Commit
•
166c885
1
Parent(s):
7c14ed3
final training
Browse files- all_results.json +15 -15
- eval_pretrained_results.json +3 -3
- pytorch_model.bin +1 -1
- requirements.txt +199 -4
- test_finetuned_results.json +5 -5
- test_pretrained_results.json +3 -3
- train_results.json +4 -4
- trainer_state.json +275 -275
- training_args.bin +1 -1
all_results.json
CHANGED
|
@@ -2,31 +2,31 @@
|
|
| 2 |
"epoch": 2.32,
|
| 3 |
"eval_loss": 1.6191972494125366,
|
| 4 |
"eval_pretrained_loss": 1.710707187652588,
|
| 5 |
-
"eval_pretrained_runtime":
|
| 6 |
-
"eval_pretrained_samples_per_second": 2.
|
| 7 |
-
"eval_pretrained_steps_per_second": 0.
|
| 8 |
"eval_pretrained_wer": 258.2308797700783,
|
| 9 |
"eval_runtime": 56.3363,
|
| 10 |
"eval_samples_per_second": 0.071,
|
| 11 |
"eval_steps_per_second": 0.036,
|
| 12 |
"eval_wer": 153.2258064516129,
|
| 13 |
-
"test_finetuned_loss": 1.
|
| 14 |
-
"test_finetuned_runtime":
|
| 15 |
-
"test_finetuned_samples_per_second": 4.
|
| 16 |
-
"test_finetuned_steps_per_second": 0.
|
| 17 |
-
"test_finetuned_wer":
|
| 18 |
"test_loss": 1.7568330764770508,
|
| 19 |
"test_pretrained_loss": 1.723103642463684,
|
| 20 |
-
"test_pretrained_runtime":
|
| 21 |
-
"test_pretrained_samples_per_second":
|
| 22 |
-
"test_pretrained_steps_per_second": 0.
|
| 23 |
"test_pretrained_wer": 261.9265750514506,
|
| 24 |
"test_runtime": 37.8582,
|
| 25 |
"test_samples_per_second": 0.106,
|
| 26 |
"test_steps_per_second": 0.053,
|
| 27 |
"test_wer": 138.5964912280702,
|
| 28 |
-
"train_loss":
|
| 29 |
-
"train_runtime":
|
| 30 |
-
"train_samples_per_second": 4.
|
| 31 |
-
"train_steps_per_second": 0.
|
| 32 |
}
|
|
|
|
| 2 |
"epoch": 2.32,
|
| 3 |
"eval_loss": 1.6191972494125366,
|
| 4 |
"eval_pretrained_loss": 1.710707187652588,
|
| 5 |
+
"eval_pretrained_runtime": 125.9444,
|
| 6 |
+
"eval_pretrained_samples_per_second": 2.62,
|
| 7 |
+
"eval_pretrained_steps_per_second": 0.087,
|
| 8 |
"eval_pretrained_wer": 258.2308797700783,
|
| 9 |
"eval_runtime": 56.3363,
|
| 10 |
"eval_samples_per_second": 0.071,
|
| 11 |
"eval_steps_per_second": 0.036,
|
| 12 |
"eval_wer": 153.2258064516129,
|
| 13 |
+
"test_finetuned_loss": 1.0840972661972046,
|
| 14 |
+
"test_finetuned_runtime": 169.6782,
|
| 15 |
+
"test_finetuned_samples_per_second": 4.473,
|
| 16 |
+
"test_finetuned_steps_per_second": 0.141,
|
| 17 |
+
"test_finetuned_wer": 172.4357697669787,
|
| 18 |
"test_loss": 1.7568330764770508,
|
| 19 |
"test_pretrained_loss": 1.723103642463684,
|
| 20 |
+
"test_pretrained_runtime": 196.5873,
|
| 21 |
+
"test_pretrained_samples_per_second": 3.861,
|
| 22 |
+
"test_pretrained_steps_per_second": 0.122,
|
| 23 |
"test_pretrained_wer": 261.9265750514506,
|
| 24 |
"test_runtime": 37.8582,
|
| 25 |
"test_samples_per_second": 0.106,
|
| 26 |
"test_steps_per_second": 0.053,
|
| 27 |
"test_wer": 138.5964912280702,
|
| 28 |
+
"train_loss": 1.0522634956453527,
|
| 29 |
+
"train_runtime": 1735.2585,
|
| 30 |
+
"train_samples_per_second": 4.131,
|
| 31 |
+
"train_steps_per_second": 0.065
|
| 32 |
}
|
eval_pretrained_results.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
"eval_pretrained_loss": 1.710707187652588,
|
| 3 |
-
"eval_pretrained_runtime":
|
| 4 |
-
"eval_pretrained_samples_per_second": 2.
|
| 5 |
-
"eval_pretrained_steps_per_second": 0.
|
| 6 |
"eval_pretrained_wer": 258.2308797700783
|
| 7 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"eval_pretrained_loss": 1.710707187652588,
|
| 3 |
+
"eval_pretrained_runtime": 125.9444,
|
| 4 |
+
"eval_pretrained_samples_per_second": 2.62,
|
| 5 |
+
"eval_pretrained_steps_per_second": 0.087,
|
| 6 |
"eval_pretrained_wer": 258.2308797700783
|
| 7 |
}
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 151098921
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2027faf2910d36a40b249965d707ab19532ef0effc8d243d184de4439275106e
|
| 3 |
size 151098921
|
requirements.txt
CHANGED
|
@@ -1,100 +1,295 @@
|
|
| 1 |
absl-py==1.3.0
|
| 2 |
aiohttp==3.8.3
|
| 3 |
aiosignal==1.3.1
|
|
|
|
| 4 |
appdirs==1.4.4
|
|
|
|
|
|
|
|
|
|
| 5 |
async-timeout==4.0.2
|
|
|
|
| 6 |
attrs==22.1.0
|
| 7 |
audioread==3.0.0
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 8 |
boto3==1.26.27
|
| 9 |
botocore==1.29.27
|
|
|
|
| 10 |
cachetools==5.2.0
|
|
|
|
| 11 |
certifi==2022.12.7
|
| 12 |
cffi==1.15.1
|
|
|
|
| 13 |
charset-normalizer==2.1.1
|
| 14 |
click==8.1.3
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 15 |
contextlib2==21.6.0
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 16 |
datasets @ git+https://github.com/huggingface/datasets@6338ded243ccb495b53b1996cba0847bdc250aba
|
|
|
|
| 17 |
decorator==5.1.1
|
|
|
|
| 18 |
dill==0.3.6
|
|
|
|
|
|
|
|
|
|
|
|
|
| 19 |
docker-pycreds==0.4.0
|
|
|
|
|
|
|
| 20 |
evaluate==0.3.0
|
|
|
|
| 21 |
filelock==3.8.2
|
|
|
|
|
|
|
|
|
|
| 22 |
frozenlist==1.3.3
|
| 23 |
fsspec==2022.11.0
|
|
|
|
|
|
|
| 24 |
gitdb==4.0.10
|
| 25 |
GitPython==3.1.29
|
|
|
|
| 26 |
google-auth==2.15.0
|
| 27 |
google-auth-oauthlib==0.4.6
|
| 28 |
google-pasta==0.2.0
|
| 29 |
grpcio==1.51.1
|
|
|
|
|
|
|
|
|
|
|
|
|
| 30 |
huggingface-hub==0.11.1
|
|
|
|
|
|
|
| 31 |
idna==3.4
|
|
|
|
|
|
|
| 32 |
importlib-metadata==4.13.0
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 33 |
jiwer==2.5.1
|
| 34 |
jmespath==1.0.1
|
| 35 |
joblib==1.2.0
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 36 |
Levenshtein==0.20.2
|
| 37 |
librosa==0.9.2
|
|
|
|
| 38 |
llvmlite==0.39.1
|
|
|
|
|
|
|
|
|
|
| 39 |
Markdown==3.4.1
|
| 40 |
MarkupSafe==2.1.1
|
|
|
|
|
|
|
|
|
|
|
|
|
| 41 |
more-itertools==9.0.0
|
|
|
|
| 42 |
multidict==6.0.3
|
|
|
|
| 43 |
multiprocess==0.70.14
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 44 |
numba==0.56.4
|
|
|
|
| 45 |
numpy==1.23.5
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 46 |
oauthlib==3.2.2
|
|
|
|
|
|
|
|
|
|
| 47 |
packaging==20.9
|
| 48 |
pandas==1.5.2
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 49 |
pathos==0.3.0
|
| 50 |
pathtools==0.1.2
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 51 |
pooch==1.6.0
|
| 52 |
pox==0.3.2
|
| 53 |
ppft==1.7.6.6
|
|
|
|
| 54 |
promise==2.3
|
|
|
|
| 55 |
protobuf==3.20.3
|
| 56 |
protobuf3-to-dict==0.1.5
|
| 57 |
psutil==5.9.4
|
|
|
|
|
|
|
| 58 |
pyarrow==10.0.1
|
| 59 |
pyasn1==0.4.8
|
| 60 |
pyasn1-modules==0.2.8
|
|
|
|
| 61 |
pycparser==2.21
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 62 |
pyparsing==3.0.9
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 63 |
python-dateutil==2.8.2
|
|
|
|
|
|
|
| 64 |
pytz==2022.6
|
|
|
|
| 65 |
PyYAML==6.0
|
|
|
|
| 66 |
rapidfuzz==2.13.4
|
| 67 |
regex==2022.10.31
|
| 68 |
requests==2.28.1
|
| 69 |
requests-oauthlib==1.3.1
|
|
|
|
| 70 |
resampy==0.4.2
|
| 71 |
responses==0.18.0
|
| 72 |
rsa==4.9
|
| 73 |
s3transfer==0.6.0
|
| 74 |
sagemaker==2.121.1
|
| 75 |
schema==0.7.5
|
|
|
|
|
|
|
| 76 |
scikit-learn==1.2.0
|
| 77 |
scipy==1.9.3
|
|
|
|
|
|
|
|
|
|
| 78 |
sentry-sdk==1.11.1
|
|
|
|
| 79 |
setproctitle==1.3.2
|
| 80 |
shortuuid==1.0.11
|
|
|
|
| 81 |
six==1.16.0
|
| 82 |
smdebug-rulesconfig==1.0.1
|
| 83 |
smmap==5.0.0
|
|
|
|
|
|
|
| 84 |
soundfile==0.11.0
