End of training
Browse files- README.md +32 -5
- generation_config.json +1 -1
README.md
CHANGED
@@ -7,9 +7,24 @@ tags:
|
|
7 |
- generated_from_trainer
|
8 |
datasets:
|
9 |
- arab
|
|
|
|
|
10 |
model-index:
|
11 |
- name: Whisper Small ar - Atishay Sharma
|
12 |
-
results:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
13 |
---
|
14 |
|
15 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
@@ -18,6 +33,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
18 |
# Whisper Small ar - Atishay Sharma
|
19 |
|
20 |
This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the arabic dataset.
|
|
|
|
|
|
|
21 |
|
22 |
## Model description
|
23 |
|
@@ -37,18 +55,27 @@ More information needed
|
|
37 |
|
38 |
The following hyperparameters were used during training:
|
39 |
- learning_rate: 1e-05
|
40 |
-
- train_batch_size:
|
41 |
- eval_batch_size: 4
|
42 |
- seed: 42
|
|
|
|
|
43 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
44 |
- lr_scheduler_type: linear
|
45 |
-
- lr_scheduler_warmup_steps:
|
46 |
- training_steps: 500
|
47 |
- mixed_precision_training: Native AMP
|
48 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
49 |
### Framework versions
|
50 |
|
51 |
-
- Transformers 4.
|
52 |
-
- Pytorch 2.1
|
53 |
- Datasets 2.18.0
|
54 |
- Tokenizers 0.15.2
|
|
|
7 |
- generated_from_trainer
|
8 |
datasets:
|
9 |
- arab
|
10 |
+
metrics:
|
11 |
+
- wer
|
12 |
model-index:
|
13 |
- name: Whisper Small ar - Atishay Sharma
|
14 |
+
results:
|
15 |
+
- task:
|
16 |
+
name: Automatic Speech Recognition
|
17 |
+
type: automatic-speech-recognition
|
18 |
+
dataset:
|
19 |
+
name: arabic
|
20 |
+
type: arab
|
21 |
+
config: default
|
22 |
+
split: train
|
23 |
+
args: 'config: ar, split: test'
|
24 |
+
metrics:
|
25 |
+
- name: Wer
|
26 |
+
type: wer
|
27 |
+
value: 4.545454545454546
|
28 |
---
|
29 |
|
30 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
33 |
# Whisper Small ar - Atishay Sharma
|
34 |
|
35 |
This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the arabic dataset.
|
36 |
+
It achieves the following results on the evaluation set:
|
37 |
+
- Loss: 0.0922
|
38 |
+
- Wer: 4.5455
|
39 |
|
40 |
## Model description
|
41 |
|
|
|
55 |
|
56 |
The following hyperparameters were used during training:
|
57 |
- learning_rate: 1e-05
|
58 |
+
- train_batch_size: 8
|
59 |
- eval_batch_size: 4
|
60 |
- seed: 42
|
61 |
+
- gradient_accumulation_steps: 2
|
62 |
+
- total_train_batch_size: 16
|
63 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
64 |
- lr_scheduler_type: linear
|
65 |
+
- lr_scheduler_warmup_steps: 10
|
66 |
- training_steps: 500
|
67 |
- mixed_precision_training: Native AMP
|
68 |
|
69 |
+
### Training results
|
70 |
+
|
71 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
72 |
+
|:-------------:|:-----:|:----:|:---------------:|:------:|
|
73 |
+
| 0.0 | 500.0 | 500 | 0.0922 | 4.5455 |
|
74 |
+
|
75 |
+
|
76 |
### Framework versions
|
77 |
|
78 |
+
- Transformers 4.38.2
|
79 |
+
- Pytorch 2.2.1+cu121
|
80 |
- Datasets 2.18.0
|
81 |
- Tokenizers 0.15.2
|
generation_config.json
CHANGED
@@ -252,5 +252,5 @@
|
|
252 |
"transcribe": 50359,
|
253 |
"translate": 50358
|
254 |
},
|
255 |
-
"transformers_version": "4.
|
256 |
}
|
|
|
252 |
"transcribe": 50359,
|
253 |
"translate": 50358
|
254 |
},
|
255 |
+
"transformers_version": "4.38.2"
|
256 |
}
|