End of training
Browse files- README.md +32 -12
- generation_config.json +1 -1
README.md
CHANGED
@@ -1,5 +1,4 @@
|
|
1 |
---
|
2 |
-
library_name: transformers
|
3 |
language:
|
4 |
- or
|
5 |
license: apache-2.0
|
@@ -8,9 +7,24 @@ tags:
|
|
8 |
- generated_from_trainer
|
9 |
datasets:
|
10 |
- mozilla-foundation/common_voice_17_0
|
|
|
|
|
11 |
model-index:
|
12 |
- name: Whisper Small Or- Mudit Sharma
|
13 |
-
results:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
14 |
---
|
15 |
|
16 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
@@ -20,13 +34,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
20 |
|
21 |
This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the Common Voice 17.0 dataset.
|
22 |
It achieves the following results on the evaluation set:
|
23 |
-
-
|
24 |
-
-
|
25 |
-
- eval_runtime: 631.1053
|
26 |
-
- eval_samples_per_second: 1.103
|
27 |
-
- eval_steps_per_second: 0.138
|
28 |
-
- epoch: 7.8125
|
29 |
-
- step: 1000
|
30 |
|
31 |
## Model description
|
32 |
|
@@ -55,9 +64,20 @@ The following hyperparameters were used during training:
|
|
55 |
- training_steps: 5000
|
56 |
- mixed_precision_training: Native AMP
|
57 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
58 |
### Framework versions
|
59 |
|
60 |
-
- Transformers 4.44.
|
61 |
-
- Pytorch 2.4.
|
62 |
-
- Datasets
|
63 |
- Tokenizers 0.19.1
|
|
|
1 |
---
|
|
|
2 |
language:
|
3 |
- or
|
4 |
license: apache-2.0
|
|
|
7 |
- generated_from_trainer
|
8 |
datasets:
|
9 |
- mozilla-foundation/common_voice_17_0
|
10 |
+
metrics:
|
11 |
+
- wer
|
12 |
model-index:
|
13 |
- name: Whisper Small Or- Mudit Sharma
|
14 |
+
results:
|
15 |
+
- task:
|
16 |
+
name: Automatic Speech Recognition
|
17 |
+
type: automatic-speech-recognition
|
18 |
+
dataset:
|
19 |
+
name: Common Voice 17.0
|
20 |
+
type: mozilla-foundation/common_voice_17_0
|
21 |
+
config: or
|
22 |
+
split: None
|
23 |
+
args: 'config: or, split: test'
|
24 |
+
metrics:
|
25 |
+
- name: Wer
|
26 |
+
type: wer
|
27 |
+
value: 52.426532325776655
|
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-small](https://huggingface.co/openai/whisper-small) on the Common Voice 17.0 dataset.
|
36 |
It achieves the following results on the evaluation set:
|
37 |
+
- Loss: 0.3656
|
38 |
+
- Wer: 52.4265
|
|
|
|
|
|
|
|
|
|
|
39 |
|
40 |
## Model description
|
41 |
|
|
|
64 |
- training_steps: 5000
|
65 |
- mixed_precision_training: Native AMP
|
66 |
|
67 |
+
### Training results
|
68 |
+
|
69 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
70 |
+
|:-------------:|:-------:|:----:|:---------------:|:-------:|
|
71 |
+
| 0.0192 | 7.8125 | 1000 | 0.2476 | 56.1209 |
|
72 |
+
| 0.0011 | 15.625 | 2000 | 0.2956 | 52.8128 |
|
73 |
+
| 0.0001 | 23.4375 | 3000 | 0.3320 | 52.6616 |
|
74 |
+
| 0.0 | 31.25 | 4000 | 0.3582 | 52.6280 |
|
75 |
+
| 0.0 | 39.0625 | 5000 | 0.3656 | 52.4265 |
|
76 |
+
|
77 |
+
|
78 |
### Framework versions
|
79 |
|
80 |
+
- Transformers 4.44.0
|
81 |
+
- Pytorch 2.4.0
|
82 |
+
- Datasets 2.21.0
|
83 |
- Tokenizers 0.19.1
|
generation_config.json
CHANGED
@@ -252,5 +252,5 @@
|
|
252 |
"transcribe": 50359,
|
253 |
"translate": 50358
|
254 |
},
|
255 |
-
"transformers_version": "4.44.
|
256 |
}
|
|
|
252 |
"transcribe": 50359,
|
253 |
"translate": 50358
|
254 |
},
|
255 |
+
"transformers_version": "4.44.0"
|
256 |
}
|