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 85 |
tensorboard==2.11.0
|
| 86 |
tensorboard-data-server==0.6.1
|
| 87 |
tensorboard-plugin-wit==1.8.1
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 88 |
threadpoolctl==3.1.0
|
|
|
|
|
|
|
| 89 |
tokenizers==0.13.2
|
| 90 |
-
|
| 91 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 92 |
tqdm==4.64.1
|
|
|
|
| 93 |
transformers @ git+https://github.com/huggingface/transformers@9a6c6ef97fa5df4b1fb8dbc9e8c10ee3a9ed7e2a
|
| 94 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 95 |
urllib3==1.26.13
|
|
|
|
|
|
|
|
|
|
| 96 |
wandb==0.13.6
|
|
|
|
|
|
|
|
|
|
| 97 |
Werkzeug==2.2.2
|
|
|
|
|
|
|
|
|
|
|
|
|
| 98 |
xxhash==3.1.0
|
| 99 |
yarl==1.8.2
|
| 100 |
-
zipp==3.11.0
|
|
|
|
|
|
| 1 |
absl-py==1.3.0
|
| 2 |
aiohttp==3.8.3
|
| 3 |
aiosignal==1.3.1
|
| 4 |
+
anyio==3.6.2
|
| 5 |
appdirs==1.4.4
|
| 6 |
+
argon2-cffi==21.3.0
|
| 7 |
+
argon2-cffi-bindings==21.2.0
|
| 8 |
+
astunparse==1.6.2
|
| 9 |
async-timeout==4.0.2
|
| 10 |
+
atomicwrites==1.1.5
|
| 11 |
attrs==22.1.0
|
| 12 |
audioread==3.0.0
|
| 13 |
+
Automat==0.8.0
|
| 14 |
+
Babel==2.10.3
|
| 15 |
+
backcall==0.1.0
|
| 16 |
+
beautifulsoup4==4.8.2
|
| 17 |
+
bleach==3.1.1
|
| 18 |
+
blinker==1.4
|
| 19 |
+
blosc==1.7.0
|
| 20 |
boto3==1.26.27
|
| 21 |
botocore==1.29.27
|
| 22 |
+
bottle==0.12.15
|
| 23 |
cachetools==5.2.0
|
| 24 |
+
caffe==1.0.0
|
| 25 |
certifi==2022.12.7
|
| 26 |
cffi==1.15.1
|
| 27 |
+
chardet==3.0.4
|
| 28 |
charset-normalizer==2.1.1
|
| 29 |
click==8.1.3
|
| 30 |
+
cloud-init==22.3.4
|
| 31 |
+
cloudpickle==1.3.0
|
| 32 |
+
colorama==0.4.3
|
| 33 |
+
command-not-found==0.3
|
| 34 |
+
configobj==5.0.6
|
| 35 |
+
constantly==15.1.0
|
| 36 |
contextlib2==21.6.0
|
| 37 |
+
cryptography==2.8
|
| 38 |
+
ctop==1.0.0
|
| 39 |
+
cycler==0.10.0
|
| 40 |
+
Cython==0.29.14
|
| 41 |
+
dask==2.8.1+dfsg
|
| 42 |
datasets @ git+https://github.com/huggingface/datasets@6338ded243ccb495b53b1996cba0847bdc250aba
|
| 43 |
+
dbus-python==1.2.16
|
| 44 |
decorator==5.1.1
|
| 45 |
+
defusedxml==0.6.0
|
| 46 |
dill==0.3.6
|
| 47 |
+
distlib==0.3.0
|
| 48 |
+
distro==1.4.0
|
| 49 |
+
distro-info===0.23ubuntu1
|
| 50 |
+
docker==4.1.0
|
| 51 |
docker-pycreds==0.4.0
|
| 52 |
+
entrypoints==0.3
|
| 53 |
+
et-xmlfile==1.0.1
|
| 54 |
evaluate==0.3.0
|
| 55 |
+
fastjsonschema==2.16.2
|
| 56 |
filelock==3.8.2
|
| 57 |
+
flake8==3.7.9
|
| 58 |
+
flatbuffers==1.12
|
| 59 |
+
fonttools==4.38.0
|
| 60 |
frozenlist==1.3.3
|
| 61 |
fsspec==2022.11.0
|
| 62 |
+
future==0.18.2
|
| 63 |
+
gast==0.4.0
|
| 64 |
gitdb==4.0.10
|
| 65 |
GitPython==3.1.29
|
| 66 |
+
Glances==3.1.3
|
| 67 |
google-auth==2.15.0
|
| 68 |
google-auth-oauthlib==0.4.6
|
| 69 |
google-pasta==0.2.0
|
| 70 |
grpcio==1.51.1
|
| 71 |
+
h5py==2.10.0
|
| 72 |
+
html5lib==1.0.1
|
| 73 |
+
htmlmin==0.1.12
|
| 74 |
+
httplib2==0.14.0
|
| 75 |
huggingface-hub==0.11.1
|
| 76 |
+
hyperlink==19.0.0
|
| 77 |
+
icdiff==1.9.5
|
| 78 |
idna==3.4
|
| 79 |
+
ImageHash==4.3.1
|
| 80 |
+
imageio==2.4.1
|
| 81 |
importlib-metadata==4.13.0
|
| 82 |
+
incremental==16.10.1
|
| 83 |
+
influxdb==5.2.0
|
| 84 |
+
iotop==0.6
|
| 85 |
+
ipykernel==5.2.0
|
| 86 |
+
ipython==7.13.0
|
| 87 |
+
ipython_genutils==0.2.0
|
| 88 |
+
ipywidgets==8.0.2
|
| 89 |
+
jdcal==1.0
|
| 90 |
+
jedi==0.15.2
|
| 91 |
+
Jinja2==3.1.2
|
| 92 |
jiwer==2.5.1
|
| 93 |
jmespath==1.0.1
|
| 94 |
joblib==1.2.0
|
| 95 |
+
json5==0.9.10
|
| 96 |
+
jsonpatch==1.22
|
| 97 |
+
jsonpointer==2.0
|
| 98 |
+
jsonschema==3.2.0
|
| 99 |
+
jupyter-console==6.0.0
|
| 100 |
+
jupyter-server==1.21.0
|
| 101 |
+
jupyter_client==7.4.4
|
| 102 |
+
jupyter_core==4.11.2
|
| 103 |
+
jupyterlab==3.5.0
|
| 104 |
+
jupyterlab-pygments==0.2.2
|
| 105 |
+
jupyterlab-widgets==3.0.3
|
| 106 |
+
jupyterlab_server==2.16.1
|
| 107 |
+
kaptan==0.5.10
|
| 108 |
+
keras==2.9.0
|
| 109 |
+
Keras-Preprocessing==1.1.2
|
| 110 |
+
keyring==18.0.1
|
| 111 |
+
kiwisolver==1.0.1
|
| 112 |
+
language-selector==0.1
|
| 113 |
+
launchpadlib==1.10.13
|
| 114 |
+
lazr.restfulclient==0.14.2
|
| 115 |
+
lazr.uri==1.0.3
|
| 116 |
Levenshtein==0.20.2
|
| 117 |
librosa==0.9.2
|
| 118 |
+
libtmux==0.8.2
|
| 119 |
llvmlite==0.39.1
|
| 120 |
+
locket==0.2.0
|
| 121 |
+
lxml==4.5.0
|
| 122 |
+
Mako==1.1.0
|
| 123 |
Markdown==3.4.1
|
| 124 |
MarkupSafe==2.1.1
|
| 125 |
+
matplotlib==3.5.3
|
| 126 |
+
mccabe==0.6.1
|
| 127 |
+
missingno==0.5.1
|
| 128 |
+
mistune==2.0.4
|
| 129 |
more-itertools==9.0.0
|
| 130 |
+
mpi4py==3.0.3
|
| 131 |
multidict==6.0.3
|
| 132 |
+
multimethod==1.9
|
| 133 |
multiprocess==0.70.14
|
| 134 |
+
nbclassic==0.4.7
|
| 135 |
+
nbclient==0.7.0
|
| 136 |
+
nbconvert==7.2.2
|
| 137 |
+
nbformat==5.7.0
|
| 138 |
+
nest-asyncio==1.5.6
|
| 139 |
+
netifaces==0.10.4
|
| 140 |
+
networkx==2.4
|
| 141 |
+
nose==1.3.7
|
| 142 |
+
notebook==6.0.3
|
| 143 |
+
notebook_shim==0.2.0
|
| 144 |
numba==0.56.4
|
| 145 |
+
numexpr==2.7.1
|
| 146 |
numpy==1.23.5
|
| 147 |
+
nvidia-cublas-cu11==11.10.3.66
|
| 148 |
+
nvidia-cuda-nvrtc-cu11==11.7.99
|
| 149 |
+
nvidia-cuda-runtime-cu11==11.7.99
|
| 150 |
+
nvidia-cudnn-cu11==8.5.0.96
|
| 151 |
+
nvidia-ml-py==7.352.0
|
| 152 |
oauthlib==3.2.2
|
| 153 |
+
olefile==0.46
|
| 154 |
+
openpyxl==3.0.3
|
| 155 |
+
opt-einsum==3.3.0
|
| 156 |
packaging==20.9
|
| 157 |
pandas==1.5.2
|
| 158 |
+
pandas-profiling==3.4.0
|
| 159 |
+
pandocfilters==1.4.2
|
| 160 |
+
parameterized==0.7.0
|
| 161 |
+
parso==0.5.2
|
| 162 |
+
partd==1.0.0
|
| 163 |
pathos==0.3.0
|
| 164 |
pathtools==0.1.2
|
| 165 |
+
patsy==0.5.3
|
| 166 |
+
pexpect==4.6.0
|
| 167 |
+
phik==0.12.2
|
| 168 |
+
pickleshare==0.7.5
|
| 169 |
+
Pillow==7.0.0
|
| 170 |
+
pluggy==0.13.0
|
| 171 |
+
ply==3.11
|
| 172 |
pooch==1.6.0
|
| 173 |
pox==0.3.2
|
| 174 |
ppft==1.7.6.6
|
| 175 |
+
prometheus-client==0.7.1
|
| 176 |
promise==2.3
|
| 177 |
+
prompt-toolkit==2.0.10
|
| 178 |
protobuf==3.20.3
|
| 179 |
protobuf3-to-dict==0.1.5
|
| 180 |
psutil==5.9.4
|
| 181 |
+
ptyprocess==0.7.0
|
| 182 |
+
py==1.8.1
|
| 183 |
pyarrow==10.0.1
|
| 184 |
pyasn1==0.4.8
|
| 185 |
pyasn1-modules==0.2.8
|
| 186 |
+
pycodestyle==2.5.0
|
| 187 |
pycparser==2.21
|
| 188 |
+
pycryptodomex==3.6.1
|
| 189 |
+
pycuda==2019.1.2
|
| 190 |
+
pydantic==1.10.2
|
| 191 |
+
pydot==1.4.1
|
| 192 |
+
pyflakes==2.1.1
|
| 193 |
+
Pygments==2.13.0
|
| 194 |
+
PyGObject==3.36.0
|
| 195 |
+
pygpu==0.7.6
|
| 196 |
+
PyHamcrest==1.9.0
|
| 197 |
+
pyinotify==0.9.6
|
| 198 |
+
PyJWT==1.7.1
|
| 199 |
+
pymacaroons==0.13.0
|
| 200 |
+
PyNaCl==1.3.0
|
| 201 |
+
pyOpenSSL==19.0.0
|
| 202 |
pyparsing==3.0.9
|
| 203 |
+
pyrsistent==0.15.5
|
| 204 |
+
pyserial==3.4
|
| 205 |
+
pysmi==0.3.2
|
| 206 |
+
pysnmp==4.4.6
|
| 207 |
+
pystache==0.5.4
|
| 208 |
+
pytest==4.6.9
|
| 209 |
+
python-apt==2.0.0+ubuntu0.20.4.8
|
| 210 |
python-dateutil==2.8.2
|
| 211 |
+
python-debian===0.1.36ubuntu1
|
| 212 |
+
pytools==2019.1.1
|
| 213 |
pytz==2022.6
|
| 214 |
+
PyWavelets==0.5.1
|
| 215 |
PyYAML==6.0
|
| 216 |
+
pyzmq==24.0.1
|
| 217 |
rapidfuzz==2.13.4
|
| 218 |
regex==2022.10.31
|
| 219 |
requests==2.28.1
|
| 220 |
requests-oauthlib==1.3.1
|
| 221 |
+
requests-unixsocket==0.2.0
|
| 222 |
