reecursion commited on
Commit
9689953
·
verified ·
1 Parent(s): e6e4dec

Upload fine-tuned OWSM model with exp/finetune directory

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. .gitattributes +2 -0
  2. README.md +3 -4
  3. exp/finetune/1epoch.pth +3 -0
  4. exp/finetune/2epoch.pth +3 -0
  5. exp/finetune/3epoch.pth +3 -0
  6. exp/finetune/4epoch.pth +3 -0
  7. exp/finetune/checkpoint.pth +3 -0
  8. exp/finetune/config.yaml +1 -1
  9. exp/finetune/images/acc.png +0 -0
  10. exp/finetune/images/backward_time.png +0 -0
  11. exp/finetune/images/cer.png +0 -0
  12. exp/finetune/images/cer_ctc.png +0 -0
  13. exp/finetune/images/clip.png +0 -0
  14. exp/finetune/images/forward_time.png +0 -0
  15. exp/finetune/images/gpu_max_cached_mem_GB.png +0 -0
  16. exp/finetune/images/grad_norm.png +0 -0
  17. exp/finetune/images/iter_time.png +0 -0
  18. exp/finetune/images/loss.png +0 -0
  19. exp/finetune/images/loss_att.png +0 -0
  20. exp/finetune/images/loss_ctc.png +0 -0
  21. exp/finetune/images/loss_scale.png +0 -0
  22. exp/finetune/images/optim0_lr0.png +0 -0
  23. exp/finetune/images/optim_step_time.png +0 -0
  24. exp/finetune/images/train_time.png +0 -0
  25. exp/finetune/images/wer.png +0 -0
  26. exp/finetune/latest.pth +3 -0
  27. exp/finetune/tensorboard/train/events.out.tfevents.1745339185.babel-4-33.3934128.0 +3 -0
  28. exp/finetune/tensorboard/valid/events.out.tfevents.1745339185.babel-4-33.3934128.1 +3 -0
  29. exp/finetune/valid.acc.ave.pth +3 -0
  30. exp/finetune/valid.acc.ave_3best.pth +3 -0
  31. exp/finetune/valid.acc.best.pth +3 -0
  32. exp/finetune/wandb/debug-internal.log +17 -0
  33. exp/finetune/wandb/debug.log +0 -0
  34. exp/finetune/wandb/latest-run/files/config.yaml +0 -0
  35. exp/finetune/wandb/latest-run/files/output.log +0 -0
  36. exp/finetune/wandb/latest-run/files/requirements.txt +309 -0
  37. exp/finetune/wandb/latest-run/files/wandb-metadata.json +98 -0
  38. exp/finetune/wandb/latest-run/files/wandb-summary.json +1 -0
  39. exp/finetune/wandb/latest-run/logs/debug-core.log +15 -0
  40. exp/finetune/wandb/latest-run/logs/debug-internal.log +17 -0
  41. exp/finetune/wandb/latest-run/logs/debug.log +0 -0
  42. exp/finetune/wandb/latest-run/run-hc47k8jw.wandb +3 -0
  43. exp/finetune/wandb/run-20250422_122624-hc47k8jw/files/config.yaml +0 -0
  44. exp/finetune/wandb/run-20250422_122624-hc47k8jw/files/output.log +0 -0
  45. exp/finetune/wandb/run-20250422_122624-hc47k8jw/files/requirements.txt +309 -0
  46. exp/finetune/wandb/run-20250422_122624-hc47k8jw/files/wandb-metadata.json +98 -0
  47. exp/finetune/wandb/run-20250422_122624-hc47k8jw/files/wandb-summary.json +1 -0
  48. exp/finetune/wandb/run-20250422_122624-hc47k8jw/logs/debug-core.log +15 -0
  49. exp/finetune/wandb/run-20250422_122624-hc47k8jw/logs/debug-internal.log +17 -0
  50. exp/finetune/wandb/run-20250422_122624-hc47k8jw/logs/debug.log +0 -0
.gitattributes CHANGED
@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ exp/finetune/wandb/latest-run/run-hc47k8jw.wandb filter=lfs diff=lfs merge=lfs -text
37
+ exp/finetune/wandb/run-20250422_122624-hc47k8jw/run-hc47k8jw.wandb filter=lfs diff=lfs merge=lfs -text
README.md CHANGED
@@ -1,5 +1,4 @@
1
  # Common Accent ASR Model
2
-
3
  This is a fine-tuned ASR model based on [espnet/owsm_v3.1_ebf_base](https://huggingface.co/espnet/owsm_v3.1_ebf_base) trained on the [DTU54DL/common-accent](https://huggingface.co/datasets/DTU54DL/common-accent) dataset.
4
 
5
  ## Model details
@@ -8,7 +7,6 @@ This is a fine-tuned ASR model based on [espnet/owsm_v3.1_ebf_base](https://hugg
8
  - Task: Automatic Speech Recognition
9
 
10
  ## Usage
11
-
12
  ```python
13
  import torch
14
  import numpy as np
@@ -23,6 +21,7 @@ model = Speech2Text.from_pretrained(
23
  )
24
 
25
  # Example inference
26
- waveform = ... # Load your audio as numpy array
27
  transcription = model(waveform)
28
- print(transcription[0][0]) # Print the transcription
 
 
1
  # Common Accent ASR Model
 
2
  This is a fine-tuned ASR model based on [espnet/owsm_v3.1_ebf_base](https://huggingface.co/espnet/owsm_v3.1_ebf_base) trained on the [DTU54DL/common-accent](https://huggingface.co/datasets/DTU54DL/common-accent) dataset.
