Hugging Face's logo

ZhihCheng
/
whisper-base-with-motor_zh

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
Chinese
whisper
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
ZhihCheng commited on
Commit
14b2229
•
1 Parent(s): 90cc4b1

End of training

Browse files
Files changed (3) hide show
  1. README.md +13 -7
  2. model.safetensors +1 -1
  3. runs/Feb29_10-50-26_v2rlitctr1709172647552-ts7bt/events.out.tfevents.1709175078.v2rlitctr1709172647552-ts7bt.773.1 +2 -2
README.md CHANGED
@@ -19,8 +19,8 @@ should probably proofread and complete it, then remove this comment. -->
19
 
20
  This model is a fine-tuned version of [ZhihCheng/whisper-base-zh](https://huggingface.co/ZhihCheng/whisper-base-zh) on the None dataset.
21
  It achieves the following results on the evaluation set:
22
- - Loss: 0.1822
23
- - Wer: 75.0
24
 
25
  ## Model description
26
 
@@ -46,17 +46,23 @@ The following hyperparameters were used during training:
46
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
47
  - lr_scheduler_type: linear
48
  - lr_scheduler_warmup_steps: 25
49
- - training_steps: 200
50
  - mixed_precision_training: Native AMP
51
 
52
  ### Training results
53
 
54
  | Training Loss | Epoch | Step | Validation Loss | Wer |
55
  |:-------------:|:-----:|:----:|:---------------:|:-----:|
56
- | 0.1856 | 0.85 | 50 | 0.3361 | 81.25 |
57
- | 0.0434 | 1.69 | 100 | 0.2016 | 75.0 |
58
- | 0.0187 | 2.54 | 150 | 0.1727 | 75.0 |
59
- | 0.009 | 3.39 | 200 | 0.1822 | 75.0 |
 
 
 
 
 
 
60
 
61
 
62
  ### Framework versions
 
19
 
20
  This model is a fine-tuned version of [ZhihCheng/whisper-base-zh](https://huggingface.co/ZhihCheng/whisper-base-zh) on the None dataset.
21
  It achieves the following results on the evaluation set:
22
+ - Loss: 0.1986
23
+ - Wer: 68.75
24
 
25
  ## Model description
26
 
 
46
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
47
  - lr_scheduler_type: linear
48
  - lr_scheduler_warmup_steps: 25
49
+ - training_steps: 500
50
  - mixed_precision_training: Native AMP
51
 
52
  ### Training results
53
 
54
  | Training Loss | Epoch | Step | Validation Loss | Wer |
55
  |:-------------:|:-----:|:----:|:---------------:|:-----:|
56
+ | 0.0375 | 0.85 | 50 | 0.1724 | 68.75 |
57
+ | 0.0164 | 1.69 | 100 | 0.1800 | 68.75 |
58
+ | 0.0044 | 2.54 | 150 | 0.1788 | 68.75 |
59
+ | 0.0031 | 3.39 | 200 | 0.1927 | 68.75 |
60
+ | 0.0009 | 4.24 | 250 | 0.1736 | 62.5 |
61
+ | 0.0009 | 5.08 | 300 | 0.1984 | 75.0 |
62
+ | 0.0006 | 5.93 | 350 | 0.2019 | 75.0 |
63
+ | 0.0005 | 6.78 | 400 | 0.1988 | 75.0 |
64
+ | 0.0005 | 7.63 | 450 | 0.1989 | 68.75 |
65
+ | 0.0005 | 8.47 | 500 | 0.1986 | 68.75 |
66
 
67
 
68
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a37df2a570882fb545ee3158b77a083a5200ca487c8a27bf8802f395d65a0f5d
3
  size 290403936
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:690607880fa82ad12724afcb2848656bfa9dd43460c615422a52a602f6197b25
3
  size 290403936
runs/Feb29_10-50-26_v2rlitctr1709172647552-ts7bt/events.out.tfevents.1709175078.v2rlitctr1709172647552-ts7bt.773.1 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b16db621d9159e2ee0efc63cec46d031ff62e18e2c3f0b3bec4904d8fe29cff8
3
- size 10883
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e35df4073e9b595a874c418e07a3e1ccbd9553b9955304c811966ea287bb81ec
3
+ size 11869