End of training
Browse files- README.md +21 -11
- model.safetensors +1 -1
README.md
CHANGED
|
@@ -17,8 +17,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 17 |
|
| 18 |
This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on an unknown dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
-
- Loss:
|
| 21 |
-
- Wer:
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
@@ -44,23 +44,33 @@ The following hyperparameters were used during training:
|
|
| 44 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 45 |
- lr_scheduler_type: linear
|
| 46 |
- lr_scheduler_warmup_steps: 50
|
| 47 |
-
-
|
| 48 |
- mixed_precision_training: Native AMP
|
| 49 |
|
| 50 |
### Training results
|
| 51 |
|
| 52 |
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 53 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
| 54 |
-
| No log | 0.2 | 1 | 3.9843 | 120.0 |
|
| 55 |
-
| No log | 0.4 | 2 | 3.9621 | 120.0 |
|
| 56 |
-
| No log | 0.6 | 3 | 3.9201 | 120.0 |
|
| 57 |
-
| No log | 0.8 | 4 | 3.8624 | 120.0 |
|
| 58 |
| 3.9419 | 1.0 | 5 | 3.7933 | 120.0 |
|
| 59 |
-
| 3.9419 | 1.2 | 6 | 3.7164 | 120.0 |
|
| 60 |
-
| 3.9419 | 1.4 | 7 | 3.6329 | 120.0 |
|
| 61 |
-
| 3.9419 | 1.6 | 8 | 3.5400 | 120.0 |
|
| 62 |
-
| 3.9419 | 1.8 | 9 | 3.4343 | 120.0 |
|
| 63 |
| 3.5809 | 2.0 | 10 | 3.3116 | 115.0000 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 64 |
|
| 65 |
|
| 66 |
### Framework versions
|
|
|
|
| 17 |
|
| 18 |
This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on an unknown dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 0.0772
|
| 21 |
+
- Wer: 390.0
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
|
|
| 44 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 45 |
- lr_scheduler_type: linear
|
| 46 |
- lr_scheduler_warmup_steps: 50
|
| 47 |
+
- training_steps: 100
|
| 48 |
- mixed_precision_training: Native AMP
|
| 49 |
|
| 50 |
### Training results
|
| 51 |
|
| 52 |
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 53 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 54 |
| 3.9419 | 1.0 | 5 | 3.7933 | 120.0 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 55 |
| 3.5809 | 2.0 | 10 | 3.3116 | 115.0000 |
|
| 56 |
+
| 2.9392 | 3.0 | 15 | 2.5977 | 115.0000 |
|
| 57 |
+
| 2.4416 | 4.0 | 20 | 2.1190 | 100.0 |
|
| 58 |
+
| 1.9172 | 5.0 | 25 | 1.5509 | 100.0 |
|
| 59 |
+
| 1.3762 | 6.0 | 30 | 1.0480 | 185.0 |
|
| 60 |
+
| 0.9056 | 7.0 | 35 | 0.7003 | 6325.0 |
|
| 61 |
+
| 0.5587 | 8.0 | 40 | 0.3793 | 3885.0 |
|
| 62 |
+
| 0.3094 | 9.0 | 45 | 0.2168 | 3005.0 |
|
| 63 |
+
| 0.1747 | 10.0 | 50 | 0.1447 | 2455.0 |
|
| 64 |
+
| 0.101 | 11.0 | 55 | 0.1192 | 160.0 |
|
| 65 |
+
| 0.0621 | 12.0 | 60 | 0.0981 | 310.0 |
|
| 66 |
+
| 0.0389 | 13.0 | 65 | 0.0968 | 300.0 |
|
| 67 |
+
| 0.0252 | 14.0 | 70 | 0.0831 | 365.0 |
|
| 68 |
+
| 0.0163 | 15.0 | 75 | 0.0837 | 490.0000 |
|
| 69 |
+
| 0.0118 | 16.0 | 80 | 0.0790 | 450.0 |
|
| 70 |
+
| 0.009 | 17.0 | 85 | 0.0770 | 395.0 |
|
| 71 |
+
| 0.0076 | 18.0 | 90 | 0.0772 | 390.0 |
|
| 72 |
+
| 0.0069 | 19.0 | 95 | 0.0774 | 390.0 |
|
| 73 |
+
| 0.0066 | 20.0 | 100 | 0.0772 | 390.0 |
|
| 74 |
|
| 75 |
|
| 76 |
### Framework versions
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 151061672
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d1354c3a904fd74cf3a33981ae5441508be3823e560dc84b3562efe640aefe7e
|
| 3 |
size 151061672
|