3
 
4
  ## Model details
 
7
  - Task: Automatic Speech Recognition
8
 
9
  ## Usage
 
10
  ```python
11
  import torch
12
  import numpy as np
 
21
  )
22
 
23
  # Example inference
24
+ waveform = ... # Load your audio as numpy array
25
  transcription = model(waveform)
26
+ print(transcription[0][0]) # Print the transcription
27
+ ```
exp/finetune/1epoch.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ebbffdc800a7b5e14a4b39230f86f1cbded7c0bba91886d4852282c212a7cb74
3
+ size 404962360
exp/finetune/2epoch.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9a2346271c134d72130a43803053bfb4299c3ae4ea10c6a857992254ac7286e5
3
+ size 404962360
exp/finetune/3epoch.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7b6908f60a520456effc351ac4368dee639517118ecf1a47712473916b9375cd
3
+ size 404962360
exp/finetune/4epoch.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fc88df7bc5f7659f12a65a1d5742d80157eb1036ace66cfdf7629e28a62d11ba
3
+ size 404962360
exp/finetune/checkpoint.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6cdc80e832c1c363bc77b3554b79dbf37d12ad023cb007e90af1ac42133ab4a7
3
+ size 1214776468
exp/finetune/config.yaml CHANGED
@@ -16337,7 +16337,7 @@ token_list:
16337
  - 有效
16338
  - volle
16339
  - ▁May
16340
- - َّ
16341
  - ▁honey
16342
  - 行って
16343
  - чет
 
16337
  - 有效
16338
  - volle
16339
  - ▁May
16340
+ - َّ
16341
  - ▁honey
16342
  - 行って
16343
  - чет
exp/finetune/images/acc.png ADDED
exp/finetune/images/backward_time.png ADDED
exp/finetune/images/cer.png ADDED
exp/finetune/images/cer_ctc.png ADDED
exp/finetune/images/clip.png ADDED
exp/finetune/images/forward_time.png ADDED
exp/finetune/images/gpu_max_cached_mem_GB.png ADDED
exp/finetune/images/grad_norm.png ADDED
exp/finetune/images/iter_time.png ADDED
exp/finetune/images/loss.png ADDED
exp/finetune/images/loss_att.png ADDED
exp/finetune/images/loss_ctc.png ADDED
exp/finetune/images/loss_scale.png ADDED
exp/finetune/images/optim0_lr0.png ADDED
exp/finetune/images/optim_step_time.png ADDED
exp/finetune/images/train_time.png ADDED
exp/finetune/images/wer.png ADDED
exp/finetune/latest.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fc88df7bc5f7659f12a65a1d5742d80157eb1036ace66cfdf7629e28a62d11ba
3
+ size 404962360
exp/finetune/tensorboard/train/events.out.tfevents.1745339185.babel-4-33.3934128.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:87fddd289574d25dbb4fab9f15731ee5846cfd7e9333127eeb3f552b73abf7ea
3
+ size 127256
exp/finetune/tensorboard/valid/events.out.tfevents.1745339185.babel-4-33.3934128.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c2ce1078bcbe5a2b12b5be8993ed5d0cccaa90ea8b02d36776fa7ad6e7f79c14
3
+ size 1564
exp/finetune/valid.acc.ave.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:15823d9cb130150a9243cba36e92171093f49a3dd5f679dbb560464a4c1190c5
3
+ size 404968406
exp/finetune/valid.acc.ave_3best.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:15823d9cb130150a9243cba36e92171093f49a3dd5f679dbb560464a4c1190c5
3
+ size 404968406
exp/finetune/valid.acc.best.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ebbffdc800a7b5e14a4b39230f86f1cbded7c0bba91886d4852282c212a7cb74
3
+ size 404962360
exp/finetune/wandb/debug-internal.log ADDED
@@ -0,0 +1,17 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"time":"2025-04-22T12:26:24.699996553-04:00","level":"INFO","msg":"stream: starting","core version":"0.19.6","symlink path":"exp/finetune/wandb/run-20250422_122624-hc47k8jw/logs/debug-core.log"}
2
+ {"time":"2025-04-22T12:26:24.81223292-04:00","level":"INFO","msg":"created new stream","id":"hc47k8jw"}
3
+ {"time":"2025-04-22T12:26:24.812702295-04:00","level":"INFO","msg":"writer: Do: started","stream_id":"hc47k8jw"}
4
+ {"time":"2025-04-22T12:26:24.812796834-04:00","level":"INFO","msg":"sender: started","stream_id":"hc47k8jw"}
5
+ {"time":"2025-04-22T12:26:24.812793588-04:00","level":"INFO","msg":"handler: started","stream_id":"hc47k8jw"}
6
+ {"time":"2025-04-22T12:26:24.812797595-04:00","level":"INFO","msg":"stream: started","id":"hc47k8jw"}
7
+ {"time":"2025-04-22T12:26:25.072522218-04:00","level":"INFO","msg":"Starting system monitor"}
8
+ {"time":"2025-04-22T13:28:37.53429929-04:00","level":"INFO","msg":"api: retrying error","error":"Post \"https://api.wandb.ai/files/gganeshl-carnegie-mellon-university/ESPnet_ESPnetEZDataTask/hc47k8jw/file_stream\": dial tcp 35.186.228.49:443: connect: connection timed out"}
9
+ {"time":"2025-04-22T13:36:00.330987928-04:00","level":"INFO","msg":"Stopping system monitor"}
10
+ {"time":"2025-04-22T13:36:00.33232475-04:00","level":"INFO","msg":"Stopped system monitor"}
11
+ {"time":"2025-04-22T13:36:01.292531969-04:00","level":"INFO","msg":"handler: operation stats","stats":{}}
12
+ {"time":"2025-04-22T13:36:01.461755077-04:00","level":"INFO","msg":"fileTransfer: Close: file transfer manager closed"}
13
+ {"time":"2025-04-22T13:36:03.418249732-04:00","level":"INFO","msg":"stream: closing","id":"hc47k8jw"}
14
+ {"time":"2025-04-22T13:36:03.418297973-04:00","level":"INFO","msg":"handler: closed","stream_id":"hc47k8jw"}
15
+ {"time":"2025-04-22T13:36:03.418318932-04:00","level":"INFO","msg":"writer: Close: closed","stream_id":"hc47k8jw"}
16
+ {"time":"2025-04-22T13:36:03.418339511-04:00","level":"INFO","msg":"sender: closed","stream_id":"hc47k8jw"}
17
+ {"time":"2025-04-22T13:36:03.421317411-04:00","level":"INFO","msg":"stream: closed","id":"hc47k8jw"}
exp/finetune/wandb/debug.log ADDED
The diff for this file is too large to render. See raw diff
 
exp/finetune/wandb/latest-run/files/config.yaml ADDED
The diff for this file is too large to render. See raw diff
 
exp/finetune/wandb/latest-run/files/output.log ADDED
The diff for this file is too large to render. See raw diff
 
