End of training
Browse files
README.md
CHANGED
@@ -17,8 +17,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
17 |
|
18 |
This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on the None dataset.
|
19 |
It achieves the following results on the evaluation set:
|
20 |
-
- Loss:
|
21 |
-
- Wer:
|
22 |
|
23 |
## Model description
|
24 |
|
@@ -37,7 +37,7 @@ More information needed
|
|
37 |
### Training hyperparameters
|
38 |
|
39 |
The following hyperparameters were used during training:
|
40 |
-
- learning_rate:
|
41 |
- train_batch_size: 8
|
42 |
- eval_batch_size: 8
|
43 |
- seed: 42
|
@@ -45,65 +45,30 @@ The following hyperparameters were used during training:
|
|
45 |
- total_train_batch_size: 16
|
46 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
47 |
- lr_scheduler_type: linear
|
48 |
-
- num_epochs:
|
49 |
- mixed_precision_training: Native AMP
|
50 |
|
51 |
### Training results
|
52 |
|
53 |
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
54 |
|:-------------:|:-----:|:----:|:---------------:|:------:|
|
55 |
-
|
|
56 |
-
|
|
57 |
-
|
|
58 |
-
|
|
59 |
-
|
|
60 |
-
|
|
61 |
-
|
|
62 |
-
|
|
63 |
-
|
|
64 |
-
|
|
65 |
-
|
|
66 |
-
|
|
67 |
-
|
|
68 |
-
|
|
69 |
-
|
|
70 |
-
|
|
71 |
-
|
|
72 |
-
| 5.9649 | 10.29 | 1800 | 13.0151 | 1.0 |
|
73 |
-
| 5.7463 | 10.86 | 1900 | 12.8257 | 1.0 |
|
74 |
-
| 5.5924 | 11.43 | 2000 | 12.6365 | 1.0 |
|
75 |
-
| 5.427 | 12.0 | 2100 | 12.4524 | 1.0 |
|
76 |
-
| 5.3501 | 12.57 | 2200 | 12.2706 | 1.0 |
|
77 |
-
| 5.1945 | 13.14 | 2300 | 12.0939 | 1.0 |
|
78 |
-
| 5.0109 | 13.71 | 2400 | 11.9239 | 1.0 |
|
79 |
-
| 4.9798 | 14.29 | 2500 | 11.7596 | 1.0 |
|
80 |
-
| 4.8013 | 14.86 | 2600 | 11.6010 | 1.0 |
|
81 |
-
| 4.7351 | 15.43 | 2700 | 11.4466 | 1.0 |
|
82 |
-
| 4.5813 | 16.0 | 2800 | 11.3010 | 1.0 |
|
83 |
-
| 4.54 | 16.57 | 2900 | 11.1603 | 1.0 |
|
84 |
-
| 4.4236 | 17.14 | 3000 | 11.0255 | 1.0 |
|
85 |
-
| 4.3137 | 17.71 | 3100 | 10.8994 | 1.0 |
|
86 |
-
| 4.2944 | 18.29 | 3200 | 10.7777 | 1.0 |
|
87 |
-
| 4.179 | 18.86 | 3300 | 10.6654 | 1.0 |
|
88 |
-
| 4.1668 | 19.43 | 3400 | 10.5568 | 1.0 |
|
89 |
-
| 4.0301 | 20.0 | 3500 | 10.4562 | 1.0 |
|
90 |
-
| 4.0315 | 20.57 | 3600 | 10.3608 | 1.0 |
|
91 |
-
| 3.9657 | 21.14 | 3700 | 10.2723 | 1.0 |
|
92 |
-
| 3.8779 | 21.71 | 3800 | 10.1908 | 1.0 |
|
93 |
-
| 3.8818 | 22.29 | 3900 | 10.1142 | 1.0 |
|
94 |
-
| 3.8197 | 22.86 | 4000 | 10.0451 | 1.0 |
|
95 |
-
| 3.8041 | 23.43 | 4100 | 9.9809 | 1.0 |
|
96 |
-
| 3.7395 | 24.0 | 4200 | 9.9229 | 1.0 |
|
97 |
-
| 3.7321 | 24.57 | 4300 | 9.8707 | 1.0 |
|
98 |
-
| 3.7191 | 25.14 | 4400 | 9.8233 | 1.0 |
|
99 |
-
| 3.6529 | 25.71 | 4500 | 9.7823 | 1.0 |
|
100 |
-
| 3.6868 | 26.29 | 4600 | 9.7467 | 1.0 |
|
101 |
-
| 3.6413 | 26.86 | 4700 | 9.7163 | 1.0 |
|
102 |
-
| 3.6497 | 27.43 | 4800 | 9.6903 | 1.0 |
|
103 |
-
| 3.608 | 28.0 | 4900 | 9.6701 | 1.0 |
|
104 |
-
| 3.6195 | 28.57 | 5000 | 9.6548 | 1.0 |
|
105 |
-
| 3.6115 | 29.14 | 5100 | 9.6443 | 1.0 |
|
106 |
-
| 3.5798 | 29.71 | 5200 | 9.6392 | 1.0 |
|
107 |
|
108 |
|
109 |
### Framework versions
|
|
|
17 |
|
18 |
This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on the None dataset.