resampy==0.4.2
|
| 223 |
responses==0.18.0
|
| 224 |
rsa==4.9
|
| 225 |
s3transfer==0.6.0
|
| 226 |
sagemaker==2.121.1
|
| 227 |
schema==0.7.5
|
| 228 |
+
scikit-cuda==0.5.3
|
| 229 |
+
scikit-image==0.16.2
|
| 230 |
scikit-learn==1.2.0
|
| 231 |
scipy==1.9.3
|
| 232 |
+
seaborn==0.12.1
|
| 233 |
+
SecretStorage==2.3.1
|
| 234 |
+
Send2Trash==1.8.0
|
| 235 |
sentry-sdk==1.11.1
|
| 236 |
+
service-identity==18.1.0
|
| 237 |
setproctitle==1.3.2
|
| 238 |
shortuuid==1.0.11
|
| 239 |
+
simplejson==3.16.0
|
| 240 |
six==1.16.0
|
| 241 |
smdebug-rulesconfig==1.0.1
|
| 242 |
smmap==5.0.0
|
| 243 |
+
sniffio==1.3.0
|
| 244 |
+
sos==4.4
|
| 245 |
soundfile==0.11.0
|
| 246 |
+
soupsieve==1.9.5
|
| 247 |
+
ssh-import-id==5.10
|
| 248 |
+
statsmodels==0.13.2
|
| 249 |
+
systemd-python==234
|
| 250 |
+
tables==3.6.1
|
| 251 |
+
tangled-up-in-unicode==0.2.0
|
| 252 |
tensorboard==2.11.0
|
| 253 |
tensorboard-data-server==0.6.1
|
| 254 |
tensorboard-plugin-wit==1.8.1
|
| 255 |
+
tensorflow-estimator==2.9.0
|
| 256 |
+
tensorflow-gpu==2.9.1
|
| 257 |
+
termcolor==1.1.0
|
| 258 |
+
terminado==0.17.0
|
| 259 |
+
testpath==0.4.4
|
| 260 |
+
Theano==1.0.4
|
| 261 |
threadpoolctl==3.1.0
|
| 262 |
+
tinycss2==1.2.1
|
| 263 |
+
tmuxp==1.5.4
|
| 264 |
tokenizers==0.13.2
|
| 265 |
+
tomli==2.0.1
|
| 266 |
+
toolz==0.9.0
|
| 267 |
+
torch==1.13.1
|
| 268 |
+
torchaudio==0.13.1
|
| 269 |
+
torchvision==0.13.1
|
| 270 |
+
tornado==6.2
|
| 271 |
tqdm==4.64.1
|
| 272 |
+
traitlets==5.5.0
|
| 273 |
transformers @ git+https://github.com/huggingface/transformers@9a6c6ef97fa5df4b1fb8dbc9e8c10ee3a9ed7e2a
|
| 274 |
+
Twisted==18.9.0
|
| 275 |
+
typing_extensions==4.4.0
|
| 276 |
+
ubuntu-advantage-tools==27.11.2
|
| 277 |
+
ufw==0.36
|
| 278 |
+
unattended-upgrades==0.1
|
| 279 |
urllib3==1.26.13
|
| 280 |
+
virtualenv==20.0.17
|
| 281 |
+
visions==0.7.5
|
| 282 |
+
wadllib==1.3.3
|
| 283 |
wandb==0.13.6
|
| 284 |
+
wcwidth==0.1.8
|
| 285 |
+
webencodings==0.5.1
|
| 286 |
+
websocket-client==0.53.0
|
| 287 |
Werkzeug==2.2.2
|
| 288 |
+
widgetsnbextension==4.0.3
|
| 289 |
+
wrapt==1.11.2
|
| 290 |
+
xlrd==1.1.0
|
| 291 |
+
xlwt==1.3.0
|
| 292 |
xxhash==3.1.0
|
| 293 |
yarl==1.8.2
|
| 294 |
+
zipp==3.11.0
|
| 295 |
+
zope.interface==4.7.1
|
test_finetuned_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 2.32,
|
| 3 |
-
"test_finetuned_loss": 1.
|
| 4 |
-
"test_finetuned_runtime":
|
| 5 |
-
"test_finetuned_samples_per_second": 4.
|
| 6 |
-
"test_finetuned_steps_per_second": 0.
|
| 7 |
-
"test_finetuned_wer":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 2.32,
|
| 3 |
+
"test_finetuned_loss": 1.0840972661972046,
|
| 4 |
+
"test_finetuned_runtime": 169.6782,
|
| 5 |
+
"test_finetuned_samples_per_second": 4.473,
|
| 6 |
+
"test_finetuned_steps_per_second": 0.141,
|
| 7 |
+
"test_finetuned_wer": 172.4357697669787
|
| 8 |
}
|
test_pretrained_results.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
"test_pretrained_loss": 1.723103642463684,
|
| 3 |
-
"test_pretrained_runtime":
|
| 4 |
-
"test_pretrained_samples_per_second":
|
| 5 |
-
"test_pretrained_steps_per_second": 0.
|
| 6 |
"test_pretrained_wer": 261.9265750514506
|
| 7 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"test_pretrained_loss": 1.723103642463684,
|
| 3 |
+
"test_pretrained_runtime": 196.5873,
|
| 4 |
+
"test_pretrained_samples_per_second": 3.861,
|
| 5 |
+
"test_pretrained_steps_per_second": 0.122,
|
| 6 |
"test_pretrained_wer": 261.9265750514506
|
| 7 |
}
|
train_results.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 2.32,
|
| 3 |
-
"train_loss":
|
| 4 |
-
"train_runtime":
|
| 5 |
-
"train_samples_per_second": 4.
|
| 6 |
-
"train_steps_per_second": 0.
|
| 7 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 2.32,
|
| 3 |
+
"train_loss": 1.0522634956453527,
|
| 4 |
+
"train_runtime": 1735.2585,
|
| 5 |
+
"train_samples_per_second": 4.131,
|
| 6 |
+
"train_steps_per_second": 0.065
|
| 7 |
}
|
trainer_state.json
CHANGED
|
@@ -1,5 +1,5 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric":
|
| 3 |
"best_model_checkpoint": "./checkpoint-88",
|
| 4 |
"epoch": 2.3214285714285716,
|
| 5 |
"global_step": 112,
|
|
@@ -15,768 +15,768 @@
|
|
| 15 |
},
|
| 16 |
{
|
| 17 |
"epoch": 0.02,
|
| 18 |
-
"learning_rate":
|
| 19 |
"loss": 1.6564,
|
| 20 |
"step": 2
|
| 21 |
},
|
| 22 |
{
|
| 23 |
"epoch": 0.03,
|
| 24 |
-
"learning_rate":
|
| 25 |
"loss": 1.6518,
|
| 26 |
"step": 3
|
| 27 |
},
|
| 28 |
{
|
| 29 |
"epoch": 0.04,
|
| 30 |
-
"learning_rate":
|
| 31 |
-
"loss": 1.
|
| 32 |
"step": 4
|
| 33 |
},
|
| 34 |
{
|
| 35 |
"epoch": 0.04,
|
| 36 |
-
"learning_rate":
|
| 37 |
-
"loss": 1.
|
| 38 |
"step": 5
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 0.05,
|
| 42 |
-
"learning_rate":
|
| 43 |
-
"loss": 1.
|
| 44 |
"step": 6
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.06,
|
| 48 |
-
"learning_rate":
|
| 49 |
-
"loss": 1.
|
| 50 |
"step": 7
|
| 51 |
},
|
| 52 |
{
|
| 53 |
"epoch": 0.07,
|
| 54 |
-
"learning_rate":
|
| 55 |
-
"loss": 1.
|
| 56 |
"step": 8
|
| 57 |
},
|
| 58 |
{
|
| 59 |
"epoch": 0.08,
|
| 60 |
-
"learning_rate":
|
| 61 |
-
"loss": 1.
|
| 62 |
"step": 9
|
| 63 |
},
|
| 64 |
{
|
| 65 |
"epoch": 0.09,
|
| 66 |
-
"learning_rate":
|
| 67 |
-
"loss": 1.
|
| 68 |
"step": 10
|
| 69 |
},
|
| 70 |
{
|
| 71 |
"epoch": 0.1,
|
| 72 |
-
"learning_rate":
|
| 73 |
-
"loss": 1.
|
| 74 |
"step": 11
|
| 75 |
},
|
| 76 |
{
|
| 77 |
"epoch": 0.1,
|
| 78 |
-
"eval_loss": 1.
|
| 79 |
-
"eval_runtime":
|
| 80 |
-
"eval_samples_per_second":
|
| 81 |
-
"eval_steps_per_second": 0.
|
| 82 |
-
"eval_wer":
|
| 83 |
"step": 11
|
| 84 |
},
|
| 85 |
{
|
| 86 |
"epoch": 0.11,
|
| 87 |
-
"learning_rate":
|
| 88 |
-
"loss": 1.
|
| 89 |
"step": 12
|
| 90 |
},
|
| 91 |
{
|
| 92 |
"epoch": 0.12,
|
| 93 |
-
"learning_rate":
|
| 94 |
-
"loss": 1.
|
| 95 |
"step": 13
|
| 96 |
},
|
| 97 |
{
|
| 98 |
"epoch": 0.12,
|
| 99 |
-
"learning_rate":
|
| 100 |
-
"loss": 1.
|
| 101 |
"step": 14
|
| 102 |
},
|
| 103 |
{
|
| 104 |
"epoch": 0.13,
|
| 105 |
-
"learning_rate":
|
| 106 |
-
"loss": 1.
|
| 107 |
"step": 15
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 0.14,
|
| 111 |
-
"learning_rate":
|
| 112 |
-
"loss": 1.
|
| 113 |
"step": 16
|
| 114 |
},
|
| 115 |
{
|
| 116 |
"epoch": 0.15,
|
| 117 |
-
"learning_rate":
|
| 118 |
-
"loss": 1.
|
| 119 |
"step": 17
|
| 120 |
},
|
| 121 |
{
|
| 122 |
"epoch": 0.16,
|
| 123 |
-
"learning_rate":
|
| 124 |
-
"loss": 1.
|
| 125 |
"step": 18
|
| 126 |
},
|
| 127 |
{
|
| 128 |
"epoch": 0.17,
|
| 129 |
-
"learning_rate":
|
| 130 |
-
"loss": 1.
|
| 131 |
"step": 19
|
| 132 |
},
|
| 133 |
{
|
| 134 |
"epoch": 0.18,
|
| 135 |
-
"learning_rate":
|
| 136 |
-
"loss": 1.
|
| 137 |
"step": 20
|
| 138 |
},
|
| 139 |
{
|
| 140 |
"epoch": 0.19,
|
| 141 |
-
"learning_rate":
|
| 142 |
-
"loss": 1.
|
| 143 |
"step": 21
|
| 144 |
},
|
| 145 |
{
|
| 146 |
"epoch": 0.2,
|
| 147 |
-
"learning_rate":
|
| 148 |
-
"loss": 1.
|
| 149 |
"step": 22
|
| 150 |
},
|
| 151 |
{
|
| 152 |
"epoch": 0.2,
|
| 153 |
-
"eval_loss": 1.
|
| 154 |
-
"eval_runtime":
|
| 155 |
-
"eval_samples_per_second": 2.
|
| 156 |
-
"eval_steps_per_second": 0.
