End of training
Browse files- README.md +19 -8
- generation_config.json +0 -1
README.md
CHANGED
@@ -4,6 +4,8 @@ license: apache-2.0
|
|
4 |
base_model: openai/whisper-tiny
|
5 |
tags:
|
6 |
- generated_from_trainer
|
|
|
|
|
7 |
model-index:
|
8 |
- name: whisper-tiny-akan
|
9 |
results: []
|
@@ -16,13 +18,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
16 |
|
17 |
This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the None dataset.
|
18 |
It achieves the following results on the evaluation set:
|
19 |
-
-
|
20 |
-
-
|
21 |
-
- eval_runtime: 101.6449
|
22 |
-
- eval_samples_per_second: 1.968
|
23 |
-
- eval_steps_per_second: 0.246
|
24 |
-
- epoch: 20.0
|
25 |
-
- step: 500
|
26 |
|
27 |
## Model description
|
28 |
|
@@ -42,7 +39,7 @@ More information needed
|
|
42 |
|
43 |
The following hyperparameters were used during training:
|
44 |
- learning_rate: 0.0001
|
45 |
-
- train_batch_size:
|
46 |
- eval_batch_size: 8
|
47 |
- seed: 42
|
48 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
@@ -51,6 +48,20 @@ The following hyperparameters were used during training:
|
|
51 |
- training_steps: 2000
|
52 |
- mixed_precision_training: Native AMP
|
53 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
54 |
### Framework versions
|
55 |
|
56 |
- Transformers 4.44.2
|
|
|
4 |
base_model: openai/whisper-tiny
|
5 |
tags:
|
6 |
- generated_from_trainer
|
7 |
+
metrics:
|
8 |
+
- wer
|
9 |
model-index:
|
10 |
- name: whisper-tiny-akan
|
11 |
results: []
|
|
|
18 |
|
19 |
This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the None dataset.
|
20 |
It achieves the following results on the evaluation set:
|
21 |
+
- Loss: 1.1400
|
22 |
+
- Wer: 49.1687
|
|
|
|
|
|
|
|
|
|
|
23 |
|
24 |
## Model description
|
25 |
|
|
|
39 |
|
40 |
The following hyperparameters were used during training:
|
41 |
- learning_rate: 0.0001
|
42 |
+
- train_batch_size: 16
|
43 |
- eval_batch_size: 8
|
44 |
- seed: 42
|
45 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
|
|
48 |
- training_steps: 2000
|
49 |
- mixed_precision_training: Native AMP
|
50 |
|
51 |
+
### Training results
|
52 |
+
|
53 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
54 |
+
|:-------------:|:-----:|:----:|:---------------:|:-------:|
|
55 |
+
| 0.3923 | 5.0 | 250 | 0.8148 | 61.9777 |
|
56 |
+
| 0.1045 | 10.0 | 500 | 0.9445 | 53.9829 |
|
57 |
+
| 0.0414 | 15.0 | 750 | 1.0241 | 49.9639 |
|
58 |
+
| 0.0187 | 20.0 | 1000 | 1.0862 | 51.7999 |
|
59 |
+
| 0.0089 | 25.0 | 1250 | 1.1386 | 51.1927 |
|
60 |
+
| 0.0025 | 30.0 | 1500 | 1.1284 | 49.1832 |
|
61 |
+
| 0.0004 | 35.0 | 1750 | 1.1355 | 48.7061 |
|
62 |
+
| 0.0004 | 40.0 | 2000 | 1.1400 | 49.1687 |
|
63 |
+
|
64 |
+
|
65 |
### Framework versions
|
66 |
|
67 |
- Transformers 4.44.2
|
generation_config.json
CHANGED
@@ -31,7 +31,6 @@
|
|
31 |
],
|
32 |
"bos_token_id": 50257,
|
33 |
"decoder_start_token_id": 50258,
|
34 |
-
"dropout": 0.1,
|
35 |
"eos_token_id": 50257,
|
36 |
"is_multilingual": true,
|
37 |
"lang_to_id": {
|
|
|
31 |
],
|
32 |
"bos_token_id": 50257,
|
33 |
"decoder_start_token_id": 50258,
|
|
|
34 |
"eos_token_id": 50257,
|
35 |
"is_multilingual": true,
|
36 |
"lang_to_id": {
|