Hugging Face's logo

Jbautistas
/
checkpoints

Automatic Speech Recognition
Transformers
Safetensors
Spanish
whisper
Generated from Trainer
8-bit precision
bitsandbytes
Model card Files
xet
Community
Jbautistas commited on
Commit
f89ca32
·
verified ·
1 Parent(s): 0690774

End of training

Browse files
Files changed (3) hide show
  1. README.md +13 -7
  2. model.safetensors +2 -2
  3. training_args.bin +1 -1
README.md CHANGED
@@ -6,6 +6,8 @@ license: apache-2.0
6
  base_model: openai/whisper-large-v3
7
  tags:
8
  - generated_from_trainer
 
 
9
  model-index:
10
  - name: Whisper large LoRA Merged Es - Jbautistas
11
  results: []
@@ -18,7 +20,8 @@ should probably proofread and complete it, then remove this comment. -->
18
 
19
  This model is a fine-tuned version of [openai/whisper-large-v3](https://huggingface.co/openai/whisper-large-v3) on the None dataset.
20
  It achieves the following results on the evaluation set:
21
- - Loss: 0.6809
 
22
 
23
  ## Model description
24
 
@@ -41,19 +44,22 @@ The following hyperparameters were used during training:
41
  - train_batch_size: 8
42
  - eval_batch_size: 8
43
  - seed: 42
 
 
44
  - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
45
  - lr_scheduler_type: linear
 
46
  - num_epochs: 30
47
  - mixed_precision_training: Native AMP
48
 
49
  ### Training results
50
 
51
- | Training Loss | Epoch | Step | Validation Loss |
52
- |:-------------:|:-------:|:----:|:---------------:|
53
- | 0.294 | 6.8966 | 200 | 0.5184 |
54
- | 0.1686 | 13.7931 | 400 | 0.6167 |
55
- | 0.1306 | 20.6897 | 600 | 0.6527 |
56
- | 0.1143 | 27.5862 | 800 | 0.6809 |
57
 
58
 
59
  ### Framework versions
 
6
  base_model: openai/whisper-large-v3
7
  tags:
8
  - generated_from_trainer
9
+ metrics:
10
+ - wer
11
  model-index:
12
  - name: Whisper large LoRA Merged Es - Jbautistas
13
  results: []
 
20
 
21
  This model is a fine-tuned version of [openai/whisper-large-v3](https://huggingface.co/openai/whisper-large-v3) on the None dataset.
22
  It achieves the following results on the evaluation set:
23
+ - Loss: 0.7153
24
+ - Wer: 58.9379
25
 
26
  ## Model description
27
 
 
44
  - train_batch_size: 8
45
  - eval_batch_size: 8
46
  - seed: 42
47
+ - gradient_accumulation_steps: 4
48
+ - total_train_batch_size: 32
49
  - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
50
  - lr_scheduler_type: linear
51
+ - lr_scheduler_warmup_steps: 500
52
  - num_epochs: 30
53
  - mixed_precision_training: Native AMP
54
 
55
  ### Training results
56
 
57
+ | Training Loss | Epoch | Step | Validation Loss | Wer |
58
+ |:-------------:|:-------:|:----:|:---------------:|:-------:|
59
+ | 1.4943 | 6.2759 | 50 | 1.1849 | 58.2648 |
60
+ | 1.1692 | 12.5517 | 100 | 1.0537 | 56.2453 |
61
+ | 0.9958 | 18.8276 | 150 | 0.8947 | 54.8990 |
62
+ | 0.7308 | 25.0 | 200 | 0.7153 | 58.9379 |
63
 
64
 
65
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:734bffb12c25b8ab44708e1e609175de30d7a9534d6a2cbf09aa349e924a8296
3
- size 1773492440
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8e0d45115d84043722ccc82be41e1aef7ddb20d56253af49d622abc7417ae91c
3
+ size 1773489880
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4033888a576f3eee17d7c24bdd78a83b77ccc818cbffd2e67a63d01b504b6425
3
  size 5496
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:943ac7bc7fc365f2500a10c2faaa689a01076ff704329da2ae31ffd48da1b409
3
  size 5496