Hugging Face's logo

r-f
/
wav2vec-english-speech-emotion-recognition

Automatic Speech Recognition
Transformers
PyTorch
wav2vec2
Generated from Trainer
Model card Files Community
4
r-f commited on
Commit
574f7d9
·
1 Parent(s): 20217d5

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +17 -17
README.md CHANGED
@@ -48,20 +48,20 @@ The following hyperparameters were used during training:
48
  - mixed_precision_training: Native AMP
49
 
50
  ### Training results
51
-
52
- Step Training Loss Validation Loss Accuracy
53
- 500 1.812400 1.365212 0.486258
54
- 1000 0.887200 0.773145 0.797040
55
- 1500 0.703500 0.574954 0.852008
56
- 2000 0.687900 1.286738 0.775899
57
- 2500 0.649800 0.697455 0.832981
58
- 3000 0.569600 0.337240 0.892178
59
- 3500 0.421800 0.307072 0.911205
60
- 4000 0.308800 0.374443 0.930233
61
- 4500 0.268800 0.260444 0.936575
62
- 5000 0.297300 0.302985 0.923890
63
- 5500 0.176500 0.165439 0.961945
64
- 6000 0.147500 0.170199 0.961945
65
- 6500 0.127400 0.155310 0.966173
66
- 7000 0.069900 0.103882 0.976744
67
- 7500 0.083000 0.104075 0.974630
 
48
  - mixed_precision_training: Native AMP
49
 
50
  ### Training results
51
+ | Step | Training Loss | Validation Loss | Accuracy |
52
+ | ---- | ------------- | --------------- | -------- |
53
+ | 500 | 1.8124 | 1.365212 | 0.486258 |
54
+ | 1000 | 0.8872 | 0.773145 | 0.79704 |
55
+ | 1500 | 0.7035 | 0.574954 | 0.852008 |
56
+ | 2000 | 0.6879 | 1.286738 | 0.775899 |
57
+ | 2500 | 0.6498 | 0.697455 | 0.832981 |
58
+ | 3000 | 0.5696 | 0.33724 | 0.892178 |
59
+ | 3500 | 0.4218 | 0.307072 | 0.911205 |
60
+ | 4000 | 0.3088 | 0.374443 | 0.930233 |
61
+ | 4500 | 0.2688 | 0.260444 | 0.936575 |
62
+ | 5000 | 0.2973 | 0.302985 | 0.92389 |
63
+ | 5500 | 0.1765 | 0.165439 | 0.961945 |
64
+ | 6000 | 0.1475 | 0.170199 | 0.961945 |
65
+ | 6500 | 0.1274 | 0.15531 | 0.966173 |
66
+ | 7000 | 0.0699 | 0.103882 | 0.976744 |
67
+ | 7500 | 0.083 | 0.104075 | 0.97463 |