|
| 157 |
-
"eval_wer":
|
| 158 |
"step": 22
|
| 159 |
},
|
| 160 |
{
|
| 161 |
"epoch": 0.21,
|
| 162 |
-
"learning_rate":
|
| 163 |
-
"loss": 1.
|
| 164 |
"step": 23
|
| 165 |
},
|
| 166 |
{
|
| 167 |
"epoch": 0.21,
|
| 168 |
-
"learning_rate":
|
| 169 |
-
"loss": 1.
|
| 170 |
"step": 24
|
| 171 |
},
|
| 172 |
{
|
| 173 |
"epoch": 0.22,
|
| 174 |
-
"learning_rate":
|
| 175 |
-
"loss": 1.
|
| 176 |
"step": 25
|
| 177 |
},
|
| 178 |
{
|
| 179 |
"epoch": 0.23,
|
| 180 |
-
"learning_rate":
|
| 181 |
-
"loss": 1.
|
| 182 |
"step": 26
|
| 183 |
},
|
| 184 |
{
|
| 185 |
"epoch": 0.24,
|
| 186 |
-
"learning_rate":
|
| 187 |
-
"loss": 1.
|
| 188 |
"step": 27
|
| 189 |
},
|
| 190 |
{
|
| 191 |
"epoch": 0.25,
|
| 192 |
-
"learning_rate":
|
| 193 |
-
"loss": 1.
|
| 194 |
"step": 28
|
| 195 |
},
|
| 196 |
{
|
| 197 |
"epoch": 0.26,
|
| 198 |
-
"learning_rate":
|
| 199 |
-
"loss":
|
| 200 |
"step": 29
|
| 201 |
},
|
| 202 |
{
|
| 203 |
"epoch": 0.27,
|
| 204 |
-
"learning_rate":
|
| 205 |
-
"loss": 1.
|
| 206 |
"step": 30
|
| 207 |
},
|
| 208 |
{
|
| 209 |
"epoch": 0.28,
|
| 210 |
-
"learning_rate":
|
| 211 |
-
"loss": 1.
|
| 212 |
"step": 31
|
| 213 |
},
|
| 214 |
{
|
| 215 |
"epoch": 0.29,
|
| 216 |
-
"learning_rate":
|
| 217 |
-
"loss": 1.
|
| 218 |
"step": 32
|
| 219 |
},
|
| 220 |
{
|
| 221 |
"epoch": 0.29,
|
| 222 |
-
"learning_rate":
|
| 223 |
-
"loss": 1.
|
| 224 |
"step": 33
|
| 225 |
},
|
| 226 |
{
|
| 227 |
"epoch": 0.29,
|
| 228 |
-
"eval_loss": 1.
|
| 229 |
-
"eval_runtime":
|
| 230 |
-
"eval_samples_per_second":
|
| 231 |
-
"eval_steps_per_second": 0.
|
| 232 |
-
"eval_wer":
|
| 233 |
"step": 33
|
| 234 |
},
|
| 235 |
{
|
| 236 |
"epoch": 0.3,
|
| 237 |
-
"learning_rate":
|
| 238 |
-
"loss": 1.
|
| 239 |
"step": 34
|
| 240 |
},
|
| 241 |
{
|
| 242 |
"epoch": 0.31,
|
| 243 |
-
"learning_rate":
|
| 244 |
-
"loss": 1.
|
| 245 |
"step": 35
|
| 246 |
},
|
| 247 |
{
|
| 248 |
"epoch": 0.32,
|
| 249 |
-
"learning_rate":
|
| 250 |
-
"loss": 1.
|
| 251 |
"step": 36
|
| 252 |
},
|
| 253 |
{
|
| 254 |
"epoch": 0.33,
|
| 255 |
-
"learning_rate":
|
| 256 |
-
"loss": 1.
|
| 257 |
"step": 37
|
| 258 |
},
|
| 259 |
{
|
| 260 |
"epoch": 0.34,
|
| 261 |
-
"learning_rate":
|
| 262 |
-
"loss": 1.
|
| 263 |
"step": 38
|
| 264 |
},
|
| 265 |
{
|
| 266 |
"epoch": 1.01,
|
| 267 |
-
"learning_rate":
|
| 268 |
-
"loss":
|
| 269 |
"step": 39
|
| 270 |
},
|
| 271 |
{
|
| 272 |
"epoch": 1.02,
|
| 273 |
-
"learning_rate":
|
| 274 |
-
"loss": 1.
|
| 275 |
"step": 40
|
| 276 |
},
|
| 277 |
{
|
| 278 |
"epoch": 1.03,
|
| 279 |
-
"learning_rate":
|
| 280 |
-
"loss":
|
| 281 |
"step": 41
|
| 282 |
},
|
| 283 |
{
|
| 284 |
"epoch": 1.04,
|
| 285 |
-
"learning_rate":
|
| 286 |
-
"loss":
|
| 287 |
"step": 42
|
| 288 |
},
|
| 289 |
{
|
| 290 |
"epoch": 1.04,
|
| 291 |
-
"learning_rate":
|
| 292 |
-
"loss":
|
| 293 |
"step": 43
|
| 294 |
},
|
| 295 |
{
|
| 296 |
"epoch": 1.05,
|
| 297 |
-
"learning_rate":
|
| 298 |
-
"loss":
|
| 299 |
"step": 44
|
| 300 |
},
|
| 301 |
{
|
| 302 |
"epoch": 1.05,
|
| 303 |
-
"eval_loss": 1.
|
| 304 |
-
"eval_runtime":
|
| 305 |
-
"eval_samples_per_second": 3.
|
| 306 |
-
"eval_steps_per_second": 0.
|
| 307 |
-
"eval_wer":
|
| 308 |
"step": 44
|
| 309 |
},
|
| 310 |
{
|
| 311 |
"epoch": 1.06,
|
| 312 |
-
"learning_rate":
|
| 313 |
-
"loss":
|
| 314 |
"step": 45
|
| 315 |
},
|
| 316 |
{
|
| 317 |
"epoch": 1.07,
|
| 318 |
-
"learning_rate":
|
| 319 |
-
"loss": 0.
|
| 320 |
"step": 46
|
| 321 |
},
|
| 322 |
{
|
| 323 |
"epoch": 1.08,
|
| 324 |
-
"learning_rate":
|
| 325 |
-
"loss": 0.
|
| 326 |
"step": 47
|
| 327 |
},
|
| 328 |
{
|
| 329 |
"epoch": 1.09,
|
| 330 |
-
"learning_rate":
|
| 331 |
-
"loss":
|
| 332 |
"step": 48
|
| 333 |
},
|
| 334 |
{
|
| 335 |
"epoch": 1.1,
|
| 336 |
-
"learning_rate":
|
| 337 |
-
"loss": 0.
|
| 338 |
"step": 49
|
| 339 |
},
|
| 340 |
{
|
| 341 |
"epoch": 1.11,
|
| 342 |
-
"learning_rate":
|
| 343 |
-
"loss": 0.
|
| 344 |
"step": 50
|
| 345 |
},
|
| 346 |
{
|
| 347 |
"epoch": 1.12,
|
| 348 |
-
"learning_rate":
|
| 349 |
-
"loss": 0.
|
| 350 |
"step": 51
|
| 351 |
},
|
| 352 |
{
|
| 353 |
"epoch": 1.12,
|
| 354 |
-
"learning_rate":
|
| 355 |
-
"loss":
|
| 356 |
"step": 52
|
| 357 |
},
|
| 358 |
{
|
| 359 |
"epoch": 1.13,
|
| 360 |
-
"learning_rate":
|
| 361 |
-
"loss":
|
| 362 |
"step": 53
|
| 363 |
},
|
| 364 |
{
|
| 365 |
"epoch": 1.14,
|
| 366 |
-
"learning_rate":
|
| 367 |
-
"loss": 0.
|
| 368 |
"step": 54
|
| 369 |
},
|
| 370 |
{
|
| 371 |
"epoch": 1.15,
|
| 372 |
-
"learning_rate":
|
| 373 |
-
"loss": 0.
|
| 374 |
"step": 55
|
| 375 |
},
|
| 376 |
{
|
| 377 |
"epoch": 1.15,
|
| 378 |
-
"eval_loss": 1.
|
| 379 |
-
"eval_runtime":
|
| 380 |
-
"eval_samples_per_second":
|
| 381 |
-
"eval_steps_per_second": 0.
|
| 382 |
-
"eval_wer":
|
| 383 |
"step": 55
|
| 384 |
},
|
| 385 |
{
|
| 386 |
"epoch": 1.16,
|
| 387 |
-
"learning_rate":
|
| 388 |
-
"loss": 0.
|
| 389 |
"step": 56
|
| 390 |
},
|
| 391 |
{
|
| 392 |
"epoch": 1.17,
|
| 393 |
-
"learning_rate":
|
| 394 |
-
"loss": 0.
|
| 395 |
"step": 57
|
| 396 |
},
|
| 397 |
{
|
| 398 |
"epoch": 1.18,
|
| 399 |
-
"learning_rate":
|
| 400 |
-
"loss": 0.
|
| 401 |
"step": 58
|
| 402 |
},
|
| 403 |
{
|
| 404 |
"epoch": 1.19,
|
| 405 |
-
"learning_rate":
|
| 406 |
-
"loss": 0.
|
| 407 |
"step": 59
|
| 408 |
},
|
| 409 |
{
|
| 410 |
"epoch": 1.2,
|
| 411 |
-
"learning_rate": 5.
|
| 412 |
-
"loss": 0.
|
| 413 |
"step": 60
|
| 414 |
},
|
| 415 |
{
|
| 416 |
"epoch": 1.21,
|
| 417 |
-
"learning_rate":
|
| 418 |
-
"loss": 0.
|
| 419 |
"step": 61
|
| 420 |
},
|
| 421 |
{
|
| 422 |
"epoch": 1.21,
|
| 423 |
-
"learning_rate":
|
| 424 |
-
"loss": 0.
|
| 425 |
"step": 62
|
| 426 |
},
|
| 427 |
{
|
| 428 |
"epoch": 1.22,
|
| 429 |
-
"learning_rate":
|
| 430 |
-
"loss": 0.
|
| 431 |
"step": 63
|
| 432 |
},
|
| 433 |
{
|
| 434 |
"epoch": 1.23,
|
| 435 |
-
"learning_rate":
|
| 436 |
-
"loss": 0.
|
| 437 |
"step": 64
|
| 438 |
},
|
| 439 |
{
|
| 440 |
"epoch": 1.24,
|
| 441 |
-
"learning_rate":
|
| 442 |
-
"loss": 0.
|
| 443 |
"step": 65
|
| 444 |
},
|
| 445 |
{
|
| 446 |
"epoch": 1.25,
|
| 447 |
-
"learning_rate":
|
| 448 |
-
"loss": 0.
|
| 449 |
"step": 66
|
| 450 |
},
|
| 451 |
{
|
| 452 |
"epoch": 1.25,
|
| 453 |
-
"eval_loss": 1.
|
| 454 |
-
"eval_runtime":
|
| 455 |
-
"eval_samples_per_second": 3.
|
| 456 |
-
"eval_steps_per_second": 0.
|
| 457 |
-
"eval_wer":
|
| 458 |
"step": 66
|
| 459 |
},
|
| 460 |
{
|
| 461 |
"epoch": 1.26,
|
| 462 |
-
"learning_rate":
|
| 463 |
-
"loss": 0.
