11.cudnn82
+jedi==0.18.1
+jinja2==3.0.3
+jiwer==2.3.0
+joblib==1.1.0
+json5==0.9.6
+jsonschema==4.4.0
+jupyter-client==7.1.2
+jupyter-console==6.4.0
+jupyter-core==4.9.1
+jupyter-server==1.13.5
+jupyter==1.0.0
+jupyterlab-pygments==0.1.2
+jupyterlab-server==2.10.3
+jupyterlab-widgets==1.0.2
+jupyterlab==3.2.9
+kiwisolver==1.3.2
+librosa==0.8.1
+llvmlite==0.38.0
+markdown==3.3.6
+markupsafe==2.0.1
+matplotlib-inline==0.1.3
+matplotlib==3.5.1
+mistune==0.8.4
+msgpack==1.0.3
+multidict==6.0.2
+multiprocess==0.70.12.2
+mypy-extensions==0.4.3
+nbclassic==0.3.5
+nbclient==0.5.10
+nbconvert==6.4.1
+nbformat==5.1.3
+nest-asyncio==1.5.4
+notebook==6.4.8
+numba==0.55.1
+numpy==1.21.5
+oauthlib==3.2.0
+opt-einsum==3.3.0
+optax==0.1.0
+packaging==21.3
+pandas==1.4.0
+pandocfilters==1.5.0
+parso==0.8.3
+pathspec==0.9.0
+pathtools==0.1.2
+pexpect==4.8.0
+phonemizer==3.0.1
+pickleshare==0.7.5
+pillow==9.0.0
+pip==22.0.2
+pkg-resources==0.0.0
+platformdirs==2.4.1
+pooch==1.6.0
+prometheus-client==0.13.1
+promise==2.3
+prompt-toolkit==3.0.26
+protobuf==3.19.4
+psutil==5.9.0
+ptyprocess==0.7.0
+pure-eval==0.2.2
+pyarrow==6.0.1
+pyasn1-modules==0.2.8
+pyasn1==0.4.8
+pycparser==2.21
+pyctcdecode==0.3.0
+pygments==2.11.2
+pygtrie==2.4.2
+pyparsing==3.0.7
+pyrsistent==0.18.1
+python-dateutil==2.8.2
+python-levenshtein==0.12.2
+pytz==2021.3
+pyyaml==6.0
+pyzmq==22.3.0
+qtconsole==5.2.2
+qtpy==2.0.1
+regex==2022.1.18
+requests-oauthlib==1.3.1
+requests==2.27.1
+resampy==0.2.2
+rfc3986==2.0.0
+rsa==4.8
+sacremoses==0.0.47
+scikit-learn==1.0.2
+scipy==1.7.3
+segments==2.2.0
+send2trash==1.8.0
+sentry-sdk==1.5.6
+setuptools==44.1.1
+shortuuid==1.0.8
+six==1.16.0
+smmap==5.0.0
+sniffio==1.2.0
+sortedcontainers==2.4.0
+soundfile==0.10.3.post1
+stack-data==0.1.4
+tabulate==0.8.9
+tensorboard-data-server==0.6.1
+tensorboard-plugin-wit==1.8.1
+tensorboard==2.8.0
+termcolor==1.1.0
+terminado==0.13.1
+testpath==0.5.0
+threadpoolctl==3.1.0
+tokenizers==0.11.4
+toml==0.10.2
+tomli==2.0.0
+toolz==0.11.2
+torch==1.10.2+cu113
+torchaudio==0.10.2+cu113
+tornado==6.1
+tqdm==4.62.3
+traitlets==5.1.1
+transformers==4.17.0.dev0
+typing-extensions==3.10.0.2
+uritemplate==4.1.1
+urllib3==1.26.8
+wandb==0.12.10
+wcwidth==0.2.5
+webencodings==0.5.1
+websocket-client==1.2.3
+werkzeug==2.0.2
+wheel==0.37.1
+widgetsnbextension==3.5.2
+xxhash==2.0.2
+yarl==1.7.2
+yaspin==2.1.0
+zipp==3.7.0

wandb/run-20220324_170220-1u8cita3/files/wandb-metadata.json

@@ -0,0 +1,58 @@
+{
+    "os": "Linux-5.11.0-1028-gcp-x86_64-with-glibc2.33",
+    "python": "3.9.5",
+    "heartbeatAt": "2022-03-24T17:02:22.232891",
+    "startedAt": "2022-03-24T17:02:20.543386",
+    "docker": null,
+    "gpu": "Tesla V100-SXM2-16GB",
+    "gpu_count": 2,
+    "cpu_count": 16,
+    "cuda": null,
+    "args": [
+        "--dataset_name=librispeech_asr",
+        "--model_name_or_path=./",
+        "--tokenizer_name=./",
+        "--dataset_config_name=clean",
+        "--train_split_name=train.100",
+        "--eval_split_name=validation",
+        "--output_dir=./",
+        "--preprocessing_num_workers=1",
+        "--length_column_name=input_length",
+        "--num_train_epochs=25",
+        "--per_device_train_batch_size=8",
+        "--per_device_eval_batch_size=8",