exp/finetune/wandb/latest-run/files/requirements.txt ADDED
@@ -0,0 +1,309 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ monai==1.4.0
2
+ nvidia-nvjitlink-cu12==12.4.127
3
+ jiter==0.8.2
4
+ accelerate==1.3.0
5
+ nvidia-nvtx-cu12==12.4.127
6
+ gitdb==4.0.12
7
+ rfc3986-validator==0.1.1
8
+ proto-plus==1.26.0
9
+ jupyterlab==4.3.5
10
+ prometheus-fastapi-instrumentator==7.0.2
11
+ tomli==2.2.1
12
+ json5==0.10.0
13
+ sniffio==1.3.1
14
+ google-auth==2.38.0
15
+ msgspec==0.19.0
16
+ nvidia-cusparse-cu12==12.3.1.170
17
+ traitlets==5.14.3
18
+ torchvision==0.20.1
19
+ referencing==0.36.2
20
+ astor==0.8.1
21
+ decorator==5.1.1
22
+ importlib_resources==6.4.5
23
+ debugpy==1.8.13
24
+ compressed-tensors==0.9.1
25
+ dill==0.3.8
26
+ evaluate==0.4.3
27
+ pycparser==2.22
28
+ nvidia-nccl-cu12==2.21.5
29
+ mpmath==1.3.0
30
+ ptyprocess==0.7.0
31
+ cffi==1.17.1
32
+ multidict==6.1.0
33
+ bleach==6.2.0
34
+ fastjsonschema==2.21.1
35
+ blake3==1.0.4
36
+ partial-json-parser==0.2.1.1.post5
37
+ jedi==0.19.2
38
+ xgrammar==0.1.11
39
+ nvidia-cusolver-cu12==11.6.1.9
40
+ asttokens==3.0.0
41
+ rfc3339-validator==0.1.4
42
+ aiolimiter==1.2.1
43
+ smart-open==7.1.0
44
+ pyarrow==18.0.0
45
+ torchaudio==2.5.1
46
+ charset-normalizer==3.4.0
47
+ GitPython==3.1.44
48
+ sentencepiece==0.2.0
49
+ nvidia-cuda-runtime-cu12==12.4.127
50
+ huggingface-hub==0.26.2
51
+ pandas==2.2.3
52
+ opencv-python-headless==4.11.0.86
53
+ pyasn1==0.6.1
54
+ msgpack==1.1.0
55
+ frozenlist==1.5.0
56
+ regex==2024.11.6
57
+ pybind11==2.13.6
58
+ smmap==5.0.2
59
+ xxhash==3.5.0
60
+ llvmlite==0.43.0
61
+ uri-template==1.3.0
62
+ multiprocess==0.70.16
63
+ argon2-cffi-bindings==21.2.0
64
+ webencodings==0.5.1
65
+ mistune==3.1.2
66
+ webcolors==24.11.1
67
+ filelock==3.16.1
68
+ tenacity==9.0.0
69
+ fonttools==4.55.1
70
+ nvidia-cuda-cupti-cu12==12.4.127
71
+ widgetsnbextension==4.0.13
72
+ aiohappyeyeballs==2.4.3
73
+ sglang==0.4.2.post3
74
+ yarl==1.18.0
75
+ pandocfilters==1.5.1
76
+ iniconfig==2.0.0
77
+ nbconvert==7.16.6
78
+ ipykernel==6.29.5
79
+ jupyter_core==5.7.2
80
+ vllm==0.7.3.dev57+g2ae88905
81
+ notebook==7.3.2
82
+ certifi==2024.8.30
83
+ websockets==14.2
84
+ googleapis-common-protos==1.67.0rc1
85
+ jupyter_client==8.6.3
86
+ safetensors==0.4.5
87
+ jupyter-lsp==2.2.5
88
+ scikit-learn==1.5.2
89
+ kiwisolver==1.4.7
90
+ einops==0.8.0
91
+ eval_type_backport==0.2.2
92
+ notebook_shim==0.2.4
93
+ pyasn1_modules==0.4.1
94
+ wcwidth==0.2.13
95
+ nvidia-cufft-cu12==11.2.1.3
96
+ jupyterlab_widgets==3.0.13
97
+ pyzmq==26.2.1
98
+ pure_eval==0.2.3
99
+ wandb==0.19.6
100
+ docker-pycreds==0.4.0
101
+ executing==2.2.0
102
+ ipython==8.18.1
103
+ distlib==0.3.9
104
+ uvloop==0.21.0
105
+ python-dotenv==1.0.1
106
+ matplotlib==3.9.3
107
+ tiktoken==0.8.0
108
+ requests==2.32.3
109
+ protobuf==5.29.3
110
+ distro==1.9.0
111
+ wrapt==1.17.2
112
+ pydantic==2.10.6
113
+ soupsieve==2.6
114
+ rouge-score==0.1.2
115
+ pycountry==24.6.1
116
+ terminado==0.18.1
117
+ fsspec==2024.9.0
118
+ Send2Trash==1.8.3
119
+ py-cpuinfo==9.0.0
120
+ sentry-sdk==2.20.0
121
+ scipy==1.13.1
122
+ transformers==4.48.3
123
+ tokenizers==0.21.0
124
+ rsa==4.9
125
+ jupyter-events==0.12.0
126
+ slicer==0.0.8
127
+ async-lru==2.0.4
128
+ openai==1.60.2
129
+ platformdirs==4.3.6
130
+ pytest==8.3.4
131
+ bitsandbytes==0.45.5
132
+ async-timeout==5.0.1
133
+ joblib==1.4.2
134
+ tinycss2==1.4.0
135
+ seaborn==0.13.2
136
+ fqdn==1.5.1
137
+ isoduration==20.11.0
138
+ httpx==0.28.1
139
+ numba==0.60.0
140
+ beautifulsoup4==4.13.3
141
+ jupyter_server_terminals==0.5.3
142
+ sympy==1.13.1
143
+ ray==2.42.0
144
+ httpcore==1.0.7
145
+ depyf==0.18.0
146
+ overrides==7.7.0
147
+ nbformat==5.10.4
148
+ babel==2.17.0
149
+ argon2-cffi==23.1.0
150
+ nvidia-curand-cu12==10.3.5.147
151
+ python-dateutil==2.9.0.post0
152
+ uvicorn==0.34.0
153
+ aiohttp-cors==0.7.0
154
+ nvidia-cuda-nvrtc-cu12==12.4.127
155
+ starlette==0.45.3
156
+ aiosignal==1.3.1
157
+ shap==0.46.0
158
+ matplotlib-inline==0.1.7
159
+ absl-py==2.1.0
160
+ jupyterlab_server==2.27.3
161
+ cloudpickle==3.1.1
162
+ datasets==3.1.0
163
+ jsonschema-specifications==2024.10.1
164
+ comm==0.2.2
165
+ parso==0.8.4
166
+ httptools==0.6.4
167
+ websocket-client==1.8.0
168
+ nvidia-cublas-cu12==12.4.5.8
169
+ six==1.17.0
170
+ outlines==0.1.11
171
+ google-api-core==2.24.1
172
+ jupyter_server==2.15.0
173
+ cachetools==5.5.1
174
+ annotated-types==0.7.0
175
+ virtualenv==20.29.1
176
+ threadpoolctl==3.5.0
177
+ prompt_toolkit==3.0.50
178
+ torch==2.5.1
179
+ watchfiles==1.0.4
180
+ urllib3==2.3.0
181
+ attrs==24.2.0
182
+ nltk==3.9.1
183
+ propcache==0.2.0
184
+ jsonpointer==3.0.0
185
+ aiohttp==3.11.7
186
+ pillow==11.0.0
187
+ outlines_core==0.1.26
188
+ airportsdata==20241001
189
+ pydantic_core==2.27.2
190
+ defusedxml==0.7.1
191
+ lm-format-enforcer==0.10.9
192
+ opencensus==0.11.4
193
+ python-json-logger==3.2.1
194
+ stack-data==0.6.3
195
+ Pygments==2.19.1
196
+ tqdm==4.67.1
197
+ h11==0.14.0
198
+ pluggy==1.5.0
199
+ triton==3.1.0
200
+ rpds-py==0.22.3
201
+ jupyterlab_pygments==0.3.0
202
+ exceptiongroup==1.2.2
203
+ contourpy==1.3.0
204
+ interegular==0.3.3
205
+ jsonschema==4.23.0
206
+ nvidia-cudnn-cu12==9.1.0.70
207
+ pexpect==4.9.0
208
+ jupyter==1.1.1
209
+ opencensus-context==0.2.dev0
210
+ anyio==4.8.0
211
+ ipywidgets==8.1.5
212
+ mistral_common==1.5.3
213
+ nbclient==0.10.2
214
+ gguf==0.10.0
215
+ nest-asyncio==1.6.0
216
+ xformers==0.0.28.post3
217
+ jupyter-console==6.6.3
218
+ tornado==6.4.2
219
+ packaging==24.2
220
+ arrow==1.3.0
221
+ py-spy==0.4.0
222
+ grpcio==1.70.0
223
+ networkx==3.2.1
224
+ typing_extensions==4.12.2
225
+ lark==1.2.2
226
+ colorful==0.6.0a1
227
+ fastapi==0.115.8
