Hugging Face's logo

ghazikhanihamed
/
TooT-PLM-P2S

Transformers
PyTorch
t5
text2text-generation
Generated from Trainer
text-generation-inference
Model card Files
xet
 Community
ghazikhanihamed commited on
Commit
4e81753
·
1 Parent(s): feef391

PLM-Secondary-Structure-Generation

Browse files
Files changed (4) hide show
  1. README.md +30 -22
  2. pytorch_model.bin +2 -2
  3. tokenizer.json +2 -2
  4. training_args.bin +2 -2
README.md CHANGED
@@ -15,8 +15,8 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  This model is a fine-tuned version of [ElnaggarLab/ankh-base](https://huggingface.co/ElnaggarLab/ankh-base) on the None dataset.
17
  It achieves the following results on the evaluation set:
18
- - Loss: 0.1649
19
- - Q3 Accuracy: 0.5470
20
 
21
  ## Model description
22
 
@@ -35,36 +35,44 @@ More information needed
35
  ### Training hyperparameters
36
 
37
  The following hyperparameters were used during training:
38
- - learning_rate: 0.001
39
  - train_batch_size: 1
40
  - eval_batch_size: 8
41
  - seed: 42
42
- - distributed_type: multi-GPU
43
- - num_devices: 4
44
- - gradient_accumulation_steps: 4
45
- - total_train_batch_size: 16
46
- - total_eval_batch_size: 32
47
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
48
  - lr_scheduler_type: linear
49
- - lr_scheduler_warmup_ratio: 0.3
50
- - num_epochs: 1
51
 
52
  ### Training results
53
 
54
  | Training Loss | Epoch | Step | Validation Loss | Q3 Accuracy |
55
  |:-------------:|:-----:|:----:|:---------------:|:-----------:|
56
- | 1.0993 | 0.08 | 50 | 0.3957 | 0.4526 |
57
- | 0.4077 | 0.16 | 100 | 0.3564 | 0.5574 |
58
- | 0.3475 | 0.24 | 150 | 0.3082 | 0.4327 |
59
- | 0.2474 | 0.31 | 200 | 0.2171 | 0.5477 |
60
- | 0.2567 | 0.39 | 250 | 0.3058 | 0.4924 |
61
- | 0.2948 | 0.47 | 300 | 0.1828 | 0.5470 |
62
- | 0.2148 | 0.55 | 350 | 0.1873 | 0.5470 |
63
- | 0.209 | 0.63 | 400 | 0.1832 | 0.5470 |
64
- | 0.1761 | 0.71 | 450 | 0.1791 | 0.5103 |
65
- | 0.1791 | 0.79 | 500 | 0.1700 | 0.5470 |
66
- | 0.1641 | 0.87 | 550 | 0.1684 | 0.5470 |
67
- | 0.1653 | 0.94 | 600 | 0.1649 | 0.5470 |
 
 
 
 
 
 
 
 
 
 
 
68
 
69
 
70
  ### Framework versions
 
15
 
16
  This model is a fine-tuned version of [ElnaggarLab/ankh-base](https://huggingface.co/ElnaggarLab/ankh-base) on the None dataset.
17
  It achieves the following results on the evaluation set:
18
+ - Loss: 0.2497
19
+ - Q3 Accuracy: 0.3837
20
 
21
  ## Model description
22
 
 
35
  ### Training hyperparameters
36
 
37
  The following hyperparameters were used during training:
38
+ - learning_rate: 0.00043
39
  - train_batch_size: 1
40
  - eval_batch_size: 8
41
  - seed: 42
42
+ - gradient_accumulation_steps: 2
43
+ - total_train_batch_size: 2
 
 
 
44
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
45
  - lr_scheduler_type: linear
46
+ - lr_scheduler_warmup_ratio: 0.4
47
+ - num_epochs: 10
48
 
49
  ### Training results
50
 
51
  | Training Loss | Epoch | Step | Validation Loss | Q3 Accuracy |
52
  |:-------------:|:-----:|:----:|:---------------:|:-----------:|
53
+ | 7.2902 | 0.01 | 50 | 6.9889 | 0.0256 |
54
+ | 5.3162 | 0.02 | 100 | 4.3754 | 0.0189 |
55
+ | 3.1825 | 0.03 | 150 | 2.8131 | 0.0226 |
56
+ | 2.3638 | 0.04 | 200 | 2.2350 | 0.0292 |
57
+ | 1.7805 | 0.05 | 250 | 1.9359 | 0.0337 |
58
+ | 1.809 | 0.06 | 300 | 1.7340 | 0.0272 |
59
+ | 1.4045 | 0.06 | 350 | 1.5957 | 0.0201 |
60
+ | 1.3332 | 0.07 | 400 | 1.4751 | 0.0256 |
61
+ | 1.3069 | 0.08 | 450 | 1.3454 | 0.0323 |
62
+ | 1.1402 | 0.09 | 500 | 1.1687 | 0.0628 |
63
+ | 0.9261 | 0.1 | 550 | 0.9721 | 0.0807 |
64
+ | 0.7871 | 0.11 | 600 | 0.8113 | 0.3194 |
65
+ | 0.6571 | 0.12 | 650 | 0.6875 | 0.3076 |
66
+ | 0.5671 | 0.13 | 700 | 0.5642 | 0.3395 |
67
+ | 0.5278 | 0.14 | 750 | 0.4711 | 0.3233 |
68
+ | 0.4502 | 0.15 | 800 | 0.4269 | 0.3471 |
69
+ | 0.4006 | 0.16 | 850 | 0.3827 | 0.3495 |
70
+ | 0.3446 | 0.17 | 900 | 0.3413 | 0.3303 |
71
+ | 0.3375 | 0.18 | 950 | 0.3217 | 0.3433 |
72
+ | 0.3065 | 0.19 | 1000 | 0.3054 | 0.3459 |
73
+ | 0.2858 | 0.19 | 1050 | 0.3106 | 0.3519 |
74
+ | 0.2861 | 0.2 | 1100 | 0.2863 | 0.3696 |
75
+ | 0.2258 | 0.21 | 1150 | 0.2497 | 0.3837 |
76
 
77
 
78
  ### Framework versions
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:18fdf824df75a1a0b4b3b77df233d6eb7d72ba594951b3071eda50b2b6bbd0b0
3
- size 2946957610
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7c19569f3482c43061d22a1bde94cdf837d26c1e97e82e6d8e58f0b10664e05b
3
+ size 2946083298
tokenizer.json CHANGED
@@ -2,13 +2,13 @@
2
  "version": "1.0",
3
  "truncation": {
4
  "direction": "Right",
5
- "max_length": 825,
6
  "strategy": "LongestFirst",
7
  "stride": 0
8
  },
9
  "padding": {
10
  "strategy": {
11
- "Fixed": 825
12
  },
13
  "direction": "Right",
14
  "pad_to_multiple_of": null,
 
2
  "version": "1.0",
3
  "truncation": {
4
  "direction": "Right",
5
+ "max_length": 809,
6
  "strategy": "LongestFirst",
7
  "stride": 0
8
  },
9
  "padding": {
10
  "strategy": {
11
+ "Fixed": 809
12
  },
13
  "direction": "Right",
14
  "pad_to_multiple_of": null,
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:52c0ad81e3b3de1651910d8e6ca794e1eef1803026e9e4eb120b48685a1838ff
3
- size 5880
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:524d4852fa44265f2af5909b950bc6c223e873d0dcd65e962de4dd34e964148d
3
+ size 4536