Hugging Face's logo

arslanarjumand
/
wav2vec-reptiles

Audio Classification
Transformers
Safetensors
wav2vec2-bert
Generated from Trainer
Inference Endpoints
Model card Files Community
arslanarjumand commited on
Commit
be1de82
•
1 Parent(s): ed9d65a

arslanarjumand/wav2vec-reptiles

Browse files
Files changed (4) hide show
  1. README.md +22 -17
  2. config.json +1 -1
  3. model.safetensors +1 -1
  4. training_args.bin +1 -1
README.md CHANGED
@@ -1,6 +1,6 @@
1
  ---
2
  license: mit
3
- base_model: facebook/w2v-bert-2.0
4
  tags:
5
  - generated_from_trainer
6
  model-index:
@@ -13,13 +13,13 @@ should probably proofread and complete it, then remove this comment. -->
13
 
14
  # wav2vec-reptiles
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
  It achieves the following results on the evaluation set:
18
- - Loss: 483.1183
19
- - Pcc Accuracy: -0.0911
20
- - Pcc Fluency: -0.0702
21
- - Pcc Total Score: -0.0823
22
- - Pcc Content: -0.0799
23
 
24
  ## Model description
25
 
@@ -47,22 +47,27 @@ The following hyperparameters were used during training:
47
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
48
  - lr_scheduler_type: cosine
49
  - lr_scheduler_warmup_ratio: 0.4
50
- - num_epochs: 10
51
  - mixed_precision_training: Native AMP
52
 
53
  ### Training results
54
 
55
  | Training Loss | Epoch | Step | Validation Loss | Pcc Accuracy | Pcc Fluency | Pcc Total Score | Pcc Content |
56
  |:-------------:|:-----:|:----:|:---------------:|:------------:|:-----------:|:---------------:|:-----------:|
57
- | 3259.2658 | 1.07 | 500 | 2804.9829 | -0.2619 | -0.2376 | -0.2620 | -0.2464 |
58
- | 1980.8576 | 2.13 | 1000 | 2444.1970 | -0.2591 | -0.2359 | -0.2573 | -0.2394 |
59
- | 1728.8568 | 3.2 | 1500 | 1480.2877 | -0.2539 | -0.2310 | -0.2519 | -0.2337 |
60
- | 490.517 | 4.27 | 2000 | 577.7591 | -0.2334 | -0.2102 | -0.2303 | -0.2141 |
61
- | 761.9771 | 5.34 | 2500 | 503.8331 | -0.1885 | -0.1646 | -0.1829 | -0.1714 |
62
- | 286.5463 | 6.4 | 3000 | 496.2837 | -0.1460 | -0.1234 | -0.1390 | -0.1315 |
63
- | 341.2555 | 7.47 | 3500 | 488.9790 | -0.1150 | -0.0937 | -0.1072 | -0.1026 |
64
- | 544.4691 | 8.54 | 4000 | 484.3144 | -0.0955 | -0.0745 | -0.0870 | -0.0841 |
65
- | 377.6802 | 9.61 | 4500 | 483.1183 | -0.0911 | -0.0702 | -0.0823 | -0.0799 |
 
 
 
 
 
66
 
67
 
68
  ### Framework versions
 
1
  ---
2
  license: mit
3
+ base_model: arslanarjumand/wav2vec-reptiles
4
  tags:
5
  - generated_from_trainer
6
  model-index:
 
13
 
14
  # wav2vec-reptiles
15
 
16
+ This model is a fine-tuned version of [arslanarjumand/wav2vec-reptiles](https://huggingface.co/arslanarjumand/wav2vec-reptiles) on the None dataset.
17
  It achieves the following results on the evaluation set:
18
+ - Loss: 390.7498
19
+ - Pcc Accuracy: 0.3249
20
+ - Pcc Fluency: 0.3309
21
+ - Pcc Total Score: 0.3483
22
+ - Pcc Content: 0.3289
23
 
24
  ## Model description
25
 
 
47
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
48
  - lr_scheduler_type: cosine
49
  - lr_scheduler_warmup_ratio: 0.4
50
+ - num_epochs: 15
51
  - mixed_precision_training: Native AMP
52
 
53
  ### Training results
54
 
55
  | Training Loss | Epoch | Step | Validation Loss | Pcc Accuracy | Pcc Fluency | Pcc Total Score | Pcc Content |
56
  |:-------------:|:-----:|:----:|:---------------:|:------------:|:-----------:|:---------------:|:-----------:|
57
+ | 295.2316 | 1.07 | 500 | 461.7437 | -0.1108 | -0.1167 | -0.1176 | -0.1085 |
58
+ | 597.6253 | 2.13 | 1000 | 457.3701 | -0.0901 | -0.0972 | -0.0963 | -0.0889 |
59
+ | 398.6829 | 3.2 | 1500 | 451.0632 | -0.0516 | -0.0600 | -0.0562 | -0.0518 |
60
+ | 268.3367 | 4.27 | 2000 | 442.0888 | -0.0001 | -0.0087 | -0.0019 | -0.0014 |
61
+ | 672.6757 | 5.34 | 2500 | 430.7292 | 0.0784 | 0.0684 | 0.0804 | 0.0751 |
62
+ | 356.5551 | 6.4 | 3000 | 420.3665 | 0.1491 | 0.1413 | 0.1562 | 0.1459 |
63
+ | 439.0877 | 7.47 | 3500 | 418.4001 | 0.1997 | 0.1944 | 0.2109 | 0.1975 |
64
+ | 486.5083 | 8.54 | 4000 | 403.9485 | 0.2559 | 0.2552 | 0.2724 | 0.2560 |
65
+ | 282.6038 | 9.61 | 4500 | 398.6847 | 0.2874 | 0.2885 | 0.3064 | 0.2882 |
66
+ | 397.4536 | 10.67 | 5000 | 394.2357 | 0.3064 | 0.3098 | 0.3274 | 0.3088 |
67
+ | 347.7374 | 11.74 | 5500 | 392.6132 | 0.3168 | 0.3217 | 0.3392 | 0.3201 |
68
+ | 343.9868 | 12.81 | 6000 | 391.1635 | 0.3226 | 0.3286 | 0.3460 | 0.3265 |
69
+ | 704.0403 | 13.87 | 6500 | 390.9952 | 0.3244 | 0.3304 | 0.3477 | 0.3284 |
70
+ | 333.3059 | 14.94 | 7000 | 390.7498 | 0.3249 | 0.3309 | 0.3483 | 0.3289 |
71
 
72
 
73
  ### Framework versions
config.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "_name_or_path": "facebook/w2v-bert-2.0",
3
  "activation_dropout": 0.006,
4
  "adapter_act": "relu",
5
  "adapter_kernel_size": 3,
 
1
  {
2
+ "_name_or_path": "arslanarjumand/wav2vec-reptiles",
3
  "activation_dropout": 0.006,
4
  "adapter_act": "relu",
5
  "adapter_kernel_size": 3,
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3fddaf81529524678e9557f62263777143f3e524c988e8d1ac11a21b67626ea1
3
  size 2325236000
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7ace668030e70cf126b06672bad9d71a32f570ae1554b10045acb9db038754a6
3
  size 2325236000
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2537ca84b04b67a116773480a8d78d6e854b65ac1d2475dca6ed6ed742bed2f0
3
  size 4728
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:119e926d6f273268a6dc07f037caa94cf98bb5346794c83fc7bdbc6d45e95cbf
3
  size 4728