End of training
Browse files- README.md +72 -14
- model.safetensors +1 -1
README.md
CHANGED
@@ -15,7 +15,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
15 |
|
16 |
This model was trained from scratch on the common_voice_17_0 dataset.
|
17 |
It achieves the following results on the evaluation set:
|
18 |
-
- Loss: 0.
|
19 |
|
20 |
## Model description
|
21 |
|
@@ -34,28 +34,86 @@ More information needed
|
|
34 |
### Training hyperparameters
|
35 |
|
36 |
The following hyperparameters were used during training:
|
37 |
-
- learning_rate: 0.
|
38 |
-
- train_batch_size:
|
39 |
- eval_batch_size: 8
|
40 |
- seed: 42
|
41 |
-
- gradient_accumulation_steps:
|
42 |
-
- total_train_batch_size:
|
43 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
44 |
- lr_scheduler_type: linear
|
45 |
- lr_scheduler_warmup_steps: 500
|
46 |
-
- num_epochs:
|
47 |
- mixed_precision_training: Native AMP
|
48 |
|
49 |
### Training results
|
50 |
|
51 |
-
| Training Loss | Epoch
|
52 |
-
|
53 |
-
|
|
54 |
-
|
|
55 |
-
|
|
56 |
-
|
|
57 |
-
|
|
58 |
-
| 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
59 |
|
60 |
|
61 |
### Framework versions
|
|
|
15 |
|
16 |
This model was trained from scratch on the common_voice_17_0 dataset.
|
17 |
It achieves the following results on the evaluation set:
|
18 |
+
- Loss: 0.7309
|
19 |
|
20 |
## Model description
|
21 |
|
|
|
34 |
### Training hyperparameters
|
35 |
|
36 |
The following hyperparameters were used during training:
|
37 |
+
- learning_rate: 0.0001
|
38 |
+
- train_batch_size: 6
|
39 |
- eval_batch_size: 8
|
40 |
- seed: 42
|
41 |
+
- gradient_accumulation_steps: 4
|
42 |
+
- total_train_batch_size: 24
|
43 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
44 |
- lr_scheduler_type: linear
|
45 |
- lr_scheduler_warmup_steps: 500
|
46 |
+
- num_epochs: 240
|
47 |
- mixed_precision_training: Native AMP
|
48 |
|
49 |
### Training results
|
50 |
|
51 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
52 |
+
|:-------------:|:--------:|:-----:|:---------------:|
|
53 |
+
| 0.1677 | 3.6866 | 200 | 1.0381 |
|
54 |
+
| 0.1214 | 7.3733 | 400 | 0.5607 |
|
55 |
+
| 0.1272 | 11.0599 | 600 | 0.5442 |
|
56 |
+
| 0.135 | 14.7465 | 800 | 0.5933 |
|
57 |
+
| 0.0824 | 18.4332 | 1000 | 0.6316 |
|
58 |
+
| 0.0711 | 22.1198 | 1200 | 0.5971 |
|
59 |
+
| 0.0653 | 25.8065 | 1400 | 0.6050 |
|
60 |
+
| 0.0499 | 29.4931 | 1600 | 0.6699 |
|
61 |
+
| 0.0516 | 33.1797 | 1800 | 0.6940 |
|
62 |
+
| 0.0507 | 36.8664 | 2000 | 0.7045 |
|
63 |
+
| 0.0478 | 40.5530 | 2200 | 0.7603 |
|
64 |
+
| 0.045 | 44.2396 | 2400 | 0.7415 |
|
65 |
+
| 0.0419 | 47.9263 | 2600 | 0.7341 |
|
66 |
+