228
+ monai-weekly==1.5.dev2505
229
+ diskcache==5.6.3
230
+ types-python-dateutil==2.9.0.20241206
231
+ setproctitle==1.3.4
232
+ prometheus_client==0.21.1
233
+ cycler==0.12.1
234
+ tzdata==2024.2
235
+ pytz==2025.2
236
+ pyparsing==3.2.3
237
+ Werkzeug==3.1.3
238
+ ctc_segmentation==1.7.4
239
+ espnet-tts-frontend==0.0.3
240
+ audioread==3.0.1
241
+ numpy==1.26.4
242
+ h5py==3.13.0
243
+ idna==3.10
244
+ librosa==0.9.2
245
+ zipp==3.21.0
246
+ omegaconf==2.3.0
247
+ lightning==2.5.1
248
+ resampy==0.4.3
249
+ jiwer==3.1.0
250
+ pooch==1.8.2
251
+ jaconv==0.4.0
252
+ pypinyin==0.44.0
253
+ g2p-en==2.1.0
254
+ asteroid-filterbanks==0.4.0
255
+ PySocks==1.7.1
256
+ ci_sdr==0.0.2
257
+ fast_bss_eval==0.1.3
258
+ Markdown==3.8
259
+ hydra-core==1.3.2
260
+ espnet==202503
261
+ setuptools==73.0.1
262
+ espnet-model-zoo==0.1.7
263
+ PyYAML==6.0.2
264
+ torch-complex==0.4.4
265
+ ConfigArgParse==1.7
266
+ Unidecode==1.3.8
267
+ soundfile==0.13.1
268
+ jamo==0.4.1
269
+ Cython==3.0.12
270
+ kaldiio==2.18.1
271
+ editdistance==0.8.1
272
+ humanfriendly==10.0
273
+ psutil==7.0.0
274
+ Distance==0.1.3
275
+ tensorboard-data-server==0.7.2
276
+ click==8.1.8
277
+ gdown==5.2.0
278
+ typeguard==4.4.2
279
+ lightning-utilities==0.14.3
280
+ MarkupSafe==3.0.2
281
+ pyworld==0.3.5
282
+ pytorch-lightning==2.5.1
283
+ pip==25.0
284
+ torchmetrics==1.7.1
285
+ tensorboard==2.19.0
286
+ more-itertools==10.6.0
287
+ Jinja2==3.1.6
288
+ importlib-metadata==4.13.0
289
+ RapidFuzz==3.13.0
290
+ wheel==0.45.1
291
+ antlr4-python3-runtime==4.9.3
292
+ inflect==7.5.0
293
+ opt_einsum==3.4.0
294
+ importlib_resources==6.4.0
295
+ wheel==0.43.0
296
+ backports.tarfile==1.2.0
297
+ autocommand==2.2.2
298
+ inflect==7.3.1
299
+ typing_extensions==4.12.2
300
+ jaraco.text==3.12.1
301
+ zipp==3.19.2
302
+ more-itertools==10.3.0
303
+ platformdirs==4.2.2
304
+ packaging==24.1
305
+ typeguard==4.3.0
306
+ jaraco.functools==4.0.1
307
+ jaraco.context==5.3.0
308
+ importlib_metadata==8.0.0
309
+ tomli==2.0.1
exp/finetune/wandb/latest-run/files/wandb-metadata.json ADDED
@@ -0,0 +1,98 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "os": "Linux-5.14.0-427.42.1.el9_4.x86_64-x86_64-with-glibc2.34",
3
+ "python": "CPython 3.9.6",
4
+ "startedAt": "2025-04-22T16:26:24.696915Z",
5
+ "program": "/home/gganeshl/speech/asr_ft.py",
6
+ "codePath": "asr_ft.py",
7
+ "email": "gganeshl@andrew.cmu.edu",
8
+ "root": "exp/finetune",
9
+ "host": "babel-4-33",
10
+ "executable": "/home/gganeshl/miniconda3/envs/speech/bin/python",
11
+ "codePathLocal": "asr_ft.py",
12
+ "cpu_count": 32,
13
+ "cpu_count_logical": 64,
14
+ "gpu": "NVIDIA RTX A6000",
15
+ "gpu_count": 1,
16
+ "disk": {
17
+ "/": {
18
+ "total": "236663791616",
19
+ "used": "88929800192"
20
+ }
21
+ },
22
+ "memory": {
23
+ "total": "540351823872"
24
+ },
25
+ "cpu": {
26
+ "count": 32,
27
+ "countLogical": 64
28
+ },
29
+ "gpu_nvidia": [
30
+ {
31
+ "name": "NVIDIA RTX A6000",
32
+ "memoryTotal": "51527024640",
33
+ "cudaCores": 10752,
34
+ "architecture": "Ampere"
35
+ }
36
+ ],
37
+ "slurm": {
38
+ "cluster_name": "babel",
39
+ "conf": "/var/spool/slurmd/conf-cache/slurm.conf",
40
+ "cpu_bind": "quiet,mask_cpu:0x00000001000002000000000100000200",
41
+ "cpu_bind_list": "0x00000001000002000000000100000200",
42
+ "cpu_bind_type": "mask_cpu:",
43
+ "cpu_bind_verbose": "quiet",
44
+ "cpus_on_node": "4",
45
+ "cpus_per_task": "1",
46
+ "distribution": "cyclic,pack",
47
+ "gpus_on_node": "1",
48
+ "gtids": "0",
49
+ "job_account": "cp3a",
50
+ "job_cpus_per_node": "4",
51
+ "job_end_time": "1745511243",
52
+ "job_gid": "2709914",
53
+ "job_gpus": "5",
54
+ "job_group": "gganeshl",
55
+ "job_id": "4624778",
56
+ "job_name": "speech_asr",
57
+ "job_nodelist": "babel-4-33",
58
+ "job_num_nodes": "1",
59
+ "job_partition": "general",
60
+ "job_qos": "normal",
61
+ "job_start_time": "1745338743",
62
+ "job_uid": "2709914",
63
+ "job_user": "gganeshl",
64
+ "jobid": "4624778",
65
+ "launch_node_ipaddr": "172.16.1.5",
66
+ "localid": "0",
67
+ "mem_per_node": "32768",
68
+ "nnodes": "1",
69
+ "nodeid": "0",
70
+ "nodelist": "babel-4-33",
71
+ "nprocs": "1",
72
+ "ntasks": "1",
73
+ "prio_process": "0",
74
+ "procid": "0",
75
+ "pty_port": "37249",
76
+ "pty_win_col": "134",
77
+ "pty_win_row": "25",
78
+ "script_context": "prolog_task",
79
+ "srun_comm_host": "172.16.1.5",
80
+ "srun_comm_port": "32883",
81
+ "step_gpus": "3",
82
+ "step_id": "0",
83
+ "step_launcher_port": "32883",
84
+ "step_nodelist": "babel-6-21",
85
+ "step_num_nodes": "1",
86
+ "step_num_tasks": "1",
87
+ "step_tasks_per_node": "1",
88
+ "stepid": "0",
89
+ "submit_dir": "/home/gganeshl/speech",
90
+ "submit_host": "babel-6-21",
91
+ "task_pid": "3934047",
92
+ "tasks_per_node": "4",
93
+ "topology_addr": "babel-4-33",
94
+ "topology_addr_pattern": "node",
95
+ "tres_per_task": "cpu=4"
96
+ },
97
+ "cudaVersion": "12.6"
98
+ }
