gokuls commited on
Commit
94eb393
1 Parent(s): 5af65cd

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +14 -18
README.md CHANGED
@@ -1,6 +1,4 @@
1
  ---
2
- language:
3
- - en
4
  tags:
5
  - generated_from_trainer
6
  datasets:
@@ -14,7 +12,7 @@ model-index:
14
  name: Text Classification
15
  type: text-classification
16
  dataset:
17
- name: GLUE STSB
18
  type: glue
19
  config: stsb
20
  split: validation
@@ -22,7 +20,7 @@ model-index:
22
  metrics:
23
  - name: Spearmanr
24
  type: spearmanr
25
- value: 0.06376297357455894
26
  ---
27
 
28
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -30,12 +28,12 @@ should probably proofread and complete it, then remove this comment. -->
30
 
31
  # hBERTv1_new_pretrain_w_init__stsb
32
 
33
- This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1_complete_training_new_wt_init](https://huggingface.co/gokuls/bert_12_layer_model_v1_complete_training_new_wt_init) on the GLUE STSB dataset.
34
  It achieves the following results on the evaluation set:
35
- - Loss: 2.2578
36
- - Pearson: 0.0646
37
- - Spearmanr: 0.0638
38
- - Combined Score: 0.0642
39
 
40
  ## Model description
41
 
@@ -54,7 +52,7 @@ More information needed
54
  ### Training hyperparameters
55
 
56
  The following hyperparameters were used during training:
57
- - learning_rate: 0.0005
58
  - train_batch_size: 128
59
  - eval_batch_size: 128
60
  - seed: 10
@@ -67,14 +65,12 @@ The following hyperparameters were used during training:
67
 
68
  | Training Loss | Epoch | Step | Validation Loss | Pearson | Spearmanr | Combined Score |
69
  |:-------------:|:-----:|:----:|:---------------:|:-------:|:---------:|:--------------:|
70
- | 61.8804 | 1.0 | 45 | 2.3512 | -0.0440 | 0.0212 | -0.0114 |
71
- | 2.6713 | 2.0 | 90 | 3.0563 | 0.0479 | 0.0125 | 0.0302 |
72
- | 2.3803 | 3.0 | 135 | 2.2578 | 0.0646 | 0.0638 | 0.0642 |
73
- | 2.2516 | 4.0 | 180 | 2.3252 | 0.0646 | 0.0638 | 0.0642 |
74
- | 2.243 | 5.0 | 225 | 2.2928 | 0.0646 | 0.0638 | 0.0642 |
75
- | 2.1764 | 6.0 | 270 | 2.2826 | 0.0632 | 0.0058 | 0.0345 |
76
- | 2.1933 | 7.0 | 315 | 2.3738 | 0.0495 | 0.0497 | 0.0496 |
77
- | 2.2067 | 8.0 | 360 | 2.2834 | 0.0372 | -0.0148 | 0.0112 |
78
 
79
 
80
  ### Framework versions
 
1
  ---
 
 
2
  tags:
3
  - generated_from_trainer
4
  datasets:
 
12
  name: Text Classification
13
  type: text-classification
14
  dataset:
15
+ name: glue
16
  type: glue
17
  config: stsb
18
  split: validation
 
20
  metrics:
21
  - name: Spearmanr
22
  type: spearmanr
23
+ value: 0.4406348486823185
24
  ---
25
 
26
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
28
 
29
  # hBERTv1_new_pretrain_w_init__stsb
30
 
31
+ This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1_complete_training_new_wt_init](https://huggingface.co/gokuls/bert_12_layer_model_v1_complete_training_new_wt_init) on the glue dataset.
32
  It achieves the following results on the evaluation set:
33
+ - Loss: 2.8004
34
+ - Pearson: 0.4571
35
+ - Spearmanr: 0.4406
36
+ - Combined Score: 0.4489
37
 
38
  ## Model description
39
 
 
52
  ### Training hyperparameters
53
 
54
  The following hyperparameters were used during training:
55
+ - learning_rate: 4e-05
56
  - train_batch_size: 128
57
  - eval_batch_size: 128
58
  - seed: 10
 
65
 
66
  | Training Loss | Epoch | Step | Validation Loss | Pearson | Spearmanr | Combined Score |
67
  |:-------------:|:-----:|:----:|:---------------:|:-------:|:---------:|:--------------:|
68
+ | 2.5056 | 1.0 | 45 | 2.2584 | 0.0949 | 0.0892 | 0.0920 |
69
+ | 2.1254 | 2.0 | 90 | 2.6871 | 0.1250 | 0.1231 | 0.1241 |
70
+ | 1.9839 | 3.0 | 135 | 2.2709 | 0.1790 | 0.1840 | 0.1815 |
71
+ | 1.6299 | 4.0 | 180 | 2.5115 | 0.2691 | 0.2797 | 0.2744 |
72
+ | 1.3155 | 5.0 | 225 | 2.4555 | 0.3453 | 0.3437 | 0.3445 |
73
+ | 0.9686 | 6.0 | 270 | 2.8004 | 0.4571 | 0.4406 | 0.4489 |
 
 
74
 
75
 
76
  ### Framework versions