| 0.0344 | 51.6129 | 2800 | 0.7328 |
|
67 |
+
| 0.0354 | 55.2995 | 3000 | 0.8550 |
|
68 |
+
| 0.0268 | 58.9862 | 3200 | 0.7838 |
|
69 |
+
| 0.0383 | 62.6728 | 3400 | 0.7995 |
|
70 |
+
| 0.0371 | 66.3594 | 3600 | 0.7765 |
|
71 |
+
| 0.0264 | 70.0461 | 3800 | 0.8186 |
|
72 |
+
| 0.0212 | 73.7327 | 4000 | 0.7439 |
|
73 |
+
| 0.0177 | 77.4194 | 4200 | 0.7830 |
|
74 |
+
| 0.0204 | 81.1060 | 4400 | 0.8145 |
|
75 |
+
| 0.0254 | 84.7926 | 4600 | 0.8149 |
|
76 |
+
| 0.0257 | 88.4793 | 4800 | 0.7663 |
|
77 |
+
| 0.0126 | 92.1659 | 5000 | 0.7704 |
|
78 |
+
| 0.0196 | 95.8525 | 5200 | 0.7660 |
|
79 |
+
| 0.0185 | 99.5392 | 5400 | 0.8580 |
|
80 |
+
| 0.0236 | 103.2258 | 5600 | 0.8169 |
|
81 |
+
| 0.0141 | 106.9124 | 5800 | 0.8222 |
|
82 |
+
| 0.0142 | 110.5991 | 6000 | 0.9001 |
|
83 |
+
| 0.0098 | 114.2857 | 6200 | 0.8509 |
|
84 |
+
| 0.0372 | 117.9724 | 6400 | 0.7734 |
|
85 |
+
| 0.0075 | 121.6590 | 6600 | 0.8911 |
|
86 |
+
| 0.0118 | 125.3456 | 6800 | 0.8347 |
|
87 |
+
| 0.0115 | 129.0323 | 7000 | 0.8926 |
|
88 |
+
| 0.0164 | 132.7189 | 7200 | 0.7985 |
|
89 |
+
| 0.006 | 136.4055 | 7400 | 0.7571 |
|
90 |
+
| 0.0124 | 140.0922 | 7600 | 0.8476 |
|
91 |
+
| 0.0141 | 143.7788 | 7800 | 0.8071 |
|
92 |
+
| 0.0065 | 147.4654 | 8000 | 0.7630 |
|
93 |
+
| 0.0095 | 151.1521 | 8200 | 0.7161 |
|
94 |
+
| 0.0063 | 154.8387 | 8400 | 0.8165 |
|
95 |
+
| 0.0107 | 158.5253 | 8600 | 0.7411 |
|
96 |
+
| 0.0037 | 162.2120 | 8800 | 0.7424 |
|
97 |
+
| 0.0045 | 165.8986 | 9000 | 0.7611 |
|
98 |
+
| 0.0044 | 169.5853 | 9200 | 0.7278 |
|
99 |
+
| 0.0043 | 173.2719 | 9400 | 0.7396 |
|
100 |
+
| 0.0025 | 176.9585 | 9600 | 0.7215 |
|
101 |
+
| 0.0029 | 180.6452 | 9800 | 0.7551 |
|
102 |
+
| 0.0067 | 184.3318 | 10000 | 0.7518 |
|
103 |
+
| 0.0062 | 188.0184 | 10200 | 0.7668 |
|
104 |
+
| 0.0065 | 191.7051 | 10400 | 0.7433 |
|
105 |
+
| 0.0024 | 195.3917 | 10600 | 0.7942 |
|
106 |
+
| 0.0039 | 199.0783 | 10800 | 0.7448 |
|
107 |
+
| 0.0024 | 202.7650 | 11000 | 0.7290 |
|
108 |
+
| 0.0036 | 206.4516 | 11200 | 0.7678 |
|
109 |
+
| 0.0001 | 210.1382 | 11400 | 0.7390 |
|
110 |
+
| 0.0009 | 213.8249 | 11600 | 0.7292 |
|
111 |
+
| 0.0008 | 217.5115 | 11800 | 0.7383 |
|
112 |
+
| 0.0009 | 221.1982 | 12000 | 0.7435 |
|
113 |
+
| 0.0009 | 224.8848 | 12200 | 0.7324 |
|
114 |
+
| 0.0007 | 228.5714 | 12400 | 0.7444 |
|
115 |
+
| 0.0002 | 232.2581 | 12600 | 0.7228 |
|
116 |
+
| 0.0005 | 235.9447 | 12800 | 0.7309 |
|
117 |
|
118 |
|
119 |
### Framework versions
|
model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 3850792988
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:391683654c3aa59a2b680898a0e8f2f03700826c0240dec688840c2c1e714722
|
3 |
size 3850792988
|