|
| 464 |
"step": 67
|
| 465 |
},
|
| 466 |
{
|
| 467 |
"epoch": 1.27,
|
| 468 |
-
"learning_rate":
|
| 469 |
-
"loss": 0.
|
| 470 |
"step": 68
|
| 471 |
},
|
| 472 |
{
|
| 473 |
"epoch": 1.28,
|
| 474 |
-
"learning_rate": 4.
|
| 475 |
-
"loss": 0.
|
| 476 |
"step": 69
|
| 477 |
},
|
| 478 |
{
|
| 479 |
"epoch": 1.29,
|
| 480 |
-
"learning_rate": 4.
|
| 481 |
-
"loss": 0.
|
| 482 |
"step": 70
|
| 483 |
},
|
| 484 |
{
|
| 485 |
"epoch": 1.29,
|
| 486 |
-
"learning_rate": 4.
|
| 487 |
-
"loss": 0.
|
| 488 |
"step": 71
|
| 489 |
},
|
| 490 |
{
|
| 491 |
"epoch": 1.3,
|
| 492 |
-
"learning_rate":
|
| 493 |
-
"loss": 0.
|
| 494 |
"step": 72
|
| 495 |
},
|
| 496 |
{
|
| 497 |
"epoch": 1.31,
|
| 498 |
-
"learning_rate":
|
| 499 |
-
"loss": 0.
|
| 500 |
"step": 73
|
| 501 |
},
|
| 502 |
{
|
| 503 |
"epoch": 1.32,
|
| 504 |
-
"learning_rate":
|
| 505 |
-
"loss": 0.
|
| 506 |
"step": 74
|
| 507 |
},
|
| 508 |
{
|
| 509 |
"epoch": 1.33,
|
| 510 |
-
"learning_rate":
|
| 511 |
-
"loss": 0.
|
| 512 |
"step": 75
|
| 513 |
},
|
| 514 |
{
|
| 515 |
"epoch": 1.34,
|
| 516 |
-
"learning_rate":
|
| 517 |
-
"loss": 0.
|
| 518 |
"step": 76
|
| 519 |
},
|
| 520 |
{
|
| 521 |
"epoch": 2.01,
|
| 522 |
-
"learning_rate":
|
| 523 |
-
"loss": 0.
|
| 524 |
"step": 77
|
| 525 |
},
|
| 526 |
{
|
| 527 |
"epoch": 2.01,
|
| 528 |
-
"eval_loss": 1.
|
| 529 |
-
"eval_runtime":
|
| 530 |
-
"eval_samples_per_second": 3.
|
| 531 |
-
"eval_steps_per_second": 0.
|
| 532 |
-
"eval_wer":
|
| 533 |
"step": 77
|
| 534 |
},
|
| 535 |
{
|
| 536 |
"epoch": 2.02,
|
| 537 |
-
"learning_rate": 3.
|
| 538 |
-
"loss": 0.
|
| 539 |
"step": 78
|
| 540 |
},
|
| 541 |
{
|
| 542 |
"epoch": 2.03,
|
| 543 |
-
"learning_rate": 3.
|
| 544 |
-
"loss": 0.
|
| 545 |
"step": 79
|
| 546 |
},
|
| 547 |
{
|
| 548 |
"epoch": 2.04,
|
| 549 |
-
"learning_rate": 3.
|
| 550 |
-
"loss": 0.
|
| 551 |
"step": 80
|
| 552 |
},
|
| 553 |
{
|
| 554 |
"epoch": 2.04,
|
| 555 |
-
"learning_rate": 3.
|
| 556 |
-
"loss": 0.
|
| 557 |
"step": 81
|
| 558 |
},
|
| 559 |
{
|
| 560 |
"epoch": 2.05,
|
| 561 |
-
"learning_rate":
|
| 562 |
-
"loss": 0.
|
| 563 |
"step": 82
|
| 564 |
},
|
| 565 |
{
|
| 566 |
"epoch": 2.06,
|
| 567 |
-
"learning_rate":
|
| 568 |
-
"loss": 0.
|
| 569 |
"step": 83
|
| 570 |
},
|
| 571 |
{
|
| 572 |
"epoch": 2.07,
|
| 573 |
-
"learning_rate":
|
| 574 |
-
"loss": 0.
|
| 575 |
"step": 84
|
| 576 |
},
|
| 577 |
{
|
| 578 |
"epoch": 2.08,
|
| 579 |
-
"learning_rate":
|
| 580 |
-
"loss": 0.
|
| 581 |
"step": 85
|
| 582 |
},
|
| 583 |
{
|
| 584 |
"epoch": 2.09,
|
| 585 |
-
"learning_rate":
|
| 586 |
-
"loss": 0.
|
| 587 |
"step": 86
|
| 588 |
},
|
| 589 |
{
|
| 590 |
"epoch": 2.1,
|
| 591 |
-
"learning_rate": 2.
|
| 592 |
-
"loss": 0.
|
| 593 |
"step": 87
|
| 594 |
},
|
| 595 |
{
|
| 596 |
"epoch": 2.11,
|
| 597 |
-
"learning_rate": 2.
|
| 598 |
-
"loss": 0.
|
| 599 |
"step": 88
|
| 600 |
},
|
| 601 |
{
|
| 602 |
"epoch": 2.11,
|
| 603 |
-
"eval_loss": 1.
|
| 604 |
-
"eval_runtime":
|
| 605 |
-
"eval_samples_per_second":
|
| 606 |
-
"eval_steps_per_second": 0.
|
| 607 |
-
"eval_wer":
|
| 608 |
"step": 88
|
| 609 |
},
|
| 610 |
{
|
| 611 |
"epoch": 2.12,
|
| 612 |
-
"learning_rate": 2.
|
| 613 |
-
"loss": 0.
|
| 614 |
"step": 89
|
| 615 |
},
|
| 616 |
{
|
| 617 |
"epoch": 2.12,
|
| 618 |
-
"learning_rate": 2.
|
| 619 |
-
"loss": 0.
|
| 620 |
"step": 90
|
| 621 |
},
|
| 622 |
{
|
| 623 |
"epoch": 2.13,
|
| 624 |
-
"learning_rate": 2.
|
| 625 |
-
"loss": 0.
|
| 626 |
"step": 91
|
| 627 |
},
|
| 628 |
{
|
| 629 |
"epoch": 2.14,
|
| 630 |
-
"learning_rate": 2.
|
| 631 |
-
"loss": 0.
|
| 632 |
"step": 92
|
| 633 |
},
|
| 634 |
{
|
| 635 |
"epoch": 2.15,
|
| 636 |
-
"learning_rate":
|
| 637 |
-
"loss": 0.
|
| 638 |
"step": 93
|
| 639 |
},
|
| 640 |
{
|
| 641 |
"epoch": 2.16,
|
| 642 |
-
"learning_rate":
|
| 643 |
-
"loss": 0.
|
| 644 |
"step": 94
|
| 645 |
},
|
| 646 |
{
|
| 647 |
"epoch": 2.17,
|
| 648 |
-
"learning_rate":
|
| 649 |
-
"loss": 0.
|
| 650 |
"step": 95
|
| 651 |
},
|
| 652 |
{
|
| 653 |
"epoch": 2.18,
|
| 654 |
-
"learning_rate": 1.
|
| 655 |
-
"loss": 0.
|
| 656 |
"step": 96
|
| 657 |
},
|
| 658 |
{
|
| 659 |
"epoch": 2.19,
|
| 660 |
-
"learning_rate": 1.
|
| 661 |
-
"loss": 0.
|
| 662 |
"step": 97
|
| 663 |
},
|
| 664 |
{
|
| 665 |
"epoch": 2.2,
|
| 666 |
-
"learning_rate": 1.
|
| 667 |
-
"loss": 0.
|
| 668 |
"step": 98
|
| 669 |
},
|
| 670 |
{
|
| 671 |
"epoch": 2.21,
|
| 672 |
-
"learning_rate": 1.
|
| 673 |
-
"loss": 0.
|
| 674 |
"step": 99
|
| 675 |
},
|
| 676 |
{
|
| 677 |
"epoch": 2.21,
|
| 678 |
-
"eval_loss": 1.
|
| 679 |
-
"eval_runtime":
|
| 680 |
-
"eval_samples_per_second": 3.
|
| 681 |
-
"eval_steps_per_second": 0.
|
| 682 |
-
"eval_wer":
|
| 683 |
"step": 99
|
| 684 |
},
|
| 685 |
{
|
| 686 |
"epoch": 2.21,
|
| 687 |
-
"learning_rate": 1.
|
| 688 |
-
"loss": 0.
|
| 689 |
"step": 100
|
| 690 |
},
|
| 691 |
{
|
| 692 |
"epoch": 2.22,
|
| 693 |
-
"learning_rate": 1.
|
| 694 |
-
"loss": 0.
|
| 695 |
"step": 101
|
| 696 |
},
|
| 697 |
{
|
| 698 |
"epoch": 2.23,
|
| 699 |
-
"learning_rate": 1.
|
| 700 |
-
"loss": 0.
|
| 701 |
"step": 102
|
| 702 |
},
|
| 703 |
{
|
| 704 |
"epoch": 2.24,
|
| 705 |
-
"learning_rate":
|
| 706 |
-
"loss": 0.
|
| 707 |
"step": 103
|
| 708 |
},
|
| 709 |
{
|
| 710 |
"epoch": 2.25,
|
| 711 |
-
"learning_rate":
|
| 712 |
-
"loss": 0.
|
| 713 |
"step": 104
|
| 714 |
},
|
| 715 |
{
|
| 716 |
"epoch": 2.26,
|
| 717 |
-
"learning_rate":
|
| 718 |
-
"loss": 0.
|
| 719 |
"step": 105
|
| 720 |
},
|
| 721 |
{
|
| 722 |
"epoch": 2.27,
|
| 723 |
-
"learning_rate":
|
| 724 |
-
"loss": 0.
|
| 725 |
"step": 106
|
| 726 |
},
|
| 727 |
{
|
| 728 |
"epoch": 2.28,
|
| 729 |
-
"learning_rate":
|
| 730 |
-
"loss": 0.
|
| 731 |
"step": 107
|
| 732 |
},
|
| 733 |
{
|
| 734 |
"epoch": 2.29,
|
| 735 |
-
"learning_rate":
|
| 736 |
-
"loss": 0.
|
| 737 |
"step": 108
|
| 738 |
},
|
| 739 |
{
|
| 740 |
"epoch": 2.29,
|
| 741 |
-
"learning_rate":
|
| 742 |
-
"loss": 0.
|
| 743 |
"step": 109
|
| 744 |
},
|
| 745 |
{
|
| 746 |
"epoch": 2.3,
|
| 747 |
-
"learning_rate":
|
| 748 |
-
"loss": 0.
|
| 749 |
"step": 110
|
| 750 |
},
|
| 751 |
{
|
| 752 |
"epoch": 2.3,
|
| 753 |
-
"eval_loss": 1.
|
| 754 |
-
"eval_runtime":
|
| 755 |
-
"eval_samples_per_second": 3.
|
| 756 |
-
"eval_steps_per_second": 0.
|
| 757 |
-
"eval_wer":
|
| 758 |
"step": 110
|
| 759 |
},
|
| 760 |
{
|
| 761 |
"epoch": 2.31,
|
| 762 |
-
"learning_rate":
|
| 763 |
-
"loss": 0.
|
| 764 |
"step": 111
|
| 765 |
},
|
| 766 |
{
|
| 767 |
"epoch": 2.32,
|
| 768 |
-
"learning_rate":
|
| 769 |
-
"loss": 0.
