gcmsrc commited on
Commit
d30535f
1 Parent(s): 170ce08

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +13 -12
README.md CHANGED
@@ -16,14 +16,14 @@ model-index:
16
  dataset:
17
  name: emotion
18
  type: emotion
19
- args: default
20
  metrics:
21
  - name: Accuracy
22
  type: accuracy
23
- value: 0.9245
24
  - name: F1
25
  type: f1
26
- value: 0.9247712048482103
27
  ---
28
 
29
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -33,9 +33,9 @@ should probably proofread and complete it, then remove this comment. -->
33
 
34
  This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the emotion dataset.
35
  It achieves the following results on the evaluation set:
36
- - Loss: 0.2179
37
- - Accuracy: 0.9245
38
- - F1: 0.9248
39
 
40
  ## Model description
41
 
@@ -54,25 +54,26 @@ More information needed
54
  ### Training hyperparameters
55
 
56
  The following hyperparameters were used during training:
57
- - learning_rate: 2e-05
58
  - train_batch_size: 64
59
  - eval_batch_size: 64
60
  - seed: 42
61
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
62
  - lr_scheduler_type: linear
63
- - num_epochs: 2
64
 
65
  ### Training results
66
 
67
  | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
68
  |:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|
69
- | 0.8178 | 1.0 | 250 | 0.3219 | 0.9035 | 0.8996 |
70
- | 0.2526 | 2.0 | 500 | 0.2179 | 0.9245 | 0.9248 |
 
71
 
72
 
73
  ### Framework versions
74
 
75
  - Transformers 4.13.0
76
- - Pytorch 1.12.1+cu113
77
- - Datasets 1.16.1
78
  - Tokenizers 0.10.3
 
16
  dataset:
17
  name: emotion
18
  type: emotion
19
+ args: split
20
  metrics:
21
  - name: Accuracy
22
  type: accuracy
23
+ value: 0.9355
24
  - name: F1
25
  type: f1
26
+ value: 0.9356480877541032
27
  ---
28
 
29
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
33
 
34
  This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the emotion dataset.
35
  It achieves the following results on the evaluation set:
36
+ - Loss: 0.1424
37
+ - Accuracy: 0.9355
38
+ - F1: 0.9356
39
 
40
  ## Model description
41
 
 
54
  ### Training hyperparameters
55
 
56
  The following hyperparameters were used during training:
57
+ - learning_rate: 5e-05
58
  - train_batch_size: 64
59
  - eval_batch_size: 64
60
  - seed: 42
61
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
62
  - lr_scheduler_type: linear
63
+ - num_epochs: 3.0
64
 
65
  ### Training results
66
 
67
  | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
68
  |:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|
69
+ | 0.5311 | 1.0 | 250 | 0.1817 | 0.932 | 0.9317 |
70
+ | 0.14 | 2.0 | 500 | 0.1483 | 0.9365 | 0.9368 |
71
+ | 0.0915 | 3.0 | 750 | 0.1424 | 0.9355 | 0.9356 |
72
 
73
 
74
  ### Framework versions
75
 
76
  - Transformers 4.13.0
77
+ - Pytorch 1.10.2+cu102
78
+ - Datasets 2.8.0
79
  - Tokenizers 0.10.3