Hugging Face's logo

emonidi
/
whisper-tiny-order

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
whisper
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
emonidi commited on
Commit
4faf701
•
1 Parent(s): 2d4f76e

emonidi/whisper-tiny-order

Browse files
Files changed (12) hide show
  1. README.md +41 -17
  2. config.json +1 -1
  3. generation_config.json +1 -2
  4. model.safetensors +1 -1
  5. runs/Dec22_08-21-42_0871cb1adcb1/events.out.tfevents.1703233309.0871cb1adcb1.1346.0 +3 -0
  6. runs/Dec22_08-22-22_0871cb1adcb1/events.out.tfevents.1703233347.0871cb1adcb1.1346.1 +3 -0
  7. runs/Dec22_08-23-00_0871cb1adcb1/events.out.tfevents.1703233392.0871cb1adcb1.1346.2 +3 -0
  8. runs/Dec22_08-30-16_1d25928f9346/events.out.tfevents.1703233831.1d25928f9346.1005.0 +3 -0
  9. runs/Dec23_16-17-46_fe9daa3159a6/events.out.tfevents.1703348274.fe9daa3159a6.6951.1 +3 -0
  10. runs/Dec23_16-49-38_fe9daa3159a6/events.out.tfevents.1703350184.fe9daa3159a6.6951.2 +3 -0
  11. runs/Mar02_13-16-50_0785b549392f/events.out.tfevents.1709385418.0785b549392f.6415.0 +3 -0
  12. training_args.bin +2 -2
README.md CHANGED
@@ -17,7 +17,7 @@ should probably proofread and complete it, then remove this comment. -->
17
 
18
  This model is a fine-tuned version of [nandovallec/whisper-tiny-bg-l](https://huggingface.co/nandovallec/whisper-tiny-bg-l) on the None dataset.
19
  It achieves the following results on the evaluation set:
20
- - Loss: 0.0111
21
  - Wer: 0.0
22
 
23
  ## Model description
@@ -38,31 +38,55 @@ More information needed
38
 
39
  The following hyperparameters were used during training:
40
  - learning_rate: 1e-05
41
- - train_batch_size: 32
42
- - eval_batch_size: 32
43
  - seed: 42
44
- - gradient_accumulation_steps: 4
45
- - total_train_batch_size: 128
46
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
47
  - lr_scheduler_type: linear
48
- - training_steps: 30
49
  - mixed_precision_training: Native AMP
50
 
51
  ### Training results
52
 
53
- | Training Loss | Epoch | Step | Validation Loss | Wer |
54
- |:-------------:|:-----:|:----:|:---------------:|:-------:|
55
- | 0.1524 | 5.0 | 5 | 0.2217 | 11.1111 |
56
- | 0.0766 | 10.0 | 10 | 0.1033 | 8.8889 |
57
- | 0.0347 | 15.0 | 15 | 0.0563 | 11.1111 |
58
- | 0.0166 | 20.0 | 20 | 0.0274 | 2.2222 |
59
- | 0.0084 | 25.0 | 25 | 0.0145 | 0.0 |
60
- | 0.0051 | 30.0 | 30 | 0.0111 | 0.0 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
61
 
62
 
63
  ### Framework versions
64
 
65
- - Transformers 4.35.2
66
  - Pytorch 2.1.0+cu121
67
- - Datasets 2.15.0
68
- - Tokenizers 0.15.0
 
17
 
18
  This model is a fine-tuned version of [nandovallec/whisper-tiny-bg-l](https://huggingface.co/nandovallec/whisper-tiny-bg-l) on the None dataset.
19
  It achieves the following results on the evaluation set:
20
+ - Loss: 0.0015
21
  - Wer: 0.0
22
 
23
  ## Model description
 
38
 
39
  The following hyperparameters were used during training:
40
  - learning_rate: 1e-05
41
+ - train_batch_size: 16
42
+ - eval_batch_size: 8
43
  - seed: 42
44
+ - gradient_accumulation_steps: 16
45
+ - total_train_batch_size: 256
46
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
47
  - lr_scheduler_type: linear
48
+ - training_steps: 150
49
  - mixed_precision_training: Native AMP
50
 