|
| 770 |
"step": 112
|
| 771 |
},
|
| 772 |
{
|
| 773 |
"epoch": 2.32,
|
| 774 |
"step": 112,
|
| 775 |
"total_flos": 1.7415399333888e+17,
|
| 776 |
-
"train_loss":
|
| 777 |
-
"train_runtime":
|
| 778 |
-
"train_samples_per_second": 4.
|
| 779 |
-
"train_steps_per_second": 0.
|
| 780 |
}
|
| 781 |
],
|
| 782 |
"max_steps": 112,
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 161.45617116397892,
|
| 3 |
"best_model_checkpoint": "./checkpoint-88",
|
| 4 |
"epoch": 2.3214285714285716,
|
| 5 |
"global_step": 112,
|
|
|
|
| 15 |
},
|
| 16 |
{
|
| 17 |
"epoch": 0.02,
|
| 18 |
+
"learning_rate": 2.2058823529411765e-07,
|
| 19 |
"loss": 1.6564,
|
| 20 |
"step": 2
|
| 21 |
},
|
| 22 |
{
|
| 23 |
"epoch": 0.03,
|
| 24 |
+
"learning_rate": 4.411764705882353e-07,
|
| 25 |
"loss": 1.6518,
|
| 26 |
"step": 3
|
| 27 |
},
|
| 28 |
{
|
| 29 |
"epoch": 0.04,
|
| 30 |
+
"learning_rate": 6.61764705882353e-07,
|
| 31 |
+
"loss": 1.6237,
|
| 32 |
"step": 4
|
| 33 |
},
|
| 34 |
{
|
| 35 |
"epoch": 0.04,
|
| 36 |
+
"learning_rate": 8.823529411764706e-07,
|
| 37 |
+
"loss": 1.7141,
|
| 38 |
"step": 5
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 0.05,
|
| 42 |
+
"learning_rate": 1.1029411764705884e-06,
|
| 43 |
+
"loss": 1.655,
|
| 44 |
"step": 6
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.06,
|
| 48 |
+
"learning_rate": 1.323529411764706e-06,
|
| 49 |
+
"loss": 1.5904,
|
| 50 |
"step": 7
|
| 51 |
},
|
| 52 |
{
|
| 53 |
"epoch": 0.07,
|
| 54 |
+
"learning_rate": 1.5441176470588234e-06,
|
| 55 |
+
"loss": 1.6232,
|
| 56 |
"step": 8
|
| 57 |
},
|
| 58 |
{
|
| 59 |
"epoch": 0.08,
|
| 60 |
+
"learning_rate": 1.7647058823529412e-06,
|
| 61 |
+
"loss": 1.6403,
|
| 62 |
"step": 9
|
| 63 |
},
|
| 64 |
{
|
| 65 |
"epoch": 0.09,
|
| 66 |
+
"learning_rate": 1.985294117647059e-06,
|
| 67 |
+
"loss": 1.6065,
|
| 68 |
"step": 10
|
| 69 |
},
|
| 70 |
{
|
| 71 |
"epoch": 0.1,
|
| 72 |
+
"learning_rate": 2.2058823529411767e-06,
|
| 73 |
+
"loss": 1.5947,
|
| 74 |
"step": 11
|
| 75 |
},
|
| 76 |
{
|
| 77 |
"epoch": 0.1,
|
| 78 |
+
"eval_loss": 1.629066824913025,
|
| 79 |
+
"eval_runtime": 125.637,
|
| 80 |
+
"eval_samples_per_second": 2.627,
|
| 81 |
+
"eval_steps_per_second": 0.088,
|
| 82 |
+
"eval_wer": 227.94188088775348,
|
| 83 |
"step": 11
|
| 84 |
},
|
| 85 |
{
|
| 86 |
"epoch": 0.11,
|
| 87 |
+
"learning_rate": 2.4264705882352943e-06,
|
| 88 |
+
"loss": 1.52,
|
| 89 |
"step": 12
|
| 90 |
},
|
| 91 |
{
|
| 92 |
"epoch": 0.12,
|
| 93 |
+
"learning_rate": 2.647058823529412e-06,
|
| 94 |
+
"loss": 1.4456,
|
| 95 |
"step": 13
|
| 96 |
},
|
| 97 |
{
|
| 98 |
"epoch": 0.12,
|
| 99 |
+
"learning_rate": 2.867647058823529e-06,
|
| 100 |
+
"loss": 1.58,
|
| 101 |
"step": 14
|
| 102 |
},
|
| 103 |
{
|
| 104 |
"epoch": 0.13,
|
| 105 |
+
"learning_rate": 3.088235294117647e-06,
|
| 106 |
+
"loss": 1.5958,
|
| 107 |
"step": 15
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 0.14,
|
| 111 |
+
"learning_rate": 3.308823529411765e-06,
|
| 112 |
+
"loss": 1.4695,
|
| 113 |
"step": 16
|
| 114 |
},
|
| 115 |
{
|
| 116 |
"epoch": 0.15,
|
| 117 |
+
"learning_rate": 3.5294117647058825e-06,
|
| 118 |
+
"loss": 1.4718,
|
| 119 |
"step": 17
|
| 120 |
},
|
| 121 |
{
|
| 122 |
"epoch": 0.16,
|
| 123 |
+
"learning_rate": 3.75e-06,
|
| 124 |
+
"loss": 1.4406,
|
| 125 |
"step": 18
|
| 126 |
},
|
| 127 |
{
|
| 128 |
"epoch": 0.17,
|
| 129 |
+
"learning_rate": 3.970588235294118e-06,
|
| 130 |
+
"loss": 1.4339,
|
| 131 |
"step": 19
|
| 132 |
},
|
| 133 |
{
|
| 134 |
"epoch": 0.18,
|
| 135 |
+
"learning_rate": 4.191176470588236e-06,
|
| 136 |
+
"loss": 1.313,
|
| 137 |
"step": 20
|
| 138 |
},
|
| 139 |
{
|
| 140 |
"epoch": 0.19,
|
| 141 |
+
"learning_rate": 4.411764705882353e-06,
|
| 142 |
+
"loss": 1.2825,
|
| 143 |
"step": 21
|
| 144 |
},
|
| 145 |
{
|
| 146 |
"epoch": 0.2,
|
| 147 |
+
"learning_rate": 4.632352941176471e-06,
|
| 148 |
+
"loss": 1.3194,
|
| 149 |
"step": 22
|
| 150 |
},
|
| 151 |
{
|
| 152 |
"epoch": 0.2,
|
| 153 |
+
"eval_loss": 1.460469126701355,
|
| 154 |
+
"eval_runtime": 118.4642,
|
| 155 |
+
"eval_samples_per_second": 2.786,
|
| 156 |
+
"eval_steps_per_second": 0.093,
|
| 157 |
+
"eval_wer": 235.94124221619032,
|
| 158 |
"step": 22
|
| 159 |
},
|
| 160 |
{
|
| 161 |
"epoch": 0.21,
|
| 162 |
+
"learning_rate": 4.852941176470589e-06,
|
| 163 |
+
"loss": 1.2301,
|
| 164 |
"step": 23
|
| 165 |
},
|
| 166 |
{
|
| 167 |
"epoch": 0.21,
|
| 168 |
+
"learning_rate": 5.073529411764706e-06,
|
| 169 |
+
"loss": 1.3231,
|
| 170 |
"step": 24
|
| 171 |
},
|
| 172 |
{
|
| 173 |
"epoch": 0.22,
|
| 174 |
+
"learning_rate": 5.294117647058824e-06,
|
| 175 |
+
"loss": 1.2689,
|
| 176 |
"step": 25
|
| 177 |
},
|
| 178 |
{
|
| 179 |
"epoch": 0.23,
|
| 180 |
+
"learning_rate": 5.5147058823529415e-06,
|
| 181 |
+
"loss": 1.2768,
|
| 182 |
"step": 26
|
| 183 |
},
|
| 184 |
{
|
| 185 |
"epoch": 0.24,
|
| 186 |
+
"learning_rate": 5.735294117647058e-06,
|
| 187 |
+
"loss": 1.2927,
|
| 188 |
"step": 27
|
| 189 |
},
|
| 190 |
{
|
| 191 |
"epoch": 0.25,
|
| 192 |
+
"learning_rate": 5.955882352941176e-06,
|
| 193 |
+
"loss": 1.3419,
|
| 194 |
"step": 28
|
| 195 |
},
|
| 196 |
{
|
| 197 |
"epoch": 0.26,
|
| 198 |
+
"learning_rate": 6.176470588235294e-06,
|
| 199 |
+
"loss": 1.141,
|
| 200 |
"step": 29
|
| 201 |
},
|
| 202 |
{
|
| 203 |
"epoch": 0.27,
|
| 204 |
+
"learning_rate": 6.397058823529412e-06,
|
| 205 |
+
"loss": 1.2907,
|
| 206 |
"step": 30
|
| 207 |
},
|
| 208 |
{
|
| 209 |
"epoch": 0.28,
|
| 210 |
+
"learning_rate": 6.61764705882353e-06,
|
| 211 |
+
"loss": 1.2248,
|
| 212 |
"step": 31
|
| 213 |
},
|
| 214 |
{
|
| 215 |
"epoch": 0.29,
|
| 216 |
+
"learning_rate": 6.838235294117647e-06,
|
| 217 |
+
"loss": 1.149,
|
| 218 |
"step": 32
|
| 219 |
},
|
| 220 |
{
|
| 221 |
"epoch": 0.29,
|
| 222 |
+
"learning_rate": 7.058823529411765e-06,
|
| 223 |
+
"loss": 1.2636,
|
| 224 |
"step": 33
|
| 225 |
},
|
| 226 |
{
|
| 227 |
"epoch": 0.29,
|
| 228 |
+
"eval_loss": 1.3149287700653076,
|
| 229 |
+
"eval_runtime": 115.745,
|
| 230 |
+
"eval_samples_per_second": 2.851,
|
| 231 |
+
"eval_steps_per_second": 0.095,
|
| 232 |
+
"eval_wer": 177.19942519559316,
|
| 233 |
"step": 33
|
| 234 |
},
|
| 235 |
{
|
| 236 |
"epoch": 0.3,
|
| 237 |
+
"learning_rate": 7.2794117647058826e-06,
|
| 238 |
+
"loss": 1.2351,
|
| 239 |
"step": 34
|
| 240 |
},
|
| 241 |
{
|
| 242 |
"epoch": 0.31,
|
| 243 |
+
"learning_rate": 7.5e-06,
|
| 244 |
+
"loss": 1.2055,
|
| 245 |
"step": 35
|
| 246 |
},
|
| 247 |
{
|
| 248 |
"epoch": 0.32,
|
| 249 |
+
"learning_rate": 7.403846153846155e-06,
|
| 250 |
+
"loss": 1.1864,
|
| 251 |
"step": 36
|
| 252 |
},
|
| 253 |
{
|
| 254 |
"epoch": 0.33,
|
| 255 |
+
"learning_rate": 7.307692307692308e-06,
|
| 256 |
+
"loss": 1.148,
|
| 257 |
"step": 37
|
| 258 |
},
|
| 259 |
{
|
| 260 |
"epoch": 0.34,
|
| 261 |
+
"learning_rate": 7.211538461538462e-06,
|
| 262 |
+
"loss": 1.1505,
|
| 263 |
