Hugging Face's logo

csikasote
/
mms-1b-all-swagen-combined-m50f50-dnn-42-0.04

Automatic Speech Recognition
Transformers
Safetensors
wav2vec2
swagen
mms
Generated from Trainer
Model card Files
xet
Community
csikasote commited on
Commit
17d3f73
·
verified ·
1 Parent(s): 5f37732

Model save

Browse files
Files changed (2) hide show
  1. README.md +15 -18
  2. model.safetensors +1 -1
README.md CHANGED
@@ -1,11 +1,8 @@
1
  ---
2
  library_name: transformers
3
  license: cc-by-nc-4.0
4
- base_model: facebook/mms-1b-all
5
  tags:
6
- - automatic-speech-recognition
7
- - swagen
8
- - mms
9
  - generated_from_trainer
10
  model-index:
11
  - name: mms-1b-all-swagen-combined-m50f50-dnn-42-0.04
@@ -17,10 +14,10 @@ should probably proofread and complete it, then remove this comment. -->
17
 
18
  # mms-1b-all-swagen-combined-m50f50-dnn-42-0.04
19
 
20
- This model is a fine-tuned version of [facebook/mms-1b-all](https://huggingface.co/facebook/mms-1b-all) on the SWAGEN - DEB dataset.
21
  It achieves the following results on the evaluation set:
22
- - Loss: 15.0948
23
- - Cer: 4.1360
24
 
25
  ## Model description
26
 
@@ -47,21 +44,21 @@ The following hyperparameters were used during training:
47
  - total_train_batch_size: 16
48
  - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
49
  - lr_scheduler_type: linear
50
- - lr_scheduler_warmup_steps: 250
51
- - num_epochs: 15.0
 
52
  - mixed_precision_training: Native AMP
53
 
54
  ### Training results
55
 
56
- | Training Loss | Epoch | Step | Validation Loss | Cer |
57
- |:-------------:|:-------:|:----:|:---------------:|:------:|
58
- | 13.0486 | 1.9723 | 250 | 15.0948 | 4.1382 |
59
- | 25.0654 | 3.9407 | 500 | 15.0948 | 4.1373 |
60
- | 17.1936 | 5.9091 | 750 | 15.0947 | 4.1391 |
61
- | 10.2055 | 7.8775 | 1000 | 15.0947 | 4.1376 |
62
- | 11.2981 | 9.8458 | 1250 | 15.0946 | 4.1384 |
63
- | 13.1665 | 11.8142 | 1500 | 15.0946 | 4.1380 |
64
- | 11.5061 | 13.7826 | 1750 | 15.0948 | 4.1370 |
65
 
66
 
67
  ### Framework versions
 
1
  ---
2
  library_name: transformers
3
  license: cc-by-nc-4.0
4
+ base_model: csikasote/mms-1b-all-swagen-combined-m50f50-dnn-42-0.04
5
  tags:
 
 
 
6
  - generated_from_trainer
7
  model-index:
8
  - name: mms-1b-all-swagen-combined-m50f50-dnn-42-0.04
 
14
 
15
  # mms-1b-all-swagen-combined-m50f50-dnn-42-0.04
16
 
17
+ This model is a fine-tuned version of [csikasote/mms-1b-all-swagen-combined-m50f50-dnn-42-0.04](https://huggingface.co/csikasote/mms-1b-all-swagen-combined-m50f50-dnn-42-0.04) on an unknown dataset.
18
  It achieves the following results on the evaluation set:
19
+ - Loss: 0.3241
20
+ - Cer: 0.0790
21
 
22
  ## Model description
23
 
 
44
  - total_train_batch_size: 16
45
  - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
46
  - lr_scheduler_type: linear
47
+ - lr_scheduler_warmup_ratio: 0.05
48
+ - lr_scheduler_warmup_steps: 100
49
+ - num_epochs: 30.0
50
  - mixed_precision_training: Native AMP
51
 
52
  ### Training results
53
 
54
+ | Training Loss | Epoch | Step | Validation Loss | Cer |
55
+ |:-------------:|:------:|:----:|:---------------:|:------:|
56
+ | 0.4391 | 0.7905 | 100 | 0.4093 | 0.0927 |
57
+ | 0.6661 | 1.5771 | 200 | 0.3816 | 0.0872 |
58
+ | 0.0181 | 2.3636 | 300 | 0.3057 | 0.0738 |
59
+ | 0.2996 | 3.1502 | 400 | 0.4619 | 0.1165 |
60
+ | 0.2144 | 3.9407 | 500 | 0.3101 | 0.0744 |
61
+ | 0.0264 | 4.7273 | 600 | 0.3241 | 0.0790 |
 
62
 
63
 
64
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7cabebdb612b033dc5d56a949045c2e4071d23369c943cc02bc654e1f92aeae0
3
  size 4209756988
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:461f7afa428aaf148250a57d3cef07740fad9e861ace99a2312fe963f889723c
3
  size 4209756988