exp/finetune/wandb/latest-run/files/wandb-summary.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"train/train_loss_ctc_epoch":8.894595814524148,"train/train_time":8.246310259787366,"_runtime":4169.55117254,"train/train_loss_epoch":1.2184081917391751,"train/train_forward_time_epoch":0.05445344569766894,"valid/valid_acc_epoch":0.8508969873189927,"train/train_optim_step_time_epoch":0.006779810173436999,"valid/valid_loss_ctc_epoch":21.5558465385437,"valid/valid_cer_ctc_epoch":0.12294024802744388,"epoch":4,"train/train_loss_att_epoch":3.15036278873905,"metrics/loss_scale":1,"metrics/iter_time":1.937151960558258,"train/train_backward_time_epoch":0.05791670780628919,"_timestamp":1.7453433542476704e+09,"train/train_iter_time_epoch":1.937151960558258,"metrics/clip":100,"train/train_clip_epoch":100,"metrics/loss":1.2184081917391751,"_step":203,"train/train_gpu_max_cached_mem_GB_epoch":10.736328125,"train/train_grad_norm_epoch":34.59821179962158,"metrics/optim_step_time":0.006779810173436999,"iteration":2000,"train/train_optim0_lr0_epoch":0.00010000000000000003,"valid/valid_loss_epoch":17.118548574447633,"train/train_train_time_epoch":8.246310259787366,"train/train_loss_scale_epoch":1,"valid/valid_wer_epoch":0.9300000011920929,"metrics/loss_ctc":8.894595814524148,"valid/valid_loss_att_epoch":15.216849050521851,"train/train_acc_epoch":0.9732946280975294,"metrics/optim0_lr0":0.00010000000000000003,"metrics/backward_time":0.05791670780628919,"valid/valid_gpu_max_cached_mem_GB_epoch":10.736328125,"metrics/forward_time":0.05445344569766894,"metrics/loss_att":3.15036278873905,"valid/valid_cer_epoch":0.09997057549655437,"_wandb":{"runtime":4175},"metrics/grad_norm":34.59821179962158,"metrics/acc":0.9732946280975294}
exp/finetune/wandb/latest-run/logs/debug-core.log ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"time":"2025-04-22T12:26:24.672048644-04:00","level":"INFO","msg":"main: starting server","port-filename":"/tmp/tmpy0krtiml/port-3934128.txt","pid":3934128,"log-level":0,"disable-analytics":false,"shutdown-on-parent-exit":false}
2
+ {"time":"2025-04-22T12:26:24.673781794-04:00","level":"INFO","msg":"Will exit if parent process dies.","ppid":3934128}
3
+ {"time":"2025-04-22T12:26:24.6746965-04:00","level":"INFO","msg":"server is running","addr":{"IP":"127.0.0.1","Port":45355,"Zone":""}}
4
+ {"time":"2025-04-22T12:26:24.688179568-04:00","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"127.0.0.1:53318"}
5
+ {"time":"2025-04-22T12:26:24.699245967-04:00","level":"INFO","msg":"handleInformInit: received","streamId":"hc47k8jw","id":"127.0.0.1:53318"}
6
+ {"time":"2025-04-22T12:26:24.812922771-04:00","level":"INFO","msg":"handleInformInit: stream started","streamId":"hc47k8jw","id":"127.0.0.1:53318"}
7
+ {"time":"2025-04-22T13:36:03.417988098-04:00","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"hc47k8jw","id":"127.0.0.1:53318"}
8
+ {"time":"2025-04-22T13:36:03.4213275-04:00","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"hc47k8jw","id":"127.0.0.1:53318"}
9
+ {"time":"2025-04-22T13:36:03.458763144-04:00","level":"INFO","msg":"handleInformTeardown: server teardown initiated","id":"127.0.0.1:53318"}
10
+ {"time":"2025-04-22T13:36:03.458778322-04:00","level":"INFO","msg":"handleInformTeardown: server shutdown complete","id":"127.0.0.1:53318"}
11
+ {"time":"2025-04-22T13:36:03.458788401-04:00","level":"INFO","msg":"server is shutting down"}
12
+ {"time":"2025-04-22T13:36:03.458799733-04:00","level":"INFO","msg":"connection: closing","id":"127.0.0.1:53318"}
13
+ {"time":"2025-04-22T13:36:03.458922374-04:00","level":"INFO","msg":"connection: closed successfully","id":"127.0.0.1:53318"}
14
+ {"time":"2025-04-22T13:36:03.458959755-04:00","level":"INFO","msg":"connection: ManageConnectionData: connection closed","id":"127.0.0.1:53318"}
15
+ {"time":"2025-04-22T13:36:03.458966237-04:00","level":"INFO","msg":"server is closed"}
exp/finetune/wandb/latest-run/logs/debug-internal.log ADDED
@@ -0,0 +1,17 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"time":"2025-04-22T12:26:24.699996553-04:00","level":"INFO","msg":"stream: starting","core version":"0.19.6","symlink path":"exp/finetune/wandb/run-20250422_122624-hc47k8jw/logs/debug-core.log"}
2
+ {"time":"2025-04-22T12:26:24.81223292-04:00","level":"INFO","msg":"created new stream","id":"hc47k8jw"}
3
+ {"time":"2025-04-22T12:26:24.812702295-04:00","level":"INFO","msg":"writer: Do: started","stream_id":"hc47k8jw"}
4
+ {"time":"2025-04-22T12:26:24.812796834-04:00","level":"INFO","msg":"sender: started","stream_id":"hc47k8jw"}
5
+ {"time":"2025-04-22T12:26:24.812793588-04:00","level":"INFO","msg":"handler: started","stream_id":"hc47k8jw"}
6
+ {"time":"2025-04-22T12:26:24.812797595-04:00","level":"INFO","msg":"stream: started","id":"hc47k8jw"}
7
+ {"time":"2025-04-22T12:26:25.072522218-04:00","level":"INFO","msg":"Starting system monitor"}
8
+ {"time":"2025-04-22T13:28:37.53429929-04:00","level":"INFO","msg":"api: retrying error","error":"Post \"https://api.wandb.ai/files/gganeshl-carnegie-mellon-university/ESPnet_ESPnetEZDataTask/hc47k8jw/file_stream\": dial tcp 35.186.228.49:443: connect: connection timed out"}
9
+ {"time":"2025-04-22T13:36:00.330987928-04:00","level":"INFO","msg":"Stopping system monitor"}
10
+ {"time":"2025-04-22T13:36:00.33232475-04:00","level":"INFO","msg":"Stopped system monitor"}
11
+ {"time":"2025-04-22T13:36:01.292531969-04:00","level":"INFO","msg":"handler: operation stats","stats":{}}
12
+ {"time":"2025-04-22T13:36:01.461755077-04:00","level":"INFO","msg":"fileTransfer: Close: file transfer manager closed"}
13
+ {"time":"2025-04-22T13:36:03.418249732-04:00","level":"INFO","msg":"stream: closing","id":"hc47k8jw"}
14
+ {"time":"2025-04-22T13:36:03.418297973-04:00","level":"INFO","msg":"handler: closed","stream_id":"hc47k8jw"}
15
+ {"time":"2025-04-22T13:36:03.418318932-04:00","level":"INFO","msg":"writer: Close: closed","stream_id":"hc47k8jw"}
16
+ {"time":"2025-04-22T13:36:03.418339511-04:00","level":"INFO","msg":"sender: closed","stream_id":"hc47k8jw"}
17
+ {"time":"2025-04-22T13:36:03.421317411-04:00","level":"INFO","msg":"stream: closed","id":"hc47k8jw"}
exp/finetune/wandb/latest-run/logs/debug.log ADDED
The diff for this file is too large to render. See raw diff
 
exp/finetune/wandb/latest-run/run-hc47k8jw.wandb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:58cd1587df06b15d93fc7ec1142d48956d866a7a17cb77d87dd2c108790412aa
3
+ size 3785184
exp/finetune/wandb/run-20250422_122624-hc47k8jw/files/config.yaml ADDED
