File size: 3,773 Bytes
c26605f
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
train 2418, val 278, test 953  (+extra from /home/sjkim/POLYEDIT/data/real/raw)
The tokenizer class you load from this checkpoint is not the same type as the class this function is called from. It may result in unexpected tokenization. 
The tokenizer class you load from this checkpoint is 'RobertaTokenizer'. 
The class this function is called from is 'PolymerSmilesTokenizer'.
Some weights of RobertaModel were not initialized from the model checkpoint at ChangwenXu/TransPolymer and are newly initialized: ['roberta.pooler.dense.bias', 'roberta.pooler.dense.weight']
You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.
wandb: Currently logged in as: promotionkim (mhchoiiii-Korea Advanced Institute of Science and Technology) to https://api.wandb.ai. Use `wandb login --relogin` to force relogin
wandb: setting up run s6oubhqd
wandb: Tracking run with wandb version 0.23.1
wandb: Run data is saved locally in /home/sjkim/POLYEDIT/wandb/run-20260803_143623-s6oubhqd
wandb: Run `wandb offline` to turn off syncing.
wandb: Syncing run polyedit-verifier-transpolymer-full80
wandb: ⭐️ View project at https://wandb.ai/promotion-kim/polyedit
wandb: πŸš€ View run at https://wandb.ai/promotion-kim/polyedit/runs/s6oubhqd
epoch   1  train 0.8608  val 0.5379
epoch   2  train 0.4393  val 0.2617
epoch   3  train 0.2513  val 0.1970
epoch   4  train 0.1835  val 0.1749
epoch   5  train 0.1486  val 0.1584
epoch   6  train 0.1276  val 0.1589
epoch   7  train 0.1059  val 0.1513
epoch   8  train 0.0944  val 0.1606
epoch   9  train 0.0811  val 0.1690
epoch  10  train 0.0734  val 0.1555
epoch  11  train 0.0675  val 0.1700
epoch  12  train 0.0622  val 0.1690
early stop at epoch 12 (best val 0.1513)
wandb: updating run metadata
wandb: uploading output.log; uploading wandb-summary.json
wandb: uploading history steps 10-12, summary, console lines 10-12
wandb: 
wandb: Run history:
wandb:                       epoch β–β–‚β–‚β–ƒβ–„β–„β–…β–…β–†β–‡β–‡β–ˆ
wandb:       test/EPS_pairwise_acc ▁
wandb:               test/EPS_rmse ▁
wandb: test/EPS_rmse_over_interval ▁
wandb:           test/EPS_spearman ▁
wandb:       test/Eat_pairwise_acc ▁
wandb:               test/Eat_rmse ▁
wandb: test/Eat_rmse_over_interval ▁
wandb:           test/Eat_spearman ▁
wandb:       test/Eea_pairwise_acc ▁
wandb:                         +25 ...
wandb: 
wandb: Run summary:
wandb:                       epoch 12
wandb:       test/EPS_pairwise_acc 0.85117
wandb:               test/EPS_rmse 0.5682
wandb: test/EPS_rmse_over_interval 5.68197
wandb:           test/EPS_spearman 0.87051
wandb:       test/Eat_pairwise_acc 0.91144
wandb:               test/Eat_rmse 0.11029
wandb: test/Eat_rmse_over_interval 1.10288
wandb:           test/Eat_spearman 0.95939
wandb:       test/Eea_pairwise_acc 0.89208
wandb:                         +25 ...
wandb: 
wandb: πŸš€ View run polyedit-verifier-transpolymer-full80 at: https://wandb.ai/promotion-kim/polyedit/runs/s6oubhqd
wandb: ⭐️ View project at: https://wandb.ai/promotion-kim/polyedit
wandb: Synced 5 W&B file(s), 0 media file(s), 0 artifact file(s) and 0 other file(s)
wandb: Find logs at: ./wandb/run-20260803_143623-s6oubhqd/logs

prop       RMSE      MAE  RMSE/interval  spearman  pairwise
Egc       0.485    0.359            4.9     0.944     0.896
Egb       0.712    0.481            7.1     0.950     0.906
Eea       0.389    0.295            3.9     0.940     0.892
Ei        0.469    0.325            4.7     0.886     0.872
EPS       0.568    0.394            5.7     0.871     0.851
Nc        0.086    0.062            1.7     0.955     0.904
Xc       24.780   20.050           12.4     0.227     0.572
Eat       0.110    0.080            1.1     0.959     0.911