51
  ### Training results
52
 
53
+ | Training Loss | Epoch | Step | Validation Loss | Wer |
54
+ |:-------------:|:-----:|:----:|:---------------:|:--------:|
55
+ | 0.5799 | 5.0 | 5 | 1.8414 | 116.3934 |
56
+ | 0.2647 | 10.0 | 10 | 0.7719 | 61.4754 |
57
+ | 0.1199 | 15.0 | 15 | 0.3593 | 34.4262 |
58
+ | 0.063 | 20.0 | 20 | 0.1827 | 18.8525 |
59
+ | 0.0257 | 25.0 | 25 | 0.0710 | 4.9180 |
60
+ | 0.0103 | 30.0 | 30 | 0.0294 | 0.8197 |
61
+ | 0.0045 | 35.0 | 35 | 0.0149 | 0.0 |
62
+ | 0.0028 | 40.0 | 40 | 0.0094 | 0.0 |
63
+ | 0.0019 | 45.0 | 45 | 0.0064 | 0.0 |
64
+ | 0.0014 | 50.0 | 50 | 0.0048 | 0.0 |
65
+ | 0.0011 | 55.0 | 55 | 0.0038 | 0.0 |
66
+ | 0.0009 | 60.0 | 60 | 0.0032 | 0.0 |
67
+ | 0.0008 | 65.0 | 65 | 0.0028 | 0.0 |
68
+ | 0.0007 | 70.0 | 70 | 0.0025 | 0.0 |
69
+ | 0.0006 | 75.0 | 75 | 0.0023 | 0.0 |
70
+ | 0.0006 | 80.0 | 80 | 0.0022 | 0.0 |
71
+ | 0.0006 | 85.0 | 85 | 0.0021 | 0.0 |
72
+ | 0.0005 | 90.0 | 90 | 0.0020 | 0.0 |
73
+ | 0.0005 | 95.0 | 95 | 0.0019 | 0.0 |
74
+ | 0.0005 | 100.0 | 100 | 0.0018 | 0.0 |
75
+ | 0.0005 | 105.0 | 105 | 0.0017 | 0.0 |
76
+ | 0.0005 | 110.0 | 110 | 0.0017 | 0.0 |
77
+ | 0.0004 | 115.0 | 115 | 0.0017 | 0.0 |
78
+ | 0.0004 | 120.0 | 120 | 0.0016 | 0.0 |
79
+ | 0.0004 | 125.0 | 125 | 0.0016 | 0.0 |
80
+ | 0.0004 | 130.0 | 130 | 0.0016 | 0.0 |
81
+ | 0.0004 | 135.0 | 135 | 0.0016 | 0.0 |
82
+ | 0.0004 | 140.0 | 140 | 0.0015 | 0.0 |
83
+ | 0.0004 | 145.0 | 145 | 0.0015 | 0.0 |
84
+ | 0.0004 | 150.0 | 150 | 0.0015 | 0.0 |
85
 
86
 
87
  ### Framework versions
88
 
89
+ - Transformers 4.38.1
90
  - Pytorch 2.1.0+cu121
91
+ - Datasets 2.18.0
92
+ - Tokenizers 0.15.2
config.json CHANGED
@@ -45,7 +45,7 @@
45
  "scale_embedding": false,
46
  "suppress_tokens": [],
47
  "torch_dtype": "float32",
48
- "transformers_version": "4.35.2",
49
  "use_cache": false,
50
  "use_weighted_layer_sum": false,
51
  "vocab_size": 51865
 
45
  "scale_embedding": false,
46
  "suppress_tokens": [],
47
  "torch_dtype": "float32",
48
+ "transformers_version": "4.38.1",
49
  "use_cache": false,
50
  "use_weighted_layer_sum": false,
51
  "vocab_size": 51865
generation_config.json CHANGED
@@ -8,7 +8,6 @@
8
  "eos_token_id": 50257,
9
  "max_length": 448,
10
  "pad_token_id": 50257,
11
- "return_timestamps": false,
12
  "suppress_tokens": [],
13
- "transformers_version": "4.35.2"
14
  }
 
8
  "eos_token_id": 50257,
9
  "max_length": 448,
10
  "pad_token_id": 50257,
 
11
  "suppress_tokens": [],
12
+ "transformers_version": "4.38.1"
13
  }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e113aa4f73e0e58d9cebfb44c014ed51287409293fd23625d0c4b12209e8aa73
3
  size 151061672
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7793251c92c891bc99d579abd0be67bbeeb0f05124ef0b60ac9de2c8834c41d8
3
  size 151061672
runs/Dec22_08-21-42_0871cb1adcb1/events.out.tfevents.1703233309.0871cb1adcb1.1346.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:038b1c24d2e3e9eacb83861f882afe7a5b46a1930b96acca319b650da3725f74
3
+ size 5050
runs/Dec22_08-22-22_0871cb1adcb1/events.out.tfevents.1703233347.0871cb1adcb1.1346.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7205360fcda0d3000d4b5520877759b7595fa9314ec0d0edcf67822a3c767445
3
+ size 5049
runs/Dec22_08-23-00_0871cb1adcb1/events.out.tfevents.1703233392.0871cb1adcb1.1346.2 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ddd62cd85b70818b5005ab62accd8501f0b96a374726dc6e7faa61d9b525d987
3
+ size 5203
runs/Dec22_08-30-16_1d25928f9346/events.out.tfevents.1703233831.1d25928f9346.1005.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1781c17ab381700f481108967451233109614abd33ef2d3f72e9ad4e069d5ab6
3
+ size 5819
runs/Dec23_16-17-46_fe9daa3159a6/events.out.tfevents.1703348274.fe9daa3159a6.6951.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b18551293a74eb6bb39cc7fc429c0546d5d6ea7cdc4069f1ff6a4ff6cda98247
3
+ size 16213
runs/Dec23_16-49-38_fe9daa3159a6/events.out.tfevents.1703350184.fe9daa3159a6.6951.2 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5f886ad68be46627e4d45dca9d937a54b3095ee079af0bf6f8521f04bc172f9c
3
+ size 37959
runs/Mar02_13-16-50_0785b549392f/events.out.tfevents.1709385418.0785b549392f.6415.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8a3ca2e80a422da01fef835a0be3ecdb50f4ec50f70167fa7559027da21904dc
3
+ size 46284
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ddc8538277a0749917809f12b0a48b85cf58b054ee86aad92bc6a622964c5831
3
- size 4728
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5694683483cff3c9c7a0f623bd2331a82b9bf74979fdc4876065f257a2991b8e
3
+ size 5048