The diff for this file is too large to render. See raw diff
 
exp/finetune/wandb/run-20250422_122624-hc47k8jw/files/output.log ADDED
The diff for this file is too large to render. See raw diff
 
exp/finetune/wandb/run-20250422_122624-hc47k8jw/files/requirements.txt ADDED
@@ -0,0 +1,309 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ monai==1.4.0
2
+ nvidia-nvjitlink-cu12==12.4.127
3
+ jiter==0.8.2
4
+ accelerate==1.3.0
5
+ nvidia-nvtx-cu12==12.4.127
6
+ gitdb==4.0.12
7
+ rfc3986-validator==0.1.1
8
+ proto-plus==1.26.0
9
+ jupyterlab==4.3.5
10
+ prometheus-fastapi-instrumentator==7.0.2
11
+ tomli==2.2.1
12
+ json5==0.10.0
13
+ sniffio==1.3.1
14
+ google-auth==2.38.0
15
+ msgspec==0.19.0
16
+ nvidia-cusparse-cu12==12.3.1.170
17
+ traitlets==5.14.3
18
+ torchvision==0.20.1
19
+ referencing==0.36.2
20
+ astor==0.8.1
21
+ decorator==5.1.1
22
+ importlib_resources==6.4.5
23
+ debugpy==1.8.13
24
+ compressed-tensors==0.9.1
25
+ dill==0.3.8
26
+ evaluate==0.4.3
27
+ pycparser==2.22
28
+ nvidia-nccl-cu12==2.21.5
29
+ mpmath==1.3.0
30
+ ptyprocess==0.7.0
31
+ cffi==1.17.1
32
+ multidict==6.1.0
33
+ bleach==6.2.0
34
+ fastjsonschema==2.21.1
35
+ blake3==1.0.4
36
+ partial-json-parser==0.2.1.1.post5
37
+ jedi==0.19.2
38
+ xgrammar==0.1.11
39
+ nvidia-cusolver-cu12==11.6.1.9
40
+ asttokens==3.0.0
41
+ rfc3339-validator==0.1.4
42
+ aiolimiter==1.2.1
43
+ smart-open==7.1.0
44
+ pyarrow==18.0.0
45
+ torchaudio==2.5.1
46
+ charset-normalizer==3.4.0
47
+ GitPython==3.1.44
48
+ sentencepiece==0.2.0
49
+ nvidia-cuda-runtime-cu12==12.4.127
50
+ huggingface-hub==0.26.2
51
+ pandas==2.2.3
52
+ opencv-python-headless==4.11.0.86
53
+ pyasn1==0.6.1
54
+ msgpack==1.1.0
55
+ frozenlist==1.5.0
56
+ regex==2024.11.6
57
+ pybind11==2.13.6
58
+ smmap==5.0.2
59
+ xxhash==3.5.0
60
+ llvmlite==0.43.0
61
+ uri-template==1.3.0
62
+ multiprocess==0.70.16
63
+ argon2-cffi-bindings==21.2.0
64
+ webencodings==0.5.1
65
+ mistune==3.1.2
66
+ webcolors==24.11.1
67
+ filelock==3.16.1
68
+ tenacity==9.0.0
69
+ fonttools==4.55.1
70
+ nvidia-cuda-cupti-cu12==12.4.127
71
+ widgetsnbextension==4.0.13
72
+ aiohappyeyeballs==2.4.3
73
+ sglang==0.4.2.post3
74
+ yarl==1.18.0
75
+ pandocfilters==1.5.1
76
+ iniconfig==2.0.0
77
+ nbconvert==7.16.6
78
+ ipykernel==6.29.5
79
+ jupyter_core==5.7.2
80
+ vllm==0.7.3.dev57+g2ae88905
81
+ notebook==7.3.2
82
+ certifi==2024.8.30
83
+ websockets==14.2
84
+ googleapis-common-protos==1.67.0rc1
85
+ jupyter_client==8.6.3
86
+ safetensors==0.4.5
87
+ jupyter-lsp==2.2.5
88
+ scikit-learn==1.5.2
89
+ kiwisolver==1.4.7
90
+ einops==0.8.0
91
+ eval_type_backport==0.2.2
92
+ notebook_shim==0.2.4
93
+ pyasn1_modules==0.4.1
94
+ wcwidth==0.2.13
95
+ nvidia-cufft-cu12==11.2.1.3
96
+ jupyterlab_widgets==3.0.13
97
+ pyzmq==26.2.1
98
+ pure_eval==0.2.3
99
+ wandb==0.19.6
100
+ docker-pycreds==0.4.0
101
+ executing==2.2.0
102
+ ipython==8.18.1
103
+ distlib==0.3.9
104
+ uvloop==0.21.0
105
+ python-dotenv==1.0.1
106
+ matplotlib==3.9.3
107
+ tiktoken==0.8.0
108
+ requests==2.32.3
109
+ protobuf==5.29.3
110
+ distro==1.9.0
111
+ wrapt==1.17.2
112
+ pydantic==2.10.6
113
+ soupsieve==2.6
114
+ rouge-score==0.1.2
115
+ pycountry==24.6.1
116
+ terminado==0.18.1
117
+ fsspec==2024.9.0
118
+ Send2Trash==1.8.3
119
+ py-cpuinfo==9.0.0
120
+ sentry-sdk==2.20.0
121
+ scipy==1.13.1
122
+ transformers==4.48.3
123
+ tokenizers==0.21.0
124
+ rsa==4.9
125
+ jupyter-events==0.12.0
126
+ slicer==0.0.8
127
+ async-lru==2.0.4
128
+ openai==1.60.2
129
+ platformdirs==4.3.6
130
+ pytest==8.3.4
131
+ bitsandbytes==0.45.5
132
+ async-timeout==5.0.1
133
+ joblib==1.4.2
134
+ tinycss2==1.4.0
135
+ seaborn==0.13.2
136
+ fqdn==1.5.1
137
+ isoduration==20.11.0
138
+ httpx==0.28.1
139
+ numba==0.60.0
140
+ beautifulsoup4==4.13.3
141
+ jupyter_server_terminals==0.5.3
142
+ sympy==1.13.1
143
+ ray==2.42.0
144
+ httpcore==1.0.7
145
+ depyf==0.18.0
146
+ overrides==7.7.0
147
+ nbformat==5.10.4
148
+ babel==2.17.0
149
+ argon2-cffi==23.1.0
150
+ nvidia-curand-cu12==10.3.5.147
151
+ python-dateutil==2.9.0.post0
152
+ uvicorn==0.34.0
153
+ aiohttp-cors==0.7.0
154
+ nvidia-cuda-nvrtc-cu12==12.4.127
155
+ starlette==0.45.3
156
+ aiosignal==1.3.1
157
+ shap==0.46.0
158
+ matplotlib-inline==0.1.7
159
+ absl-py==2.1.0
160
+ jupyterlab_server==2.27.3
161
+ cloudpickle==3.1.1
162
+ datasets==3.1.0
163
+ jsonschema-specifications==2024.10.1
164
+ comm==0.2.2
165
+ parso==0.8.4
166
+ httptools==0.6.4
167
+ websocket-client==1.8.0
168
+ nvidia-cublas-cu12==12.4.5.8
169
+ six==1.17.0
170
+ outlines==0.1.11
171
+ google-api-core==2.24.1
172
+ jupyter_server==2.15.0
173
+ cachetools==5.5.1
174
+ annotated-types==0.7.0
175
+ virtualenv==20.29.1
176
+ threadpoolctl==3.5.0
177
+ prompt_toolkit==3.0.50
178
+ torch==2.5.1
179
+ watchfiles==1.0.4
180
+ urllib3==2.3.0
181
+ attrs==24.2.0
182
+ nltk==3.9.1
183
+ propcache==0.2.0
184
+ jsonpointer==3.0.0
185
+ aiohttp==3.11.7
186
+ pillow==11.0.0
187
+ outlines_core==0.1.26
188
+ airportsdata==20241001
189
+ pydantic_core==2.27.2
190
+ defusedxml==0.7.1
191
+ lm-format-enforcer==0.10.9
192
+ opencensus==0.11.4
193
+ python-json-logger==3.2.1
194
+ stack-data==0.6.3
195
+ Pygments==2.19.1
196
+ tqdm==4.67.1
197
+ h11==0.14.0
198
+ pluggy==1.5.0
199
+ triton==3.1.0
200
+ rpds-py==0.22.3
201
+ jupyterlab_pygments==0.3.0
202
+ exceptiongroup==1.2.2
203
+ contourpy==1.3.0
204
+ interegular==0.3.3
205
+ jsonschema==4.23.0
206
+ nvidia-cudnn-cu12==9.1.0.70
207
+ pexpect==4.9.0
208
+ jupyter==1.1.1
209
+ opencensus-context==0.2.dev0
210
+ anyio==4.8.0
211
+ ipywidgets==8.1.5
212
+ mistral_common==1.5.3
213
+ nbclient==0.10.2
214
+ gguf==0.10.0
215
+ nest-asyncio==1.6.0
216
+ xformers==0.0.28.post3
217
+ jupyter-console==6.6.3
218
+ tornado==6.4.2
219
+ packaging==24.2
