Hugging Face's logo

Bajiyo
/
w2v-bert-2.0-malayalam_mixeddataset_thre

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
wav2vec2-bert
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
Bajiyo commited on
Commit
42da1d9
•
1 Parent(s): 40ba827

End of training

Browse files
Files changed (3) hide show
  1. README.md +24 -24
  2. model.safetensors +1 -1
  3. runs/Apr23_11-22-54_kudsit-dgxserver/events.out.tfevents.1713852369.kudsit-dgxserver.2902740.0 +2 -2
README.md CHANGED
@@ -1,10 +1,10 @@
1
  ---
2
  license: mit
 
3
  tags:
4
  - generated_from_trainer
5
  metrics:
6
  - wer
7
- base_model: facebook/w2v-bert-2.0
8
  model-index:
9
  - name: w2v-bert-2.0-malayalam_mixeddataset_thre
10
  results: []
@@ -17,8 +17,8 @@ should probably proofread and complete it, then remove this comment. -->
17
 
18
  This model is a fine-tuned version of [facebook/w2v-bert-2.0](https://huggingface.co/facebook/w2v-bert-2.0) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
- - Loss: 0.1738
21
- - Wer: 0.1162
22
 
23
  ## Model description
24
 
@@ -53,27 +53,27 @@ The following hyperparameters were used during training:
53
 
54
  | Training Loss | Epoch | Step | Validation Loss | Wer |
55
  |:-------------:|:-----:|:-----:|:---------------:|:------:|
56
- | 1.1027 | 0.47 | 600 | 0.3414 | 0.4257 |
57
- | 0.1646 | 0.95 | 1200 | 0.2676 | 0.3585 |
58
- | 0.1197 | 1.42 | 1800 | 0.2197 | 0.2988 |
59
- | 0.1028 | 1.9 | 2400 | 0.2114 | 0.2849 |
60
- | 0.0809 | 2.37 | 3000 | 0.1947 | 0.2478 |
61
- | 0.0738 | 2.85 | 3600 | 0.1593 | 0.2411 |
62
- | 0.0615 | 3.32 | 4200 | 0.1657 | 0.2075 |
63
- | 0.0528 | 3.8 | 4800 | 0.1587 | 0.1986 |
64
- | 0.0425 | 4.27 | 5400 | 0.1687 | 0.1749 |
65
- | 0.0377 | 4.74 | 6000 | 0.1566 | 0.1796 |
66
- | 0.0336 | 5.22 | 6600 | 0.1628 | 0.1647 |
67
- | 0.0269 | 5.69 | 7200 | 0.1580 | 0.1754 |
68
- | 0.0237 | 6.17 | 7800 | 0.1564 | 0.1530 |
69
- | 0.0189 | 6.64 | 8400 | 0.1593 | 0.1446 |
70
- | 0.0159 | 7.12 | 9000 | 0.1370 | 0.1376 |
71
- | 0.0118 | 7.59 | 9600 | 0.1586 | 0.1421 |
72
- | 0.012 | 8.07 | 10200 | 0.1567 | 0.1281 |
73
- | 0.0071 | 8.54 | 10800 | 0.1645 | 0.1227 |
74
- | 0.0072 | 9.02 | 11400 | 0.1634 | 0.1202 |
75
- | 0.0036 | 9.49 | 12000 | 0.1719 | 0.1189 |
76
- | 0.0036 | 9.96 | 12600 | 0.1738 | 0.1162 |
77
 
78
 
79
  ### Framework versions
 
1
  ---
2
  license: mit
3
+ base_model: facebook/w2v-bert-2.0
4
  tags:
5
  - generated_from_trainer
6
  metrics:
7
  - wer
 
8
  model-index:
9
  - name: w2v-bert-2.0-malayalam_mixeddataset_thre
10
  results: []
 
17
 
18
  This model is a fine-tuned version of [facebook/w2v-bert-2.0](https://huggingface.co/facebook/w2v-bert-2.0) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
+ - Loss: 0.1709
21
+ - Wer: 0.1197
22
 
23
  ## Model description
24
 
 
53
 
54
  | Training Loss | Epoch | Step | Validation Loss | Wer |
55
  |:-------------:|:-----:|:-----:|:---------------:|:------:|
56
+ | 1.1907 | 0.47 | 600 | 0.3890 | 0.4765 |
57
+ | 0.1663 | 0.95 | 1200 | 0.2528 | 0.3528 |
58
+ | 0.1207 | 1.42 | 1800 | 0.2176 | 0.2849 |
59
+ | 0.1017 | 1.9 | 2400 | 0.2021 | 0.2625 |
60
+ | 0.0833 | 2.37 | 3000 | 0.2032 | 0.2456 |
61
+ | 0.076 | 2.85 | 3600 | 0.1880 | 0.2376 |
62
+ | 0.0625 | 3.32 | 4200 | 0.1946 | 0.2247 |
63
+ | 0.0552 | 3.8 | 4800 | 0.1701 | 0.2247 |
64
+ | 0.0441 | 4.27 | 5400 | 0.1627 | 0.1759 |
65
+ | 0.0392 | 4.74 | 6000 | 0.1629 | 0.1829 |
66
+ | 0.0362 | 5.22 | 6600 | 0.1723 | 0.1605 |
67
+ | 0.0278 | 5.69 | 7200 | 0.1600 | 0.1665 |
68
+ | 0.0248 | 6.17 | 7800 | 0.1557 | 0.1446 |
69
+ | 0.0197 | 6.64 | 8400 | 0.1524 | 0.1505 |
70
+ | 0.0176 | 7.12 | 9000 | 0.1580 | 0.1339 |
71
+ | 0.0129 | 7.59 | 9600 | 0.1528 | 0.1411 |
72
+ | 0.0125 | 8.07 | 10200 | 0.1502 | 0.1299 |
73
+ | 0.0076 | 8.54 | 10800 | 0.1711 | 0.1189 |
74
+ | 0.0076 | 9.02 | 11400 | 0.1689 | 0.1237 |
75
+ | 0.0041 | 9.49 | 12000 | 0.1708 | 0.1227 |
76
+ | 0.0041 | 9.96 | 12600 | 0.1709 | 0.1197 |
77
 
78
 
79
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:269366fb88f5d5f5010a8e967caace2a8b063305ea6c9cfb3f21b5b8e76795a2
3
  size 2423130260
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e9632dc5b41030ae84c029a4fba14f714c288e72830fe0be1f90a1145b6b9792
3
  size 2423130260
runs/Apr23_11-22-54_kudsit-dgxserver/events.out.tfevents.1713852369.kudsit-dgxserver.2902740.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:59d9048e6da0115994af20e731522ac35b258b4cc75ac695104ecb513e79bb7b
3
- size 17041
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3034294e1fa59bf9a39cca90d07ac18defdf1b4767b718a2a3d95a574e1a78aa
3
+ size 17395