+        "--gradient_accumulation_steps=4",
+        "--generation_max_length=40",
+        "--generation_num_beams=1",
+        "--learning_rate=3e-5",
+        "--warmup_steps=1000",
+        "--evaluation_strategy=steps",
+        "--text_column_name=text",
+        "--save_steps=1500",
+        "--eval_steps=1500",
+        "--logging_steps=1",
+        "--save_total_limit=1",
+        "--freeze_feature_encoder",
+        "--gradient_checkpointing",
+        "--fp16",
+        "--group_by_length",
+        "--predict_with_generate",
+        "--do_lower_case",
+        "--do_eval",
+        "--do_train",
+        "--push_to_hub",
+        "--use_auth_token"
+    ],
+    "state": "running",
+    "program": "/home/sanchit_huggingface_co/wav2vec2-2-rnd-no-adapter-regularisation/run_speech_recognition_seq2seq.py",
+    "codePath": "run_speech_recognition_seq2seq.py",
+    "git": {
+        "remote": "https://huggingface.co/sanchit-gandhi/wav2vec2-2-rnd-no-adapter-regularisation",
+        "commit": "156c71831efc84fc27b6a9a52f81872c0138b990"
+    },
+    "email": "sanchit@huggingface.co",
+    "root": "/home/sanchit_huggingface_co/wav2vec2-2-rnd-no-adapter-regularisation",
+    "host": "sanchit--v100",
+    "username": "sanchit_huggingface_co",
+    "executable": "/home/sanchit_huggingface_co/gcp/bin/python"
+}

wandb/run-20220324_170220-1u8cita3/logs/debug.log

@@ -0,0 +1,27 @@
+2022-03-24 17:02:20,545 INFO    MainThread:1754191 [wandb_setup.py:_flush():75] Loading settings from /home/sanchit_huggingface_co/.config/wandb/settings
+2022-03-24 17:02:20,545 INFO    MainThread:1754191 [wandb_setup.py:_flush():75] Loading settings from /home/sanchit_huggingface_co/wav2vec2-2-rnd-no-adapter-regularisation/wandb/settings
+2022-03-24 17:02:20,545 INFO    MainThread:1754191 [wandb_setup.py:_flush():75] Loading settings from environment variables: {}
+2022-03-24 17:02:20,545 INFO    MainThread:1754191 [wandb_setup.py:_flush():75] Inferring run settings from compute environment: {'program_relpath': 'run_speech_recognition_seq2seq.py', 'program': '/home/sanchit_huggingface_co/wav2vec2-2-rnd-no-adapter-regularisation/run_speech_recognition_seq2seq.py'}
+2022-03-24 17:02:20,545 INFO    MainThread:1754191 [wandb_init.py:_log_setup():386] Logging user logs to /home/sanchit_huggingface_co/wav2vec2-2-rnd-no-adapter-regularisation/wandb/run-20220324_170220-1u8cita3/logs/debug.log
+2022-03-24 17:02:20,545 INFO    MainThread:1754191 [wandb_init.py:_log_setup():387] Logging internal logs to /home/sanchit_huggingface_co/wav2vec2-2-rnd-no-adapter-regularisation/wandb/run-20220324_170220-1u8cita3/logs/debug-internal.log
+2022-03-24 17:02:20,545 INFO    MainThread:1754191 [wandb_init.py:init():420] calling init triggers
+2022-03-24 17:02:20,545 INFO    MainThread:1754191 [wandb_init.py:init():425] wandb.init called with sweep_config: {}
+config: {}
+2022-03-24 17:02:20,545 INFO    MainThread:1754191 [wandb_init.py:init():471] starting backend
+2022-03-24 17:02:20,545 INFO    MainThread:1754191 [backend.py:_multiprocessing_setup():99] multiprocessing start_methods=fork,spawn,forkserver, using: spawn
+2022-03-24 17:02:20,592 INFO    MainThread:1754191 [backend.py:ensure_launched():219] starting backend process...
