Update README.md
Browse files
README.md
CHANGED
@@ -8,7 +8,11 @@ library_name: peft
|
|
8 |
license: apache-2.0
|
9 |
metrics:
|
10 |
- wer
|
|
|
|
|
|
|
11 |
- bleu
|
|
|
12 |
tags:
|
13 |
- generated_from_trainer
|
14 |
model-index:
|
@@ -25,11 +29,24 @@ model-index:
|
|
25 |
args: 'config: th, split: validation'
|
26 |
metrics:
|
27 |
- type: wer
|
28 |
-
value: 0.
|
29 |
name: Wer
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
30 |
- type: bleu
|
31 |
-
value: 8.
|
32 |
name: Bleu
|
|
|
|
|
|
|
|
|
33 |
---
|
34 |
|
35 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
@@ -41,25 +58,11 @@ This model is a fine-tuned version of [openai/whisper-large-v3](https://huggingf
|
|
41 |
It achieves the following results on the evaluation set:
|
42 |
- Loss: 0.1894
|
43 |
- Wer: 0.8101
|
44 |
-
- Ter:
|
45 |
-
- Chrf:
|
46 |
- Cer: 0.1041
|
47 |
- Bleu: 8.7391
|
48 |
-
-
|
49 |
-
|
50 |
-
## Model description
|
51 |
-
|
52 |
-
More information needed
|
53 |
-
|
54 |
-
## Intended uses & limitations
|
55 |
-
|
56 |
-
More information needed
|
57 |
-
|
58 |
-
## Training and evaluation data
|
59 |
-
|
60 |
-
More information needed
|
61 |
-
|
62 |
-
## Training procedure
|
63 |
|
64 |
### Training hyperparameters
|
65 |
|
@@ -78,11 +81,11 @@ The following hyperparameters were used during training:
|
|
78 |
|
79 |
### Training results
|
80 |
|
81 |
-
| Training Loss | Epoch | Step | Validation Loss | Wer | Ter
|
82 |
-
|
83 |
-
| 0.3053 | 0.32 | 50 | 0.2052 | 0.8457 |
|
84 |
-
| 0.3752 | 0.64 | 100 | 0.1937 | 0.8323 |
|
85 |
-
| 0.2794 | 0.96 | 150 | 0.1894 | 0.8101 |
|
86 |
|
87 |
|
88 |
### Framework versions
|
|
|
8 |
license: apache-2.0
|
9 |
metrics:
|
10 |
- wer
|
11 |
+
- ter
|
12 |
+
- chrf
|
13 |
+
- cer
|
14 |
- bleu
|
15 |
+
- suber
|
16 |
tags:
|
17 |
- generated_from_trainer
|
18 |
model-index:
|
|
|
29 |
args: 'config: th, split: validation'
|
30 |
metrics:
|
31 |
- type: wer
|
32 |
+
value: 0.8101
|
33 |
name: Wer
|
34 |
+
- name: Ter
|
35 |
+
type: ter
|
36 |
+
value: 81.0089
|
37 |
+
- name: ChrF
|
38 |
+
type: chrf
|
39 |
+
value: 87.4811
|
40 |
+
- name: CER
|
41 |
+
type: cer
|
42 |
+
value: 0.1041
|
43 |
- type: bleu
|
44 |
+
value: 8.7391
|
45 |
name: Bleu
|
46 |
+
- name: SubER
|
47 |
+
type: suber
|
48 |
+
value: 0.8189
|
49 |
+
pipeline_tag: automatic-speech-recognition
|
50 |
---
|
51 |
|
52 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
58 |
It achieves the following results on the evaluation set:
|
59 |
- Loss: 0.1894
|
60 |
- Wer: 0.8101
|
61 |
+
- Ter: 81.0089
|
62 |
+
- Chrf: 87.4811
|
63 |
- Cer: 0.1041
|
64 |
- Bleu: 8.7391
|
65 |
+
- SubER: 0.8189
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
66 |
|
67 |
### Training hyperparameters
|
68 |
|
|
|
81 |
|
82 |
### Training results
|
83 |
|
84 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer | Ter | Chrf | Cer | Bleu | SubER |
|
85 |
+
|:-------------:|:-----:|:----:|:---------------:|:------:|:-------:|:-------:|:------:|:------:|:---------:|
|
86 |
+
| 0.3053 | 0.32 | 50 | 0.2052 | 0.8457 | 84.5697 | 86.6971 | 0.1115 | 7.8703 | 0.8640 |
|
87 |
+
| 0.3752 | 0.64 | 100 | 0.1937 | 0.8323 | 83.2344 | 86.9801 | 0.1087 | 8.1510 | 0.8469 |
|
88 |
+
| 0.2794 | 0.96 | 150 | 0.1894 | 0.8101 | 81.0089 | 87.4811 | 0.1041 | 8.7391 | 0.8189 |
|
89 |
|
90 |
|
91 |
### Framework versions
|