220
+ arrow==1.3.0
221
+ py-spy==0.4.0
222
+ grpcio==1.70.0
223
+ networkx==3.2.1
224
+ typing_extensions==4.12.2
225
+ lark==1.2.2
226
+ colorful==0.6.0a1
227
+ fastapi==0.115.8
228
+ monai-weekly==1.5.dev2505
229
+ diskcache==5.6.3
230
+ types-python-dateutil==2.9.0.20241206
231
+ setproctitle==1.3.4
232
+ prometheus_client==0.21.1
233
+ cycler==0.12.1
234
+ tzdata==2024.2
235
+ pytz==2025.2
236
+ pyparsing==3.2.3
237
+ Werkzeug==3.1.3
238
+ ctc_segmentation==1.7.4
239
+ espnet-tts-frontend==0.0.3
240
+ audioread==3.0.1
241
+ numpy==1.26.4
242
+ h5py==3.13.0
243
+ idna==3.10
244
+ librosa==0.9.2
245
+ zipp==3.21.0
246
+ omegaconf==2.3.0
247
+ lightning==2.5.1
248
+ resampy==0.4.3
249
+ jiwer==3.1.0
250
+ pooch==1.8.2
251
+ jaconv==0.4.0
252
+ pypinyin==0.44.0
253
+ g2p-en==2.1.0
254
+ asteroid-filterbanks==0.4.0
255
+ PySocks==1.7.1
256
+ ci_sdr==0.0.2
257
+ fast_bss_eval==0.1.3
258
+ Markdown==3.8
259
+ hydra-core==1.3.2
260
+ espnet==202503
261
+ setuptools==73.0.1
262
+ espnet-model-zoo==0.1.7
263
+ PyYAML==6.0.2
264
+ torch-complex==0.4.4
265
+ ConfigArgParse==1.7
266
+ Unidecode==1.3.8
267
+ soundfile==0.13.1
268
+ jamo==0.4.1
269
+ Cython==3.0.12
270
+ kaldiio==2.18.1
271
+ editdistance==0.8.1
272
+ humanfriendly==10.0
273
+ psutil==7.0.0
274
+ Distance==0.1.3
275
+ tensorboard-data-server==0.7.2
276
+ click==8.1.8
277
+ gdown==5.2.0
278
+ typeguard==4.4.2
279
+ lightning-utilities==0.14.3
280
+ MarkupSafe==3.0.2
281
+ pyworld==0.3.5
282
+ pytorch-lightning==2.5.1
283
+ pip==25.0
284
+ torchmetrics==1.7.1
285
+ tensorboard==2.19.0
286
+ more-itertools==10.6.0
287
+ Jinja2==3.1.6
288
+ importlib-metadata==4.13.0
289
+ RapidFuzz==3.13.0
290
+ wheel==0.45.1
291
+ antlr4-python3-runtime==4.9.3
292
+ inflect==7.5.0
293
+ opt_einsum==3.4.0
294
+ importlib_resources==6.4.0
295
+ wheel==0.43.0
296
+ backports.tarfile==1.2.0
297
+ autocommand==2.2.2
298
+ inflect==7.3.1
299
+ typing_extensions==4.12.2
300
+ jaraco.text==3.12.1
301
+ zipp==3.19.2
302
+ more-itertools==10.3.0
303
+ platformdirs==4.2.2
304
+ packaging==24.1
305
+ typeguard==4.3.0
306
+ jaraco.functools==4.0.1
307
+ jaraco.context==5.3.0
308
+ importlib_metadata==8.0.0
309
+ tomli==2.0.1
exp/finetune/wandb/run-20250422_122624-hc47k8jw/files/wandb-metadata.json ADDED
@@ -0,0 +1,98 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "os": "Linux-5.14.0-427.42.1.el9_4.x86_64-x86_64-with-glibc2.34",
3
+ "python": "CPython 3.9.6",
4
+ "startedAt": "2025-04-22T16:26:24.696915Z",
5
+ "program": "/home/gganeshl/speech/asr_ft.py",
6
+ "codePath": "asr_ft.py",
7
+ "email": "gganeshl@andrew.cmu.edu",
8
+ "root": "exp/finetune",
9
+ "host": "babel-4-33",
10
+ "executable": "/home/gganeshl/miniconda3/envs/speech/bin/python",
11
+ "codePathLocal": "asr_ft.py",
12
+ "cpu_count": 32,
13
+ "cpu_count_logical": 64,
14
+ "gpu": "NVIDIA RTX A6000",
15
+ "gpu_count": 1,
16
+ "disk": {
17
+ "/": {
18
+ "total": "236663791616",
19
+ "used": "88929800192"
20
+ }
21
+ },
22
+ "memory": {
23
+ "total": "540351823872"
24
+ },
25
+ "cpu": {
26
+ "count": 32,
27
+ "countLogical": 64
28
+ },
29
+ "gpu_nvidia": [
30
+ {
31
+ "name": "NVIDIA RTX A6000",
32
+ "memoryTotal": "51527024640",
33
+ "cudaCores": 10752,
34
+ "architecture": "Ampere"
35
+ }
36
+ ],
37
+ "slurm": {
38
+ "cluster_name": "babel",
39
+ "conf": "/var/spool/slurmd/conf-cache/slurm.conf",
40
+ "cpu_bind": "quiet,mask_cpu:0x00000001000002000000000100000200",
41
+ "cpu_bind_list": "0x00000001000002000000000100000200",
42
+ "cpu_bind_type": "mask_cpu:",
43
+ "cpu_bind_verbose": "quiet",
44
+ "cpus_on_node": "4",
45
+ "cpus_per_task": "1",
46
+ "distribution": "cyclic,pack",
47
+ "gpus_on_node": "1",
48
+ "gtids": "0",
49
+ "job_account": "cp3a",
50
+ "job_cpus_per_node": "4",
51
+ "job_end_time": "1745511243",
52
+ "job_gid": "2709914",
53
+ "job_gpus": "5",
54
+ "job_group": "gganeshl",
55
+ "job_id": "4624778",
56
+ "job_name": "speech_asr",
57
+ "job_nodelist": "babel-4-33",
58
+ "job_num_nodes": "1",
59
+ "job_partition": "general",
60
+ "job_qos": "normal",
61
+ "job_start_time": "1745338743",
62
+ "job_uid": "2709914",
63
+ "job_user": "gganeshl",
64
+ "jobid": "4624778",
65
+ "launch_node_ipaddr": "172.16.1.5",
66
+ "localid": "0",
67
+ "mem_per_node": "32768",
68
+ "nnodes": "1",
69
+ "nodeid": "0",
70
+ "nodelist": "babel-4-33",
71
+ "nprocs": "1",
72
+ "ntasks": "1",
73
+ "prio_process": "0",
74
+ "procid": "0",
75
+ "pty_port": "37249",
76
+ "pty_win_col": "134",
77
+ "pty_win_row": "25",
78
+ "script_context": "prolog_task",
79
+ "srun_comm_host": "172.16.1.5",
80
+ "srun_comm_port": "32883",
81
+ "step_gpus": "3",
82
+ "step_id": "0",
83
+ "step_launcher_port": "32883",
84
+ "step_nodelist": "babel-6-21",
85
+ "step_num_nodes": "1",
86
+ "step_num_tasks": "1",
87
+ "step_tasks_per_node": "1",
88
+ "stepid": "0",
89
+ "submit_dir": "/home/gganeshl/speech",
90
+ "submit_host": "babel-6-21",
91
+ "task_pid": "3934047",
92
+ "tasks_per_node": "4",
93
+ "topology_addr": "babel-4-33",
94
+ "topology_addr_pattern": "node",
95
+ "tres_per_task": "cpu=4"
96
+ },
97
+ "cudaVersion": "12.6"
98
+ }
