update model card README.md
Browse files
README.md
CHANGED
@@ -1,6 +1,6 @@
|
|
1 |
---
|
2 |
language:
|
3 |
-
-
|
4 |
license: apache-2.0
|
5 |
tags:
|
6 |
- whisper-event
|
@@ -24,7 +24,7 @@ model-index:
|
|
24 |
metrics:
|
25 |
- name: Wer
|
26 |
type: wer
|
27 |
-
value:
|
28 |
---
|
29 |
|
30 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
@@ -34,9 +34,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
34 |
|
35 |
This model is a fine-tuned version of [openai/whisper-base](https://huggingface.co/openai/whisper-base) on the google/fleurs dataset.
|
36 |
It achieves the following results on the evaluation set:
|
37 |
-
- Loss:
|
38 |
-
- Wer:
|
39 |
-
- Cer:
|
40 |
|
41 |
## Model description
|
42 |
|
@@ -55,7 +55,7 @@ More information needed
|
|
55 |
### Training hyperparameters
|
56 |
|
57 |
The following hyperparameters were used during training:
|
58 |
-
- learning_rate: 1e-
|
59 |
- train_batch_size: 16
|
60 |
- eval_batch_size: 8
|
61 |
- seed: 42
|
@@ -64,19 +64,23 @@ The following hyperparameters were used during training:
|
|
64 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
65 |
- lr_scheduler_type: linear
|
66 |
- lr_scheduler_warmup_steps: 30
|
67 |
-
- training_steps:
|
68 |
- mixed_precision_training: Native AMP
|
69 |
|
70 |
### Training results
|
71 |
|
72 |
-
| Training Loss | Epoch | Step | Validation Loss | Wer
|
73 |
-
|
74 |
-
|
|
75 |
-
|
|
76 |
-
|
|
77 |
-
|
|
78 |
-
|
|
79 |
-
|
|
|
|
|
|
|
|
|
|
80 |
|
81 |
|
82 |
### Framework versions
|
|
|
1 |
---
|
2 |
language:
|
3 |
+
- ps
|
4 |
license: apache-2.0
|
5 |
tags:
|
6 |
- whisper-event
|
|
|
24 |
metrics:
|
25 |
- name: Wer
|
26 |
type: wer
|
27 |
+
value: 57.611985472154956
|
28 |
---
|
29 |
|
30 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
34 |
|
35 |
This model is a fine-tuned version of [openai/whisper-base](https://huggingface.co/openai/whisper-base) on the google/fleurs dataset.
|
36 |
It achieves the following results on the evaluation set:
|
37 |
+
- Loss: 0.8723
|
38 |
+
- Wer: 57.6120
|
39 |
+
- Cer: 26.6468
|
40 |
|
41 |
## Model description
|
42 |
|
|
|
55 |
### Training hyperparameters
|
56 |
|
57 |
The following hyperparameters were used during training:
|
58 |
+
- learning_rate: 1e-05
|
59 |
- train_batch_size: 16
|
60 |
- eval_batch_size: 8
|
61 |
- seed: 42
|
|
|
64 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
65 |
- lr_scheduler_type: linear
|
66 |
- lr_scheduler_warmup_steps: 30
|
67 |
+
- training_steps: 1000
|
68 |
- mixed_precision_training: Native AMP
|
69 |
|
70 |
### Training results
|
71 |
|
72 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
|
73 |
+
|:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|
|
74 |
+
| 0.9708 | 2.38 | 100 | 0.8821 | 64.0133 | 27.3253 |
|
75 |
+
| 0.7477 | 4.75 | 200 | 0.8062 | 59.9576 | 26.4079 |
|
76 |
+
| 0.6229 | 7.14 | 300 | 0.7855 | 58.3081 | 26.3193 |
|
77 |
+
| 0.4833 | 9.52 | 400 | 0.7870 | 57.5288 | 24.8855 |
|
78 |
+
| 0.4084 | 11.89 | 500 | 0.7980 | 56.5224 | 25.2214 |
|
79 |
+
| 0.3323 | 14.28 | 600 | 0.8201 | 56.6662 | 25.3317 |
|
80 |
+
| 0.283 | 16.66 | 700 | 0.8406 | 57.7406 | 26.8674 |
|
81 |
+
| 0.2598 | 19.05 | 800 | 0.8538 | 57.2866 | 26.0386 |
|
82 |
+
| 0.2235 | 21.42 | 900 | 0.8697 | 58.2703 | 26.6819 |
|
83 |
+
| 0.2202 | 23.8 | 1000 | 0.8723 | 57.6120 | 26.6468 |
|
84 |
|
85 |
|
86 |
### Framework versions
|