Update README.md
Browse files
README.md
CHANGED
@@ -50,33 +50,7 @@ More information needed
|
|
50 |
|
51 |
## Training procedure
|
52 |
|
53 |
-
### Training hyperparameters
|
54 |
|
55 |
-
The following hyperparameters were used during training:
|
56 |
-
- learning_rate: 6.600000000000001e-06
|
57 |
-
- train_batch_size: 16
|
58 |
-
- eval_batch_size: 8
|
59 |
-
- seed: 42
|
60 |
-
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
61 |
-
- lr_scheduler_type: linear
|
62 |
-
- lr_scheduler_warmup_steps: 1
|
63 |
-
- training_steps: 1000
|
64 |
-
- mixed_precision_training: Native AMP
|
65 |
-
|
66 |
-
### Training results
|
67 |
-
|
68 |
-
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
69 |
-
|:-------------:|:-----:|:----:|:---------------:|:-------:|
|
70 |
-
| 0.4886 | 0.1 | 100 | 0.3900 | 12.3225 |
|
71 |
-
| 0.2534 | 0.2 | 200 | 0.2826 | 15.7514 |
|
72 |
-
| 0.1143 | 1.05 | 300 | 0.2457 | 16.4211 |
|
73 |
-
| 0.1134 | 1.15 | 400 | 0.2496 | 14.2245 |
|
74 |
-
| 0.1186 | 1.25 | 500 | 0.2511 | 13.5280 |
|
75 |
-
| 0.054 | 2.1 | 600 | 0.2694 | 15.6175 |
|
76 |
-
| 0.0449 | 2.2 | 700 | 0.2743 | 21.0019 |
|
77 |
-
| 0.0224 | 3.05 | 800 | 0.2812 | 24.1897 |
|
78 |
-
| 0.0241 | 3.15 | 900 | 0.2828 | 31.6368 |
|
79 |
-
| 0.0324 | 3.25 | 1000 | 0.2841 | 31.3153 |
|
80 |
|
81 |
|
82 |
### Framework versions
|
|
|
50 |
|
51 |
## Training procedure
|
52 |
|
|
|
53 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
54 |
|
55 |
|
56 |
### Framework versions
|