End of training
Browse files
README.md
CHANGED
@@ -18,10 +18,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
18 |
|
19 |
# whisper small tl - CSB05
|
20 |
|
21 |
-
This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on
|
22 |
It achieves the following results on the evaluation set:
|
23 |
-
- Loss: 0.
|
24 |
-
- Wer:
|
25 |
|
26 |
## Model description
|
27 |
|
@@ -44,7 +44,7 @@ The following hyperparameters were used during training:
|
|
44 |
- train_batch_size: 16
|
45 |
- eval_batch_size: 8
|
46 |
- seed: 42
|
47 |
-
- optimizer:
|
48 |
- lr_scheduler_type: linear
|
49 |
- lr_scheduler_warmup_steps: 500
|
50 |
- training_steps: 4000
|
@@ -54,15 +54,15 @@ The following hyperparameters were used during training:
|
|
54 |
|
55 |
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
56 |
|:-------------:|:-------:|:----:|:---------------:|:-------:|
|
57 |
-
| 0.
|
58 |
-
| 0.
|
59 |
-
| 0.
|
60 |
-
| 0.
|
61 |
|
62 |
|
63 |
### Framework versions
|
64 |
|
65 |
-
- Transformers 4.
|
66 |
-
- Pytorch 2.
|
67 |
-
- Datasets 3.0
|
68 |
-
- Tokenizers 0.20.
|
|
|
18 |
|
19 |
# whisper small tl - CSB05
|
20 |
|
21 |
+
This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the None dataset.
|
22 |
It achieves the following results on the evaluation set:
|
23 |
+
- Loss: 0.8685
|
24 |
+
- Wer: 24.4015
|
25 |
|
26 |
## Model description
|
27 |
|
|
|
44 |
- train_batch_size: 16
|
45 |
- eval_batch_size: 8
|
46 |
- seed: 42
|
47 |
+
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
|
48 |
- lr_scheduler_type: linear
|
49 |
- lr_scheduler_warmup_steps: 500
|
50 |
- training_steps: 4000
|
|
|
54 |
|
55 |
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
56 |
|:-------------:|:-------:|:----:|:---------------:|:-------:|
|
57 |
+
| 0.0158 | 8.9286 | 1000 | 0.6826 | 24.1285 |
|
58 |
+
| 0.0019 | 17.8571 | 2000 | 0.7977 | 24.7795 |
|
59 |
+
| 0.0003 | 26.7857 | 3000 | 0.8517 | 24.4645 |
|
60 |
+
| 0.0002 | 35.7143 | 4000 | 0.8685 | 24.4015 |
|
61 |
|
62 |
|
63 |
### Framework versions
|
64 |
|
65 |
+
- Transformers 4.46.1
|
66 |
+
- Pytorch 2.5.0+cu121
|
67 |
+
- Datasets 3.1.0
|
68 |
+
- Tokenizers 0.20.1
|
generation_config.json
CHANGED
@@ -250,5 +250,5 @@
|
|
250 |
"transcribe": 50359,
|
251 |
"translate": 50358
|
252 |
},
|
253 |
-
"transformers_version": "4.
|
254 |
}
|
|
|
250 |
"transcribe": 50359,
|
251 |
"translate": 50358
|
252 |
},
|
253 |
+
"transformers_version": "4.46.1"
|
254 |
}
|
model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 966995080
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d36eeca7ac590d05618468a7cbd490c04fd4181db7c778b6838b8a3e12298821
|
3 |
size 966995080
|
runs/Nov01_02-26-19_ed1deb200416/events.out.tfevents.1730427982.ed1deb200416.769.0
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1cd65214c1f30fd2ec950239fc24eaab8b14e7108086995ba0002b24555b46c1
|
3 |
+
size 42158
|