Hugging Face's logo

thrunlab
/
t5-large_cola_dense_epochs-5

Text Classification
Transformers
PyTorch
t5
Generated from Trainer
Eval Results
text-generation-inference
Model card Files Community
vxbrandon commited on
Commit
6ec6c71
·
1 Parent(s): 609f891

End of training

Browse files
Files changed (3) hide show
  1. README.md +33 -20
  2. pytorch_model.bin +1 -1
  3. training_args.bin +1 -1
README.md CHANGED
@@ -17,12 +17,12 @@ model-index:
17
  name: glue
18
  type: glue
19
  config: cola
20
- split: validation
21
  args: cola
22
  metrics:
23
  - name: Accuracy
24
  type: accuracy
25
- value: 0.8446788111217641
26
  ---
27
 
28
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -32,8 +32,8 @@ should probably proofread and complete it, then remove this comment. -->
32
 
33
  This model is a fine-tuned version of [t5-large](https://huggingface.co/t5-large) on the glue dataset.
34
  It achieves the following results on the evaluation set:
35
- - Loss: 1.1819
36
- - Accuracy: 0.8447
37
 
38
  ## Model description
39
 
@@ -53,11 +53,11 @@ More information needed
53
 
54
  The following hyperparameters were used during training:
55
  - learning_rate: 5e-05
56
- - train_batch_size: 32
57
- - eval_batch_size: 32
58
  - seed: 0
59
  - gradient_accumulation_steps: 2
60
- - total_train_batch_size: 64
61
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
62
  - lr_scheduler_type: linear
63
  - lr_scheduler_warmup_steps: 20
@@ -67,19 +67,32 @@ The following hyperparameters were used during training:
67
 
68
  | Training Loss | Epoch | Step | Validation Loss | Accuracy |
69
  |:-------------:|:-----:|:----:|:---------------:|:--------:|
70
- | 0.4141 | 0.37 | 50 | 0.5198 | 0.8025 |
71
- | 0.3263 | 0.75 | 100 | 0.5361 | 0.8245 |
72
- | 0.284 | 1.12 | 150 | 0.5351 | 0.8360 |
73
- | 0.2565 | 1.49 | 200 | 0.4917 | 0.8322 |
74
- | 0.3137 | 1.87 | 250 | 0.4415 | 0.8418 |
75
- | 0.1998 | 2.24 | 300 | 2.1507 | 0.8360 |
76
- | 0.1983 | 2.61 | 350 | 0.8461 | 0.8284 |
77
- | 0.1874 | 2.99 | 400 | 0.6088 | 0.8351 |
78
- | 0.1321 | 3.36 | 450 | 2.0126 | 0.8466 |
79
- | 0.3057 | 3.73 | 500 | 1.1819 | 0.8447 |
80
- | 0.0939 | 4.1 | 550 | 1.6535 | 0.8437 |
81
- | 0.0894 | 4.48 | 600 | 3.4434 | 0.8408 |
82
- | 0.1519 | 4.85 | 650 | 3.1519 | 0.8428 |
 
 
 
 
 
 
 
 
 
 
 
 
 
83
 
84
 
85
  ### Framework versions
 
17
  name: glue
18
  type: glue
19
  config: cola
20
+ split: train
21
  args: cola
22
  metrics:
23
  - name: Accuracy
24
  type: accuracy
25
+ value: 0.8813559322033898
26
  ---
27
 
28
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
32
 
33
  This model is a fine-tuned version of [t5-large](https://huggingface.co/t5-large) on the glue dataset.
34
  It achieves the following results on the evaluation set:
35
+ - Loss: 0.4167
36
+ - Accuracy: 0.8814
37
 
38
  ## Model description
39
 
 
53
 
54
  The following hyperparameters were used during training:
55
  - learning_rate: 5e-05
56
+ - train_batch_size: 64
57
+ - eval_batch_size: 64
58
  - seed: 0
59
  - gradient_accumulation_steps: 2
60
+ - total_train_batch_size: 128
61
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
62
  - lr_scheduler_type: linear
63
  - lr_scheduler_warmup_steps: 20
 
67
 
68
  | Training Loss | Epoch | Step | Validation Loss | Accuracy |
69
  |:-------------:|:-----:|:----:|:---------------:|:--------:|
70
+ | 0.6749 | 0.19 | 10 | 0.6270 | 0.7095 |
71
+ | 0.5772 | 0.37 | 20 | 0.5947 | 0.7101 |
72
+ | 0.6066 | 0.56 | 30 | 0.5545 | 0.7101 |
73
+ | 0.5355 | 0.75 | 40 | 0.4788 | 0.7475 |
74
+ | 0.4398 | 0.93 | 50 | 0.3992 | 0.8469 |
75
+ | 0.3932 | 1.12 | 60 | 0.3737 | 0.8638 |
76
+ | 0.3756 | 1.31 | 70 | 0.3606 | 0.8650 |
77
+ | 0.4004 | 1.5 | 80 | 0.3645 | 0.8603 |
78
+ | 0.3198 | 1.68 | 90 | 0.3201 | 0.8749 |
79
+ | 0.3129 | 1.87 | 100 | 0.3638 | 0.8697 |
80
+ | 0.2763 | 2.06 | 110 | 0.3091 | 0.8819 |
81
+ | 0.3207 | 2.24 | 120 | 0.3781 | 0.8673 |
82
+ | 0.2614 | 2.43 | 130 | 0.3351 | 0.8773 |
83
+ | 0.2909 | 2.62 | 140 | 0.3404 | 0.8662 |
84
+ | 0.2899 | 2.8 | 150 | 0.3277 | 0.8796 |
85
+ | 0.2687 | 2.99 | 160 | 0.3520 | 0.8679 |
86
+ | 0.1993 | 3.18 | 170 | 0.3319 | 0.8854 |
87
+ | 0.2584 | 3.36 | 180 | 0.3901 | 0.8732 |
88
+ | 0.2502 | 3.55 | 190 | 0.3766 | 0.8773 |
89
+ | 0.2234 | 3.74 | 200 | 0.3360 | 0.8895 |
90
+ | 0.2101 | 3.93 | 210 | 0.3334 | 0.8849 |
91
+ | 0.1708 | 4.11 | 220 | 0.3819 | 0.8714 |
92
+ | 0.1664 | 4.3 | 230 | 0.3690 | 0.8773 |
93
+ | 0.2217 | 4.49 | 240 | 0.4181 | 0.8814 |
94
+ | 0.2034 | 4.67 | 250 | 0.3607 | 0.8796 |
95
+ | 0.1948 | 4.86 | 260 | 0.4167 | 0.8814 |
96
 
97
 
98
  ### Framework versions
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:96d9650874943111370933df8327af707dc69afae354117c29224a0b4ebbec53
3
  size 2955087438
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6305bb42bebe28d40b18df6011fa959ac833a9acdac42da22818e506588b5854
3
  size 2955087438
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4339e660e9bd1d9f7fd9e092dda83b2823b78698a2e1eef9f0c3a43bbbe3d171
3
  size 4536
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:96cead93cc4c8a588405fc0e22ab38fcd9bbf72a68260e378646bb370c51ce5e
3
  size 4536