Hugging Face's logo

laura63
/
wav2vec2-base-finetuned-ks

Audio Classification
Transformers
PyTorch
TensorBoard
wav2vec2
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
laura63 commited on
Commit
578c721
•
1 Parent(s): 050d094

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +14 -16
README.md CHANGED
@@ -5,6 +5,7 @@ tags:
5
  datasets:
6
  - audiofolder
7
  metrics:
 
8
  - f1
9
  model-index:
10
  - name: wav2vec2-base-finetuned-ks
@@ -19,9 +20,12 @@ model-index:
19
  split: train
20
  args: Data_Train
21
  metrics:
 
 
 
22
  - name: F1
23
  type: f1
24
- value: 0.8016517743184772
25
  ---
26
 
27
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -31,8 +35,9 @@ should probably proofread and complete it, then remove this comment. -->
31
 
32
  This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on the audiofolder dataset.
33
  It achieves the following results on the evaluation set:
34
- - Loss: 1.1858
35
- - F1: 0.8017
 
36
 
37
  ## Model description
38
 
@@ -60,22 +65,15 @@ The following hyperparameters were used during training:
60
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
61
  - lr_scheduler_type: linear
62
  - lr_scheduler_warmup_ratio: 0.1
63
- - num_epochs: 10
64
 
65
  ### Training results
66
 
67
- | Training Loss | Epoch | Step | Validation Loss | F1 |
68
- |:-------------:|:-----:|:----:|:---------------:|:------:|
69
- | 2.6077 | 1.0 | 602 | 2.5802 | 0.1587 |
70
- | 1.9999 | 2.0 | 1204 | 2.0069 | 0.3600 |
71
- | 1.4107 | 3.0 | 1806 | 1.6226 | 0.5527 |
72
- | 0.9977 | 4.0 | 2408 | 1.2932 | 0.6719 |
73
- | 0.8132 | 5.0 | 3010 | 1.2297 | 0.7030 |
74
- | 0.5315 | 6.0 | 3612 | 1.0131 | 0.7745 |
75
- | 0.5772 | 7.0 | 4214 | 1.1444 | 0.7782 |
76
- | 0.0248 | 8.0 | 4816 | 1.1777 | 0.7850 |
77
- | 0.0186 | 9.0 | 5418 | 1.2235 | 0.7910 |
78
- | 0.2457 | 10.0 | 6020 | 1.1858 | 0.8017 |
79
 
80
 
81
  ### Framework versions
 
5
  datasets:
6
  - audiofolder
7
  metrics:
8
+ - accuracy
9
  - f1
10
  model-index:
11
  - name: wav2vec2-base-finetuned-ks
 
20
  split: train
21
  args: Data_Train
22
  metrics:
23
+ - name: Accuracy
24
+ type: accuracy
25
+ value: 0.7415185783521809
26
  - name: F1
27
  type: f1
28
+ value: 0.7412464625062914
29
  ---
30
 
31
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
35
 
36
  This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on the audiofolder dataset.
37
  It achieves the following results on the evaluation set:
38
+ - Loss: 1.2511
39
+ - Accuracy: 0.7415
40
+ - F1: 0.7412
41
 
42
  ## Model description
43
 
 
65
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
66
  - lr_scheduler_type: linear
67
  - lr_scheduler_warmup_ratio: 0.1
68
+ - num_epochs: 3
69
 
70
  ### Training results
71
 
72
+ | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
73
+ |:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|
74
+ | 0.8331 | 1.0 | 688 | 1.5519 | 0.6123 | 0.6223 |
75
+ | 0.7505 | 2.0 | 1376 | 1.3095 | 0.6947 | 0.6985 |
76
+ | 0.5294 | 3.0 | 2064 | 1.2511 | 0.7415 | 0.7412 |
 
 
 
 
 
 
 
77
 
78
 
79
  ### Framework versions