+2022-03-24 17:02:20,637 INFO    MainThread:1754191 [backend.py:ensure_launched():224] started backend process with pid: 1754401
+2022-03-24 17:02:20,639 INFO    MainThread:1754191 [wandb_init.py:init():480] backend started and connected
+2022-03-24 17:02:20,649 INFO    MainThread:1754191 [wandb_init.py:init():550] updated telemetry
+2022-03-24 17:02:20,759 INFO    MainThread:1754191 [wandb_init.py:init():581] communicating current version
+2022-03-24 17:02:21,491 INFO    MainThread:1754191 [wandb_init.py:init():586] got version response upgrade_message: "wandb version 0.12.11 is available!  To upgrade, please run:\n $ pip install wandb --upgrade"
+
+2022-03-24 17:02:21,491 INFO    MainThread:1754191 [wandb_init.py:init():596] communicating run to backend with 30 second timeout
+2022-03-24 17:02:22,227 INFO    MainThread:1754191 [wandb_init.py:init():624] starting run threads in backend
+2022-03-24 17:02:22,330 INFO    MainThread:1754191 [wandb_run.py:_console_start():1827] atexit reg
+2022-03-24 17:02:22,331 INFO    MainThread:1754191 [wandb_run.py:_redirect():1701] redirect: SettingsConsole.REDIRECT
+2022-03-24 17:02:22,331 INFO    MainThread:1754191 [wandb_run.py:_redirect():1706] Redirecting console.
+2022-03-24 17:02:22,333 INFO    MainThread:1754191 [wandb_run.py:_redirect():1762] Redirects installed.
+2022-03-24 17:02:22,333 INFO    MainThread:1754191 [wandb_init.py:init():651] run started, returning control to user process
+2022-03-24 17:02:22,335 INFO    MainThread:1754191 [wandb_run.py:_config_callback():966] config_cb None None {'return_dict': True, 'output_hidden_states': False, 'output_attentions': False, 'torchscript': False, 'torch_dtype': 'torch.float32', 'use_bfloat16': False, 'pruned_heads': {}, 'tie_word_embeddings': False, 'is_encoder_decoder': True, 'is_decoder': False, 'cross_attention_hidden_size': None, 'add_cross_attention': False, 'tie_encoder_decoder': False, 'max_length': 50, 'min_length': 0, 'do_sample': False, 'early_stopping': False, 'num_beams': 1, 'num_beam_groups': 1, 'diversity_penalty': 0.0, 'temperature': 1.0, 'top_k': 50, 'top_p': 1.0, 'repetition_penalty': 1.0, 'length_penalty': 1.0, 'no_repeat_ngram_size': 0, 'encoder_no_repeat_ngram_size': 0, 'bad_words_ids': None, 'num_return_sequences': 1, 'chunk_size_feed_forward': 0, 'output_scores': False, 'return_dict_in_generate': False, 'forced_bos_token_id': None, 'forced_eos_token_id': None, 'remove_invalid_values': False, 'architectures': ['SpeechEncoderDecoderModel'], 'finetuning_task': None, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'tokenizer_class': None, 'prefix': None, 'bos_token_id': None, 'pad_token_id': 1, 'eos_token_id': 2, 'sep_token_id': None, 'decoder_start_token_id': 0, 'task_specific_params': None, 'problem_type': None, '_name_or_path': './', 'transformers_version': None, 'decoder': {'vocab_size': 50265, 'max_position_embeddings': 1024, 'd_model': 1024, 'encoder_ffn_dim': 4096, 'encoder_layers': 12, 'encoder_attention_heads': 16, 'decoder_ffn_dim': 4096, 'decoder_layers': 12, 'decoder_attention_heads': 16, 'dropout': 0.1, 'attention_dropout': 0.1, 'activation_dropout': 0.1, 'activation_function': 'gelu', 'init_std': 0.02, 'encoder_layerdrop': 0.0, 'decoder_layerdrop': 0.0, 'classifier_dropout': 0.0, 'use_cache': False, 'num_hidden_layers': 12, 'scale_embedding': False, 'return_dict': True, 'output_hidden_states': False, 'output_attentions': False, 'torchscript': False, 'torch_dtype': 'float32', 'use_bfloat16': False, 'pruned_heads': {}, 'tie_word_embeddings': True, 'is_encoder_decoder': False, 'is_decoder': True, 