"step": 38
|
| 264 |
},
|
| 265 |
{
|
| 266 |
"epoch": 1.01,
|
| 267 |
+
"learning_rate": 7.115384615384615e-06,
|
| 268 |
+
"loss": 1.0859,
|
| 269 |
"step": 39
|
| 270 |
},
|
| 271 |
{
|
| 272 |
"epoch": 1.02,
|
| 273 |
+
"learning_rate": 7.01923076923077e-06,
|
| 274 |
+
"loss": 1.119,
|
| 275 |
"step": 40
|
| 276 |
},
|
| 277 |
{
|
| 278 |
"epoch": 1.03,
|
| 279 |
+
"learning_rate": 6.923076923076923e-06,
|
| 280 |
+
"loss": 1.0238,
|
| 281 |
"step": 41
|
| 282 |
},
|
| 283 |
{
|
| 284 |
"epoch": 1.04,
|
| 285 |
+
"learning_rate": 6.826923076923077e-06,
|
| 286 |
+
"loss": 1.0468,
|
| 287 |
"step": 42
|
| 288 |
},
|
| 289 |
{
|
| 290 |
"epoch": 1.04,
|
| 291 |
+
"learning_rate": 6.730769230769231e-06,
|
| 292 |
+
"loss": 1.061,
|
| 293 |
"step": 43
|
| 294 |
},
|
| 295 |
{
|
| 296 |
"epoch": 1.05,
|
| 297 |
+
"learning_rate": 6.6346153846153846e-06,
|
| 298 |
+
"loss": 1.0238,
|
| 299 |
"step": 44
|
| 300 |
},
|
| 301 |
{
|
| 302 |
"epoch": 1.05,
|
| 303 |
+
"eval_loss": 1.1836069822311401,
|
| 304 |
+
"eval_runtime": 109.1507,
|
| 305 |
+
"eval_samples_per_second": 3.023,
|
| 306 |
+
"eval_steps_per_second": 0.101,
|
| 307 |
+
"eval_wer": 199.80839853105542,
|
| 308 |
"step": 44
|
| 309 |
},
|
| 310 |
{
|
| 311 |
"epoch": 1.06,
|
| 312 |
+
"learning_rate": 6.538461538461539e-06,
|
| 313 |
+
"loss": 1.0016,
|
| 314 |
"step": 45
|
| 315 |
},
|
| 316 |
{
|
| 317 |
"epoch": 1.07,
|
| 318 |
+
"learning_rate": 6.442307692307692e-06,
|
| 319 |
+
"loss": 0.9635,
|
| 320 |
"step": 46
|
| 321 |
},
|
| 322 |
{
|
| 323 |
"epoch": 1.08,
|
| 324 |
+
"learning_rate": 6.3461538461538466e-06,
|
| 325 |
+
"loss": 0.9953,
|
| 326 |
"step": 47
|
| 327 |
},
|
| 328 |
{
|
| 329 |
"epoch": 1.09,
|
| 330 |
+
"learning_rate": 6.25e-06,
|
| 331 |
+
"loss": 1.0061,
|
| 332 |
"step": 48
|
| 333 |
},
|
| 334 |
{
|
| 335 |
"epoch": 1.1,
|
| 336 |
+
"learning_rate": 6.153846153846154e-06,
|
| 337 |
+
"loss": 0.9933,
|
| 338 |
"step": 49
|
| 339 |
},
|
| 340 |
{
|
| 341 |
"epoch": 1.11,
|
| 342 |
+
"learning_rate": 6.057692307692308e-06,
|
| 343 |
+
"loss": 0.9479,
|
| 344 |
"step": 50
|
| 345 |
},
|
| 346 |
{
|
| 347 |
"epoch": 1.12,
|
| 348 |
+
"learning_rate": 5.9615384615384615e-06,
|
| 349 |
+
"loss": 0.9132,
|
| 350 |
"step": 51
|
| 351 |
},
|
| 352 |
{
|
| 353 |
"epoch": 1.12,
|
| 354 |
+
"learning_rate": 5.865384615384615e-06,
|
| 355 |
+
"loss": 1.0232,
|
| 356 |
"step": 52
|
| 357 |
},
|
| 358 |
{
|
| 359 |
"epoch": 1.13,
|
| 360 |
+
"learning_rate": 5.76923076923077e-06,
|
| 361 |
+
"loss": 1.0153,
|
| 362 |
"step": 53
|
| 363 |
},
|
| 364 |
{
|
| 365 |
"epoch": 1.14,
|
| 366 |
+
"learning_rate": 5.673076923076923e-06,
|
| 367 |
+
"loss": 0.9548,
|
| 368 |
"step": 54
|
| 369 |
},
|
| 370 |
{
|
| 371 |
"epoch": 1.15,
|
| 372 |
+
"learning_rate": 5.576923076923077e-06,
|
| 373 |
+
"loss": 0.9306,
|
| 374 |
"step": 55
|
| 375 |
},
|
| 376 |
{
|
| 377 |
"epoch": 1.15,
|
| 378 |
+
"eval_loss": 1.1363615989685059,
|
| 379 |
+
"eval_runtime": 115.8861,
|
| 380 |
+
"eval_samples_per_second": 2.848,
|
| 381 |
+
"eval_steps_per_second": 0.095,
|
| 382 |
+
"eval_wer": 210.21874501037843,
|
| 383 |
"step": 55
|
| 384 |
},
|
| 385 |
{
|
| 386 |
"epoch": 1.16,
|
| 387 |
+
"learning_rate": 5.480769230769231e-06,
|
| 388 |
+
"loss": 0.9088,
|
| 389 |
"step": 56
|
| 390 |
},
|
| 391 |
{
|
| 392 |
"epoch": 1.17,
|
| 393 |
+
"learning_rate": 5.384615384615385e-06,
|
| 394 |
+
"loss": 0.8902,
|
| 395 |
"step": 57
|
| 396 |
},
|
| 397 |
{
|
| 398 |
"epoch": 1.18,
|
| 399 |
+
"learning_rate": 5.288461538461539e-06,
|
| 400 |
+
"loss": 0.861,
|
| 401 |
"step": 58
|
| 402 |
},
|
| 403 |
{
|
| 404 |
"epoch": 1.19,
|
| 405 |
+
"learning_rate": 5.192307692307692e-06,
|
| 406 |
+
"loss": 0.8047,
|
| 407 |
"step": 59
|
| 408 |
},
|
| 409 |
{
|
| 410 |
"epoch": 1.2,
|
| 411 |
+
"learning_rate": 5.096153846153847e-06,
|
| 412 |
+
"loss": 0.8459,
|
| 413 |
"step": 60
|
| 414 |
},
|
| 415 |
{
|
| 416 |
"epoch": 1.21,
|
| 417 |
+
"learning_rate": 4.9999999999999996e-06,
|
| 418 |
+
"loss": 0.8103,
|
| 419 |
"step": 61
|
| 420 |
},
|
| 421 |
{
|
| 422 |
"epoch": 1.21,
|
| 423 |
+
"learning_rate": 4.903846153846154e-06,
|
| 424 |
+
"loss": 0.8752,
|
| 425 |
"step": 62
|
| 426 |
},
|
| 427 |
{
|
| 428 |
"epoch": 1.22,
|
| 429 |
+
"learning_rate": 4.807692307692308e-06,
|
| 430 |
+
"loss": 0.8398,
|
| 431 |
"step": 63
|
| 432 |
},
|
| 433 |
{
|
| 434 |
"epoch": 1.23,
|
| 435 |
+
"learning_rate": 4.711538461538462e-06,
|
| 436 |
+
"loss": 0.8571,
|
| 437 |
"step": 64
|
| 438 |
},
|
| 439 |
{
|
| 440 |
"epoch": 1.24,
|
| 441 |
+
"learning_rate": 4.615384615384616e-06,
|
| 442 |
+
"loss": 0.8998,
|
| 443 |
"step": 65
|
| 444 |
},
|
| 445 |
{
|
| 446 |
"epoch": 1.25,
|
| 447 |
+
"learning_rate": 4.519230769230769e-06,
|
| 448 |
+
"loss": 0.9233,
|
| 449 |
"step": 66
|
| 450 |
},
|
| 451 |
{
|
| 452 |
"epoch": 1.25,
|
| 453 |
+
"eval_loss": 1.1090655326843262,
|
| 454 |
+
"eval_runtime": 104.9772,
|
| 455 |
+
"eval_samples_per_second": 3.144,
|
| 456 |
+
"eval_steps_per_second": 0.105,
|
| 457 |
+
"eval_wer": 175.20357656075362,
|
| 458 |
"step": 66
|
| 459 |
},
|
| 460 |
{
|
| 461 |
"epoch": 1.26,
|
| 462 |
+
"learning_rate": 4.423076923076924e-06,
|
| 463 |
+
"loss": 0.7878,
|
| 464 |
"step": 67
|
| 465 |
},
|
| 466 |
{
|
| 467 |
"epoch": 1.27,
|
| 468 |
+
"learning_rate": 4.3269230769230765e-06,
|
| 469 |
+
"loss": 0.8948,
|
| 470 |
"step": 68
|
| 471 |
},
|
| 472 |
{
|
| 473 |
"epoch": 1.28,
|
| 474 |
+
"learning_rate": 4.230769230769231e-06,
|
| 475 |
+
"loss": 0.8711,
|
| 476 |
"step": 69
|
| 477 |
},
|
| 478 |
{
|
| 479 |
"epoch": 1.29,
|
| 480 |
+
"learning_rate": 4.134615384615385e-06,
|
| 481 |
+
"loss": 0.8292,
|
| 482 |
"step": 70
|
| 483 |
},
|
| 484 |
{
|
| 485 |
"epoch": 1.29,
|
| 486 |
+
"learning_rate": 4.0384615384615385e-06,
|
| 487 |
+
"loss": 0.9318,
|
| 488 |
"step": 71
|
| 489 |
},
|
| 490 |
{
|
| 491 |
"epoch": 1.3,
|
| 492 |
+
"learning_rate": 3.942307692307692e-06,
|
| 493 |
+
"loss": 0.9272,
|
| 494 |
"step": 72
|
| 495 |
},
|
| 496 |
{
|
| 497 |
"epoch": 1.31,
|
| 498 |
+
"learning_rate": 3.846153846153846e-06,
|
| 499 |
+
"loss": 0.8923,
|
| 500 |
"step": 73
|
| 501 |
},
|
| 502 |
{
|
| 503 |
"epoch": 1.32,
|
| 504 |
+
"learning_rate": 3.75e-06,
|
| 505 |
+
"loss": 0.9013,
|
| 506 |
"step": 74
|
| 507 |
},
|
| 508 |
{
|
| 509 |
"epoch": 1.33,
|
| 510 |
+
"learning_rate": 3.653846153846154e-06,
|
| 511 |
+
"loss": 0.8659,
|
| 512 |
"step": 75
|
| 513 |
},
|
| 514 |
{
|
| 515 |
"epoch": 1.34,
|
| 516 |
+
"learning_rate": 3.5576923076923075e-06,
|
| 517 |
+
"loss": 0.8422,
|
| 518 |
"step": 76
|
| 519 |
},
|
| 520 |
{
|
| 521 |
"epoch": 2.01,
|
| 522 |
+
"learning_rate": 3.4615384615384617e-06,
|
| 523 |
+
"loss": 0.8482,
|
| 524 |
"step": 77
|
| 525 |
},
|
| 526 |
{
|
| 527 |
"epoch": 2.01,
|
| 528 |
+
"eval_loss": 1.0900639295578003,
|
| 529 |
+
"eval_runtime": 99.705,
