End of training
Browse files
README.md
CHANGED
@@ -32,8 +32,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
32 |
|
33 |
This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the PolyAI/minds14 dataset.
|
34 |
It achieves the following results on the evaluation set:
|
35 |
-
- Loss: 0.
|
36 |
-
- Wer Ortho: 0.
|
37 |
- Wer: 0.3282
|
38 |
|
39 |
## Model description
|
@@ -67,13 +67,13 @@ The following hyperparameters were used during training:
|
|
67 |
|
68 |
| Training Loss | Epoch | Step | Validation Loss | Wer Ortho | Wer |
|
69 |
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|
|
70 |
-
| 0.0006 | 17.86 | 500 | 0.
|
71 |
-
| 0.0002 | 35.71 | 1000 | 0.
|
72 |
|
73 |
|
74 |
### Framework versions
|
75 |
|
76 |
- Transformers 4.38.1
|
77 |
- Pytorch 2.1.0+cu121
|
78 |
-
- Datasets 2.
|
79 |
- Tokenizers 0.15.2
|
|
|
32 |
|
33 |
This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the PolyAI/minds14 dataset.
|
34 |
It achieves the following results on the evaluation set:
|
35 |
+
- Loss: 0.7236
|
36 |
+
- Wer Ortho: 0.3270
|
37 |
- Wer: 0.3282
|
38 |
|
39 |
## Model description
|
|
|
67 |
|
68 |
| Training Loss | Epoch | Step | Validation Loss | Wer Ortho | Wer |
|
69 |
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|
|
70 |
+
| 0.0006 | 17.86 | 500 | 0.6651 | 0.3257 | 0.3264 |
|
71 |
+
| 0.0002 | 35.71 | 1000 | 0.7236 | 0.3270 | 0.3282 |
|
72 |
|
73 |
|
74 |
### Framework versions
|
75 |
|
76 |
- Transformers 4.38.1
|
77 |
- Pytorch 2.1.0+cu121
|
78 |
+
- Datasets 2.18.0
|
79 |
- Tokenizers 0.15.2
|
runs/Mar04_14-31-00_728f83ee7730/events.out.tfevents.1709562661.728f83ee7730.725.0
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:dc0dac644d98a604d3e59264721d147aa1dc0154e7cb583dd3b50f2917d6a32c
|
3 |
+
size 15887
|