'cross_attention_hidden_size': None, 'add_cross_attention': True, 'tie_encoder_decoder': False, 'max_length': 20, 'min_length': 0, 'do_sample': False, 'early_stopping': True, 'num_beams': 4, 'num_beam_groups': 1, 'diversity_penalty': 0.0, 'temperature': 1.0, 'top_k': 50, 'top_p': 1.0, 'repetition_penalty': 1.0, 'length_penalty': 1.0, 'no_repeat_ngram_size': 3, 'encoder_no_repeat_ngram_size': 0, 'bad_words_ids': None, 'num_return_sequences': 1, 'chunk_size_feed_forward': 0, 'output_scores': False, 'return_dict_in_generate': False, 'forced_bos_token_id': 0, 'forced_eos_token_id': 2, 'remove_invalid_values': False, 'architectures': ['BartForCausalLM'], 'finetuning_task': None, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1', 2: 'LABEL_2'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1, 'LABEL_2': 2}, 'tokenizer_class': None, 'prefix': None, 'bos_token_id': 0, 'pad_token_id': 1, 'eos_token_id': 2, 'sep_token_id': None, 'decoder_start_token_id': 2, 'task_specific_params': {'summarization': {'length_penalty': 1.0, 'max_length': 128, 'min_length': 12, 'num_beams': 4}, 'summarization_cnn': {'length_penalty': 2.0, 'max_length': 142, 'min_length': 56, 'num_beams': 4}, 'summarization_xsum': {'length_penalty': 1.0, 'max_length': 62, 'min_length': 11, 'num_beams': 6}}, 'problem_type': None, '_name_or_path': 'decoder', 'transformers_version': '4.17.0.dev0', 'add_bias_logits': False, 'add_final_layer_norm': False, 'classif_dropout': 0.1, 'gradient_checkpointing': False, 'normalize_before': False, 'model_type': 'bart'}, 'encoder': {'return_dict': True, 'output_hidden_states': False, 'output_attentions': False, 'torchscript': False, 'torch_dtype': None, 'use_bfloat16': False, 'pruned_heads': {}, 'tie_word_embeddings': True, 'is_encoder_decoder': False, 'is_decoder': False, 'cross_attention_hidden_size': None, 'add_cross_attention': False, 'tie_encoder_decoder': False, 'max_length': 20, 'min_length': 0, 'do_sample': False, 'early_stopping': False, 'num_beams': 1, 'num_beam_groups': 1, 'diversity_penalty': 0.0, 'temperature': 1.0, 'top_k': 50, 'top_p': 1.0, 'repetition_penalty': 1.0, 'length_penalty': 1.0, 'no_repeat_ngram_size': 0, 'encoder_no_repeat_ngram_size': 0, 'bad_words_ids': None, 'num_return_sequences': 1, 'chunk_size_feed_forward': 0, 'output_scores': False, 'return_dict_in_generate': False, 'forced_bos_token_id': None, 'forced_eos_token_id': None, 'remove_invalid_values': False, 'architectures': ['Wav2Vec2ForPreTraining'], 'finetuning_task': None, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'tokenizer_class': None, 'prefix': None, 'bos_token_id': 1, 'pad_token_id': 0, 'eos_token_id': 2, 'sep_token_id': None, 'decoder_start_token_id': None, 'task_specific_params': None, 'problem_type': None, '_name_or_path': 'facebook/wav2vec2-large-lv60', 'transformers_version': '4.17.0.dev0', 'feat_extract_dropout': 0.0, 'gradient_checkpointing': False, 'hidden_dropout_prob': 0.1, 'num_feat_extract_layers': 7, 'hidden_size': 1024, 'feat_extract_norm': 'layer', 'feat_extract_activation': 'gelu', 'conv_dim': [512, 512, 512, 512, 512, 512, 512], 'conv_stride': [5, 2, 2, 2, 2, 2, 2], 'conv_kernel': [10, 3, 3, 3, 3, 2, 2], 'conv_bias': True, 'num_conv_pos_embeddings': 128, 'num_conv_pos_embedding_groups': 16, 'num_hidden_layers': 24, 'intermediate_size': 4096, 'hidden_act': 'gelu', 'num_attention_heads': 16, 'hidden_dropout': 0.1, 'attention_dropout': 0.1, 'activation_dropout': 0.1, 'feat_proj_dropout': 0.0, 'final_dropout': 0.0, 'layerdrop': 0.0, 'layer_norm_eps': 1e-05, 'initializer_range': 0.02, 'vocab_size': 32, 'do_stable_layer_norm': True, 