|
19 |
It achieves the following results on the evaluation set:
|
20 |
+
- Loss: 4.1078
|
21 |
+
- Wer: 0.9945
|
22 |
|
23 |
## Model description
|
24 |
|
|
|
37 |
### Training hyperparameters
|
38 |
|
39 |
The following hyperparameters were used during training:
|
40 |
+
- learning_rate: 0.005
|
41 |
- train_batch_size: 8
|
42 |
- eval_batch_size: 8
|
43 |
- seed: 42
|
|
|
45 |
- total_train_batch_size: 16
|
46 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
47 |
- lr_scheduler_type: linear
|
48 |
+
- num_epochs: 10
|
49 |
- mixed_precision_training: Native AMP
|
50 |
|
51 |
### Training results
|
52 |
|
53 |
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
54 |
|:-------------:|:-----:|:----:|:---------------:|:------:|
|
55 |
+
| 3.1667 | 0.57 | 100 | 4.5022 | 0.9941 |
|
56 |
+
| 2.8372 | 1.14 | 200 | 3.6561 | 0.9989 |
|
57 |
+
| 2.8065 | 1.71 | 300 | 3.9712 | 0.9943 |
|
58 |
+
| 2.7851 | 2.29 | 400 | 3.9929 | 0.9941 |
|
59 |
+
| 2.7765 | 2.86 | 500 | 4.2525 | 0.9942 |
|
60 |
+
| 2.7666 | 3.43 | 600 | 4.1306 | 0.9941 |
|
61 |
+
| 2.7647 | 4.0 | 700 | 4.1459 | 0.9941 |
|
62 |
+
| 2.7572 | 4.57 | 800 | 4.1862 | 0.9944 |
|
63 |
+
| 2.7506 | 5.14 | 900 | 4.1974 | 0.9948 |
|
64 |
+
| 2.7481 | 5.71 | 1000 | 4.3590 | 0.9944 |
|
65 |
+
| 2.7445 | 6.29 | 1100 | 4.1933 | 0.9942 |
|
66 |
+
| 2.7475 | 6.86 | 1200 | 4.2083 | 0.9948 |
|
67 |
+
| 2.7431 | 7.43 | 1300 | 4.1181 | 0.9945 |
|
68 |
+
| 2.7404 | 8.0 | 1400 | 4.1052 | 0.9944 |
|
69 |
+
| 2.7387 | 8.57 | 1500 | 4.1345 | 0.9945 |
|
70 |
+
| 2.7365 | 9.14 | 1600 | 4.0646 | 0.9945 |
|
71 |
+
| 2.7353 | 9.71 | 1700 | 4.1078 | 0.9945 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
72 |
|
73 |
|
74 |
### Framework versions
|
model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 377611120
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b3dcb5f8ca8f0c5a502e468c0e16425ae653385ca2557df46b217274157a4ed2
|
3 |
size 377611120
|
runs/May25_08-09-08_instance-til-team-31/events.out.tfevents.1716624688.instance-til-team-31
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d7bfce976a689867794af471f1ce04467017c1c84b9d1806d8328c6b80196d18
|
3 |
+
size 14404
|