HF中国镜像站's logo

iammahadev
/
w2v-bert-2-malayalam-combo-v1

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
wav2vec2-bert
Generated from Trainer
Model card Files Metrics Community
iammahadev commited on
Commit
8bf6323
·
verified ·
1 Parent(s): 8cc0a0d

End of training

Browse files
Files changed (5) hide show
  1. README.md +55 -3
  2. config.json +2 -2
  3. model.safetensors +2 -2
  4. runs/Jul19_15-49-36_n-3404bb5e-b53f-4acb-a141-8b7508ac6565-0/events.out.tfevents.1721404791.n-3404bb5e-b53f-4acb-a141-8b7508ac6565-0.429.0 +3 -0
  5. training_args.bin +2 -2
README.md CHANGED
@@ -1,8 +1,10 @@
1
  ---
2
- base_model: facebook/w2v-bert-2.0
3
  license: mit
 
4
  tags:
5
  - generated_from_trainer
 
 
6
  model-index:
7
  - name: w2v-bert-2-malayalam-combo-v1
8
  results: []
@@ -14,6 +16,9 @@ should probably proofread and complete it, then remove this comment. -->
14
  # w2v-bert-2-malayalam-combo-v1
15
 
16
  This model is a fine-tuned version of [facebook/w2v-bert-2.0](https://huggingface.co/facebook/w2v-bert-2.0) on the None dataset.
 
 
 
17
 
18
  ## Model description
19
 
@@ -44,9 +49,56 @@ The following hyperparameters were used during training:
44
  - num_epochs: 10
45
  - mixed_precision_training: Native AMP
46
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
47
  ### Framework versions
48
 
49
  - Transformers 4.42.4
50
- - Pytorch 2.2.1+cu121
51
- - Datasets 2.20.0
52
  - Tokenizers 0.19.1
 
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-malayalam-combo-v1
10
  results: []
 
16
  # w2v-bert-2-malayalam-combo-v1
17
 
18
  This model is a fine-tuned version of [facebook/w2v-bert-2.0](https://huggingface.co/facebook/w2v-bert-2.0) on the None dataset.
19
+ It achieves the following results on the evaluation set:
20
+ - Loss: inf
21
+ - Wer: 0.1007
22
 