exp/finetune/wandb/run-20250422_122624-hc47k8jw/files/wandb-summary.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"train/train_loss_ctc_epoch":8.894595814524148,"train/train_time":8.246310259787366,"_runtime":4169.55117254,"train/train_loss_epoch":1.2184081917391751,"train/train_forward_time_epoch":0.05445344569766894,"valid/valid_acc_epoch":0.8508969873189927,"train/train_optim_step_time_epoch":0.006779810173436999,"valid/valid_loss_ctc_epoch":21.5558465385437,"valid/valid_cer_ctc_epoch":0.12294024802744388,"epoch":4,"train/train_loss_att_epoch":3.15036278873905,"metrics/loss_scale":1,"metrics/iter_time":1.937151960558258,"train/train_backward_time_epoch":0.05791670780628919,"_timestamp":1.7453433542476704e+09,"train/train_iter_time_epoch":1.937151960558258,"metrics/clip":100,"train/train_clip_epoch":100,"metrics/loss":1.2184081917391751,"_step":203,"train/train_gpu_max_cached_mem_GB_epoch":10.736328125,"train/train_grad_norm_epoch":34.59821179962158,"metrics/optim_step_time":0.006779810173436999,"iteration":2000,"train/train_optim0_lr0_epoch":0.00010000000000000003,"valid/valid_loss_epoch":17.118548574447633,"train/train_train_time_epoch":8.246310259787366,"train/train_loss_scale_epoch":1,"valid/valid_wer_epoch":0.9300000011920929,"metrics/loss_ctc":8.894595814524148,"valid/valid_loss_att_epoch":15.216849050521851,"train/train_acc_epoch":0.9732946280975294,"metrics/optim0_lr0":0.00010000000000000003,"metrics/backward_time":0.05791670780628919,"valid/valid_gpu_max_cached_mem_GB_epoch":10.736328125,"metrics/forward_time":0.05445344569766894,"metrics/loss_att":3.15036278873905,"valid/valid_cer_epoch":0.09997057549655437,"_wandb":{"runtime":4175},"metrics/grad_norm":34.59821179962158,"metrics/acc":0.9732946280975294}
exp/finetune/wandb/run-20250422_122624-hc47k8jw/logs/debug-core.log ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"time":"2025-04-22T12:26:24.672048644-04:00","level":"INFO","msg":"main: starting server","port-filename":"/tmp/tmpy0krtiml/port-3934128.txt","pid":3934128,"log-level":0,"disable-analytics":false,"shutdown-on-parent-exit":false}
2
+ {"time":"2025-04-22T12:26:24.673781794-04:00","level":"INFO","msg":"Will exit if parent process dies.","ppid":3934128}
3
+ {"time":"2025-04-22T12:26:24.6746965-04:00","level":"INFO","msg":"server is running","addr":{"IP":"127.0.0.1","Port":45355,"Zone":""}}
4
+ {"time":"2025-04-22T12:26:24.688179568-04:00","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"127.0.0.1:53318"}
5
+ {"time":"2025-04-22T12:26:24.699245967-04:00","level":"INFO","msg":"handleInformInit: received","streamId":"hc47k8jw","id":"127.0.0.1:53318"}
6
+ {"time":"2025-04-22T12:26:24.812922771-04:00","level":"INFO","msg":"handleInformInit: stream started","streamId":"hc47k8jw","id":"127.0.0.1:53318"}
7
+ {"time":"2025-04-22T13:36:03.417988098-04:00","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"hc47k8jw","id":"127.0.0.1:53318"}
8
+ {"time":"2025-04-22T13:36:03.4213275-04:00","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"hc47k8jw","id":"127.0.0.1:53318"}
9
+ {"time":"2025-04-22T13:36:03.458763144-04:00","level":"INFO","msg":"handleInformTeardown: server teardown initiated","id":"127.0.0.1:53318"}
10
+ {"time":"2025-04-22T13:36:03.458778322-04:00","level":"INFO","msg":"handleInformTeardown: server shutdown complete","id":"127.0.0.1:53318"}
11
+ {"time":"2025-04-22T13:36:03.458788401-04:00","level":"INFO","msg":"server is shutting down"}
12
+ {"time":"2025-04-22T13:36:03.458799733-04:00","level":"INFO","msg":"connection: closing","id":"127.0.0.1:53318"}
13
+ {"time":"2025-04-22T13:36:03.458922374-04:00","level":"INFO","msg":"connection: closed successfully","id":"127.0.0.1:53318"}
14
+ {"time":"2025-04-22T13:36:03.458959755-04:00","level":"INFO","msg":"connection: ManageConnectionData: connection closed","id":"127.0.0.1:53318"}
15
+ {"time":"2025-04-22T13:36:03.458966237-04:00","level":"INFO","msg":"server is closed"}
exp/finetune/wandb/run-20250422_122624-hc47k8jw/logs/debug-internal.log ADDED
@@ -0,0 +1,17 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"time":"2025-04-22T12:26:24.699996553-04:00","level":"INFO","msg":"stream: starting","core version":"0.19.6","symlink path":"exp/finetune/wandb/run-20250422_122624-hc47k8jw/logs/debug-core.log"}
2
+ {"time":"2025-04-22T12:26:24.81223292-04:00","level":"INFO","msg":"created new stream","id":"hc47k8jw"}
3
+ {"time":"2025-04-22T12:26:24.812702295-04:00","level":"INFO","msg":"writer: Do: started","stream_id":"hc47k8jw"}
4
+ {"time":"2025-04-22T12:26:24.812796834-04:00","level":"INFO","msg":"sender: started","stream_id":"hc47k8jw"}
5
+ {"time":"2025-04-22T12:26:24.812793588-04:00","level":"INFO","msg":"handler: started","stream_id":"hc47k8jw"}
6
+ {"time":"2025-04-22T12:26:24.812797595-04:00","level":"INFO","msg":"stream: started","id":"hc47k8jw"}
7
+ {"time":"2025-04-22T12:26:25.072522218-04:00","level":"INFO","msg":"Starting system monitor"}
8
+ {"time":"2025-04-22T13:28:37.53429929-04:00","level":"INFO","msg":"api: retrying error","error":"Post \"https://api.wandb.ai/files/gganeshl-carnegie-mellon-university/ESPnet_ESPnetEZDataTask/hc47k8jw/file_stream\": dial tcp 35.186.228.49:443: connect: connection timed out"}
9
+ {"time":"2025-04-22T13:36:00.330987928-04:00","level":"INFO","msg":"Stopping system monitor"}
10
+ {"time":"2025-04-22T13:36:00.33232475-04:00","level":"INFO","msg":"Stopped system monitor"}
11
+ {"time":"2025-04-22T13:36:01.292531969-04:00","level":"INFO","msg":"handler: operation stats","stats":{}}
12
+ {"time":"2025-04-22T13:36:01.461755077-04:00","level":"INFO","msg":"fileTransfer: Close: file transfer manager closed"}
13
+ {"time":"2025-04-22T13:36:03.418249732-04:00","level":"INFO","msg":"stream: closing","id":"hc47k8jw"}
14
+ {"time":"2025-04-22T13:36:03.418297973-04:00","level":"INFO","msg":"handler: closed","stream_id":"hc47k8jw"}
15
+ {"time":"2025-04-22T13:36:03.418318932-04:00","level":"INFO","msg":"writer: Close: closed","stream_id":"hc47k8jw"}
16
+ {"time":"2025-04-22T13:36:03.418339511-04:00","level":"INFO","msg":"sender: closed","stream_id":"hc47k8jw"}
17
+ {"time":"2025-04-22T13:36:03.421317411-04:00","level":"INFO","msg":"stream: closed","id":"hc47k8jw"}
exp/finetune/wandb/run-20250422_122624-hc47k8jw/logs/debug.log ADDED
The diff for this file is too large to render. See raw diff