'use_weighted_layer_sum': False, 'apply_spec_augment': True, 'mask_time_prob': 0.1, 'mask_time_length': 10, 'mask_time_min_masks': 2, 'mask_feature_prob': 0.0, 'mask_feature_length': 10, 'mask_feature_min_masks': 0, 'num_codevectors_per_group': 320, 'num_codevector_groups': 2, 'contrastive_logits_temperature': 0.1, 'feat_quantizer_dropout': 0.0, 'num_negatives': 100, 'codevector_dim': 768, 'proj_codevector_dim': 768, 'diversity_loss_weight': 0.1, 'ctc_loss_reduction': 'sum', 'ctc_zero_infinity': False, 'add_adapter': False, 'adapter_kernel_size': 3, 'adapter_stride': 2, 'num_adapter_layers': 3, 'output_hidden_size': 1024, 'classifier_proj_size': 256, 'tdnn_dim': [512, 512, 512, 512, 1500], 'tdnn_kernel': [5, 3, 3, 1, 1], 'tdnn_dilation': [1, 2, 3, 1, 1], 'xvector_output_dim': 512, 'model_type': 'wav2vec2'}, 'model_type': 'speech-encoder-decoder', 'processor_class': 'Wav2Vec2Processor', 'use_cache': False, 'output_dir': './', 'overwrite_output_dir': False, 'do_train': True, 'do_eval': True, 'do_predict': False, 'evaluation_strategy': 'steps', 'prediction_loss_only': False, 'per_device_train_batch_size': 8, 'per_device_eval_batch_size': 8, 'per_gpu_train_batch_size': 'None', 'per_gpu_eval_batch_size': 'None', 'gradient_accumulation_steps': 4, 'eval_accumulation_steps': 'None', 'learning_rate': 3e-05, 'weight_decay': 0.0, 'adam_beta1': 0.9, 'adam_beta2': 0.999, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 25.0, 'max_steps': -1, 'lr_scheduler_type': 'linear', 'warmup_ratio': 0.0, 'warmup_steps': 1000, 'log_level': -1, 'log_level_replica': -1, 'log_on_each_node': True, 'logging_dir': './runs/Mar24_17-01-37_sanchit--v100', 'logging_strategy': 'steps', 'logging_first_step': False, 'logging_steps': 1, 'logging_nan_inf_filter': True, 'save_strategy': 'steps', 'save_steps': 1500, 'save_total_limit': 1, 'save_on_each_node': False, 'no_cuda': False, 'seed': 42, 'bf16': False, 'fp16': True, 'fp16_opt_level': 'O1', 'half_precision_backend': 'amp', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': 'None', 'local_rank': -1, 'xpu_backend': 'None', 'tpu_num_cores': 'None', 'tpu_metrics_debug': False, 'debug': '[]', 'dataloader_drop_last': False, 'eval_steps': 1500, 'dataloader_num_workers': 0, 'past_index': -1, 'run_name': './', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': 'None', 'load_best_model_at_end': False, 'metric_for_best_model': 'None', 'greater_is_better': 'None', 'ignore_data_skip': False, 'sharded_ddp': '[]', 'deepspeed': 'None', 'label_smoothing_factor': 0.0, 'optim': 'adamw_hf', 'adafactor': False, 'group_by_length': True, 'length_column_name': 'input_length', 'report_to': "['tensorboard', 'wandb']", 'ddp_find_unused_parameters': 'None', 'ddp_bucket_cap_mb': 'None', 'dataloader_pin_memory': True, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': True, 'resume_from_checkpoint': 'None', 'hub_model_id': 'None', 'hub_strategy': 'every_save', 'hub_token': '<HUB_TOKEN>', 'gradient_checkpointing': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': 'None', 'push_to_hub_organization': 'None', 'push_to_hub_token': '<PUSH_TO_HUB_TOKEN>', '_n_gpu': 1, 'mp_parameters': '', 'sortish_sampler': False, 'predict_with_generate': True, 'generation_max_length': 40, 'generation_num_beams': 1, 'train_batch_size': 8, 'eval_batch_size': 8}
+2022-03-24 17:02:22,338 INFO    MainThread:1754191 [wandb_watch.py:watch():43] Watching

wandb/run-20220324_170220-1u8cita3/run-1u8cita3.wandb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1c601a6b5ff5247537f483a3d4f6ce97f4213376a2c6fd3ed33265c4ed120f1e
+size 75250182