|
| 530 |
+
"eval_samples_per_second": 3.31,
|
| 531 |
+
"eval_steps_per_second": 0.11,
|
| 532 |
+
"eval_wer": 161.85534089094685,
|
| 533 |
"step": 77
|
| 534 |
},
|
| 535 |
{
|
| 536 |
"epoch": 2.02,
|
| 537 |
+
"learning_rate": 3.3653846153846154e-06,
|
| 538 |
+
"loss": 0.8925,
|
| 539 |
"step": 78
|
| 540 |
},
|
| 541 |
{
|
| 542 |
"epoch": 2.03,
|
| 543 |
+
"learning_rate": 3.2692307692307696e-06,
|
| 544 |
+
"loss": 0.8202,
|
| 545 |
"step": 79
|
| 546 |
},
|
| 547 |
{
|
| 548 |
"epoch": 2.04,
|
| 549 |
+
"learning_rate": 3.1730769230769233e-06,
|
| 550 |
+
"loss": 0.8384,
|
| 551 |
"step": 80
|
| 552 |
},
|
| 553 |
{
|
| 554 |
"epoch": 2.04,
|
| 555 |
+
"learning_rate": 3.076923076923077e-06,
|
| 556 |
+
"loss": 0.8469,
|
| 557 |
"step": 81
|
| 558 |
},
|
| 559 |
{
|
| 560 |
"epoch": 2.05,
|
| 561 |
+
"learning_rate": 2.9807692307692307e-06,
|
| 562 |
+
"loss": 0.8172,
|
| 563 |
"step": 82
|
| 564 |
},
|
| 565 |
{
|
| 566 |
"epoch": 2.06,
|
| 567 |
+
"learning_rate": 2.884615384615385e-06,
|
| 568 |
+
"loss": 0.8119,
|
| 569 |
"step": 83
|
| 570 |
},
|
| 571 |
{
|
| 572 |
"epoch": 2.07,
|
| 573 |
+
"learning_rate": 2.7884615384615386e-06,
|
| 574 |
+
"loss": 0.7811,
|
| 575 |
"step": 84
|
| 576 |
},
|
| 577 |
{
|
| 578 |
"epoch": 2.08,
|
| 579 |
+
"learning_rate": 2.6923076923076923e-06,
|
| 580 |
+
"loss": 0.819,
|
| 581 |
"step": 85
|
| 582 |
},
|
| 583 |
{
|
| 584 |
"epoch": 2.09,
|
| 585 |
+
"learning_rate": 2.596153846153846e-06,
|
| 586 |
+
"loss": 0.8324,
|
| 587 |
"step": 86
|
| 588 |
},
|
| 589 |
{
|
| 590 |
"epoch": 2.1,
|
| 591 |
+
"learning_rate": 2.4999999999999998e-06,
|
| 592 |
+
"loss": 0.8336,
|
| 593 |
"step": 87
|
| 594 |
},
|
| 595 |
{
|
| 596 |
"epoch": 2.11,
|
| 597 |
+
"learning_rate": 2.403846153846154e-06,
|
| 598 |
+
"loss": 0.7929,
|
| 599 |
"step": 88
|
| 600 |
},
|
| 601 |
{
|
| 602 |
"epoch": 2.11,
|
| 603 |
+
"eval_loss": 1.0782241821289062,
|
| 604 |
+
"eval_runtime": 111.1236,
|
| 605 |
+
"eval_samples_per_second": 2.97,
|
| 606 |
+
"eval_steps_per_second": 0.099,
|
| 607 |
+
"eval_wer": 161.45617116397892,
|
| 608 |
"step": 88
|
| 609 |
},
|
| 610 |
{
|
| 611 |
"epoch": 2.12,
|
| 612 |
+
"learning_rate": 2.307692307692308e-06,
|
| 613 |
+
"loss": 0.774,
|
| 614 |
"step": 89
|
| 615 |
},
|
| 616 |
{
|
| 617 |
"epoch": 2.12,
|
| 618 |
+
"learning_rate": 2.211538461538462e-06,
|
| 619 |
+
"loss": 0.8693,
|
| 620 |
"step": 90
|
| 621 |
},
|
| 622 |
{
|
| 623 |
"epoch": 2.13,
|
| 624 |
+
"learning_rate": 2.1153846153846155e-06,
|
| 625 |
+
"loss": 0.8628,
|
| 626 |
"step": 91
|
| 627 |
},
|
| 628 |
{
|
| 629 |
"epoch": 2.14,
|
| 630 |
+
"learning_rate": 2.0192307692307692e-06,
|
| 631 |
+
"loss": 0.8199,
|
| 632 |
"step": 92
|
| 633 |
},
|
| 634 |
{
|
| 635 |
"epoch": 2.15,
|
| 636 |
+
"learning_rate": 1.923076923076923e-06,
|
| 637 |
+
"loss": 0.7995,
|
| 638 |
"step": 93
|
| 639 |
},
|
| 640 |
{
|
| 641 |
"epoch": 2.16,
|
| 642 |
+
"learning_rate": 1.826923076923077e-06,
|
| 643 |
+
"loss": 0.7824,
|
| 644 |
"step": 94
|
| 645 |
},
|
| 646 |
{
|
| 647 |
"epoch": 2.17,
|
| 648 |
+
"learning_rate": 1.7307692307692308e-06,
|
| 649 |
+
"loss": 0.7694,
|
| 650 |
"step": 95
|
| 651 |
},
|
| 652 |
{
|
| 653 |
"epoch": 2.18,
|
| 654 |
+
"learning_rate": 1.6346153846153848e-06,
|
| 655 |
+
"loss": 0.7526,
|
| 656 |
"step": 96
|
| 657 |
},
|
| 658 |
{
|
| 659 |
"epoch": 2.19,
|
| 660 |
+
"learning_rate": 1.5384615384615385e-06,
|
| 661 |
+
"loss": 0.7037,
|
| 662 |
"step": 97
|
| 663 |
},
|
| 664 |
{
|
| 665 |
"epoch": 2.2,
|
| 666 |
+
"learning_rate": 1.4423076923076924e-06,
|
| 667 |
+
"loss": 0.7366,
|
| 668 |
"step": 98
|
| 669 |
},
|
| 670 |
{
|
| 671 |
"epoch": 2.21,
|
| 672 |
+
"learning_rate": 1.3461538461538462e-06,
|
| 673 |
+
"loss": 0.7134,
|
| 674 |
"step": 99
|
| 675 |
},
|
| 676 |
{
|
| 677 |
"epoch": 2.21,
|
| 678 |
+
"eval_loss": 1.0737253427505493,
|
| 679 |
+
"eval_runtime": 96.9018,
|
| 680 |
+
"eval_samples_per_second": 3.406,
|
| 681 |
+
"eval_steps_per_second": 0.114,
|
| 682 |
+
"eval_wer": 177.1195912501996,
|
| 683 |
"step": 99
|
| 684 |
},
|
| 685 |
{
|
| 686 |
"epoch": 2.21,
|
| 687 |
+
"learning_rate": 1.2499999999999999e-06,
|
| 688 |
+
"loss": 0.7738,
|
| 689 |
"step": 100
|
| 690 |
},
|
| 691 |
{
|
| 692 |
"epoch": 2.22,
|
| 693 |
+
"learning_rate": 1.153846153846154e-06,
|
| 694 |
+
"loss": 0.7373,
|
| 695 |
"step": 101
|
| 696 |
},
|
| 697 |
{
|
| 698 |
"epoch": 2.23,
|
| 699 |
+
"learning_rate": 1.0576923076923078e-06,
|
| 700 |
+
"loss": 0.7623,
|
| 701 |
"step": 102
|
| 702 |
},
|
| 703 |
{
|
| 704 |
"epoch": 2.24,
|
| 705 |
+
"learning_rate": 9.615384615384615e-07,
|
| 706 |
+
"loss": 0.815,
|
| 707 |
"step": 103
|
| 708 |
},
|
| 709 |
{
|
| 710 |
"epoch": 2.25,
|
| 711 |
+
"learning_rate": 8.653846153846154e-07,
|
| 712 |
+
"loss": 0.829,
|
| 713 |
"step": 104
|
| 714 |
},
|
| 715 |
{
|
| 716 |
"epoch": 2.26,
|
| 717 |
+
"learning_rate": 7.692307692307693e-07,
|
| 718 |
+
"loss": 0.7113,
|
| 719 |
"step": 105
|
| 720 |
},
|
| 721 |
{
|
| 722 |
"epoch": 2.27,
|
| 723 |
+
"learning_rate": 6.730769230769231e-07,
|
| 724 |
+
"loss": 0.8106,
|
| 725 |
"step": 106
|
| 726 |
},
|
| 727 |
{
|
| 728 |
"epoch": 2.28,
|
| 729 |
+
"learning_rate": 5.76923076923077e-07,
|
| 730 |
+
"loss": 0.7909,
|
| 731 |
"step": 107
|
| 732 |
},
|
| 733 |
{
|
| 734 |
"epoch": 2.29,
|
| 735 |
+
"learning_rate": 4.807692307692307e-07,
|
| 736 |
+
"loss": 0.7564,
|
| 737 |
"step": 108
|
| 738 |
},
|
| 739 |
{
|
| 740 |
"epoch": 2.29,
|
| 741 |
+
"learning_rate": 3.8461538461538463e-07,
|
| 742 |
+
"loss": 0.8574,
|
| 743 |
"step": 109
|
| 744 |
},
|
| 745 |
{
|
| 746 |
"epoch": 2.3,
|
| 747 |
+
"learning_rate": 2.884615384615385e-07,
|
| 748 |
+
"loss": 0.8543,
|
| 749 |
"step": 110
|
| 750 |
},
|
| 751 |
{
|
| 752 |
"epoch": 2.3,
|
| 753 |
+
"eval_loss": 1.0717276334762573,
|
| 754 |
+
"eval_runtime": 95.923,
|
| 755 |
+
"eval_samples_per_second": 3.44,
|
| 756 |
+
"eval_steps_per_second": 0.115,
|
| 757 |
+
"eval_wer": 187.5459045186013,
|
| 758 |
"step": 110
|
| 759 |
},
|
| 760 |
{
|
| 761 |
"epoch": 2.31,
|
| 762 |
+
"learning_rate": 1.9230769230769231e-07,
|
| 763 |
+
"loss": 0.824,
|
| 764 |
"step": 111
|
| 765 |
},
|
| 766 |
{
|
| 767 |
"epoch": 2.32,
|
| 768 |
+
"learning_rate": 9.615384615384616e-08,
|
| 769 |
+
"loss": 0.8352,
|
| 770 |
"step": 112
|
| 771 |
},
|
| 772 |
{
|
| 773 |
"epoch": 2.32,
|
| 774 |
"step": 112,
|
| 775 |
"total_flos": 1.7415399333888e+17,
|
| 776 |
+
"train_loss": 1.0522634956453527,
|
| 777 |
+
"train_runtime": 1735.2585,
|
| 778 |
+
"train_samples_per_second": 4.131,
|
| 779 |
+
"train_steps_per_second": 0.065
|
| 780 |
}
|
| 781 |
],
|
| 782 |
"max_steps": 112,
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 3579
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4578ee205fe35fdbb3b98224d5f855f806bd9eedb29095f0f731a29261db2d94
|
| 3 |
size 3579
|