23
  ## Model description
24
 
 
49
  - num_epochs: 10
50
  - mixed_precision_training: Native AMP
51
 
52
+ ### Training results
53
+
54
+ | Training Loss | Epoch | Step | Validation Loss | Wer |
55
+ |:-------------:|:------:|:-----:|:---------------:|:------:|
56
+ | 1.9859 | 0.2432 | 300 | inf | 0.4513 |
57
+ | 0.2903 | 0.4864 | 600 | inf | 0.4107 |
58
+ | 0.2294 | 0.7296 | 900 | inf | 0.3331 |
59
+ | 0.2075 | 0.9728 | 1200 | inf | 0.2968 |
60
+ | 0.1737 | 1.2161 | 1500 | inf | 0.2862 |
61
+ | 0.1561 | 1.4593 | 1800 | inf | 0.2603 |
62
+ | 0.1435 | 1.7025 | 2100 | inf | 0.2496 |
63
+ | 0.1388 | 1.9457 | 2400 | inf | 0.2329 |
64
+ | 0.1213 | 2.1889 | 2700 | inf | 0.2271 |
65
+ | 0.1168 | 2.4321 | 3000 | inf | 0.2202 |
66
+ | 0.1086 | 2.6753 | 3300 | inf | 0.2273 |
67
+ | 0.1131 | 2.9185 | 3600 | inf | 0.2132 |
68
+ | 0.0951 | 3.1617 | 3900 | inf | 0.2068 |
69
+ | 0.0851 | 3.4049 | 4200 | inf | 0.2075 |
70
+ | 0.0905 | 3.6482 | 4500 | inf | 0.1969 |
71
+ | 0.0811 | 3.8914 | 4800 | inf | 0.1941 |
72
+ | 0.0754 | 4.1346 | 5100 | inf | 0.1717 |
73
+ | 0.0653 | 4.3778 | 5400 | inf | 0.1704 |
74
+ | 0.0663 | 4.6210 | 5700 | inf | 0.1737 |
75
+ | 0.0635 | 4.8642 | 6000 | inf | 0.1551 |
76
+ | 0.0607 | 5.1074 | 6300 | inf | 0.1479 |
77
+ | 0.05 | 5.3506 | 6600 | inf | 0.1478 |
78
+ | 0.0519 | 5.5938 | 6900 | inf | 0.1441 |
79
+ | 0.048 | 5.8370 | 7200 | inf | 0.1410 |
80
+ | 0.0428 | 6.0803 | 7500 | inf | 0.1362 |
81
+ | 0.0344 | 6.3235 | 7800 | inf | 0.1325 |
82
+ | 0.0344 | 6.5667 | 8100 | inf | 0.1242 |
83
+ | 0.0361 | 6.8099 | 8400 | inf | 0.1247 |
84
+ | 0.031 | 7.0531 | 8700 | inf | 0.1227 |
85
+ | 0.0256 | 7.2963 | 9000 | inf | 0.1175 |
86
+ | 0.023 | 7.5395 | 9300 | inf | 0.1172 |
87
+ | 0.0223 | 7.7827 | 9600 | inf | 0.1161 |
88
+ | 0.0203 | 8.0259 | 9900 | inf | 0.1099 |
89
+ | 0.014 | 8.2692 | 10200 | inf | 0.1094 |
90
+ | 0.0158 | 8.5124 | 10500 | inf | 0.1081 |
91
+ | 0.0147 | 8.7556 | 10800 | inf | 0.1078 |
92
+ | 0.0132 | 8.9988 | 11100 | inf | 0.1049 |
93
+ | 0.008 | 9.2420 | 11400 | inf | 0.1048 |
94
+ | 0.0081 | 9.4852 | 11700 | inf | 0.1010 |
95
+ | 0.0081 | 9.7284 | 12000 | inf | 0.1010 |
96
+ | 0.0094 | 9.9716 | 12300 | inf | 0.1007 |
97
+
98
+
99
  ### Framework versions
100
 
101
  - Transformers 4.42.4
102
+ - Pytorch 2.3.1+cu121
103
+ - Datasets 2.19.1
104
  - Tokenizers 0.19.1
config.json CHANGED
@@ -47,7 +47,7 @@
47
  "num_hidden_layers": 24,
48
  "num_negatives": 100,
49
  "output_hidden_size": 1024,
50
- "pad_token_id": 75,
51
  "position_embeddings_type": "relative_key",
52
  "proj_codevector_dim": 768,
53
  "right_max_position_embeddings": 8,
@@ -77,6 +77,6 @@
77
  "transformers_version": "4.42.4",
78
  "use_intermediate_ffn_before_adapter": false,
79
  "use_weighted_layer_sum": false,
80
- "vocab_size": 78,
81
  "xvector_output_dim": 512
82
  }
 
47
  "num_hidden_layers": 24,
48
  "num_negatives": 100,
49
  "output_hidden_size": 1024,
50
+ "pad_token_id": 73,
51
  "position_embeddings_type": "relative_key",
52
  "proj_codevector_dim": 768,
53
  "right_max_position_embeddings": 8,
 
77
  "transformers_version": "4.42.4",
78
  "use_intermediate_ffn_before_adapter": false,
79
  "use_weighted_layer_sum": false,
80
+ "vocab_size": 76,
81
  "xvector_output_dim": 512
82
  }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ea65f34a8a5b2719e54d2bbcaffe8a3865db48567f85b6260685d6134fe06cc8
3
- size 2423134360
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a6dbc8d9b6bf73b0b5b1967ac5116502843ae5fbdef86b261f4dee77db2bea4d
3
+ size 2423126160
runs/Jul19_15-49-36_n-3404bb5e-b53f-4acb-a141-8b7508ac6565-0/events.out.tfevents.1721404791.n-3404bb5e-b53f-4acb-a141-8b7508ac6565-0.429.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:16bb11ae66799b97a2c9633663edd52cc81793709387b3671f152d959c2b46ec
3
+ size 28156
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2a6973d59c85288ac4f519f9b7ca4665f6766cf4c4daa55ba115e1494935a360
3
- size 5112
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:253dd23ef31ea4f503ee96f5a0b1fc18b9f9088ffe64bd50cde3a33dd91036c0
3
+ size 5176