ailab-bio/PROTAC-Splitter-EncoderDecoder-lr_cosine_restarts-rand-smiles

This model is a fine-tuned version of seyonec/ChemBERTa-zinc-base-v1 on the ailab-bio/PROTAC-Splitter-Dataset dataset. It achieves the following results on the evaluation set:

  • Loss: 0.3584
  • Linker Has Attachment Point(s): 0.9983
  • E3 Valid: 0.9934
  • E3 Graph Edit Distance Norm: inf
  • E3 Has Attachment Point(s): 0.9934
  • E3 Heavy Atoms Difference: 0.1883
  • Poi Valid: 0.9567
  • Poi Tanimoto Similarity: 0.0
  • Has All Attachment Points: 0.9939
  • Num Fragments: 2.9997
  • Reassembly: 0.6015
  • Poi Equal: 0.7927
  • Poi Graph Edit Distance: inf
  • Linker Graph Edit Distance Norm: inf
  • Linker Valid: 0.9983
  • Linker Heavy Atoms Difference Norm: 0.0036
  • Poi Has Attachment Point(s): 0.9567
  • Valid: 0.9493
  • Poi Heavy Atoms Difference Norm: 0.0441
  • Tanimoto Similarity: 0.0
  • Linker Tanimoto Similarity: 0.0
  • Linker Graph Edit Distance: inf
  • Linker Equal: 0.8512
  • Heavy Atoms Difference Norm: 0.0570
  • E3 Heavy Atoms Difference Norm: 0.0026
  • Heavy Atoms Difference: 4.2382
  • Poi Graph Edit Distance Norm: inf
  • Linker Heavy Atoms Difference: 0.2012
  • Reassembly Nostereo: 0.6357
  • E3 Graph Edit Distance: inf
  • E3 Tanimoto Similarity: 0.0
  • Has Three Substructures: 0.9997
  • E3 Equal: 0.8282
  • Poi Heavy Atoms Difference: 1.3882
  • All Ligands Equal: 0.5946

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 5e-05
  • train_batch_size: 128
  • eval_batch_size: 64
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: cosine_with_restarts
  • lr_scheduler_warmup_steps: 800
  • training_steps: 100000
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step All Ligands Equal E3 Equal E3 Graph Edit Distance E3 Graph Edit Distance Norm E3 Has Attachment Point(s) E3 Heavy Atoms Difference E3 Heavy Atoms Difference Norm E3 Tanimoto Similarity E3 Valid Has All Attachment Points Has Three Substructures Heavy Atoms Difference Heavy Atoms Difference Norm Linker Equal Linker Graph Edit Distance Linker Graph Edit Distance Norm Linker Has Attachment Point(s) Linker Heavy Atoms Difference Linker Heavy Atoms Difference Norm Linker Tanimoto Similarity Linker Valid Validation Loss Num Fragments Poi Equal Poi Graph Edit Distance Poi Graph Edit Distance Norm Poi Has Attachment Point(s) Poi Heavy Atoms Difference Poi Heavy Atoms Difference Norm Poi Tanimoto Similarity Poi Valid Reassembly Reassembly Nostereo Tanimoto Similarity Valid
0.0212 0.4932 5000 0.4396 0.7734 inf inf 0.9979 0.1237 -0.0043 0.0 0.9979 0.9903 0.9994 4.6065 0.0591 0.6416 24787535410764872823308202666266910179524056896382338942894080.0000 inf 0.9975 0.5099 0.0050 0.0 0.9975 0.2852 3.0001 0.7208 inf inf 0.9528 1.4706 0.0414 0.0 0.9528 0.4464 0.4783 0.0 0.9500
0.011 0.7398 7500 0.4827 0.7827 inf inf 0.9905 0.5439 0.0117 0.0 0.9905 0.9858 0.9990 6.6114 0.0846 0.7075 27443342776203963485595838968590031296472552664861602492710912.0000 0.0511 0.9973 0.3255 -0.0008 0.0 0.9973 0.3022 3.0001 0.7332 inf inf 0.9389 2.2561 0.0673 0.0 0.9389 0.4896 0.5176 0.0 0.9294
0.007 0.9864 10000 0.5318 0.8017 inf inf 0.9922 0.2046 -0.0014 0.0 0.9922 0.9900 0.9991 4.5184 0.0596 0.7645 inf inf 0.9980 0.3372 0.0044 0.0 0.9980 0.3093 2.9998 0.7580 inf inf 0.9534 1.3755 0.0422 0.0 0.9534 0.5402 0.5728 0.0 0.9458
0.0005 6.9047 70000 0.3363 0.9977 0.9890 inf 0.9890 0.3207 0.9573 0.0 0.9891 2.9997 0.5983 0.7900 inf inf 0.9977 0.0009 0.9573 0.9452 0.0404 0.0 0.0 23016997167138810478786188190104988023843767118069314732687360.0000 0.8439 0.0615 0.0015 4.7337 inf 0.2010 0.6325 inf 0.0 0.9996 0.8273 1.3261 0.5905
0.0005 7.1513 72500 0.3347 0.9974 0.9887 inf 0.9887 0.3786 0.9609 0.0 0.9926 3.0003 0.5987 0.7891 inf inf 0.9974 0.0033 0.9609 0.9487 0.0401 0.0 0.0 25672804532577903995569209904347871257364201785538851047997440.0000 0.8407 0.0588 0.0044 4.5035 inf 0.2400 0.6343 inf 0.0 0.9994 0.8260 1.3144 0.5911
0.0008 7.3979 75000 0.3530 0.9965 0.9897 inf 0.9897 0.2380 0.9452 0.0 0.9885 2.9999 0.5951 0.7843 inf inf 0.9965 0.0028 0.9452 0.9335 0.0534 0.0 0.0 inf 0.8362 0.0716 -0.0004 5.4047 inf 0.2346 0.6308 inf 0.0 0.9992 0.8239 1.6527 0.5876
0.0016 7.6445 77500 0.3554 0.9958 0.9887 inf 0.9887 0.4638 0.9404 0.0 0.9874 3.0001 0.5911 0.7801 inf inf 0.9958 0.0081 0.9404 0.9281 0.0595 0.0 0.0 41607648725212467950762725601724932775058748689346341705351168.0000 0.8252 0.0795 0.0085 5.9789 inf 0.3348 0.6265 inf 0.0 0.9988 0.8217 1.7751 0.5832
0.0004 7.8911 80000 0.3551 0.9988 0.9888 inf 0.9888 0.4881 0.9533 0.0 0.9912 2.9999 0.6013 0.7877 inf inf 0.9988 -0.0009 0.9533 0.9417 0.0421 0.0 0.0 inf 0.8503 0.0657 0.0099 5.0296 inf 0.1533 0.6381 inf 0.0 0.9997 0.8263 1.4026 0.5946
0.0005 8.1377 82500 0.3597 0.9981 0.9889 inf 0.9889 0.1790 0.9525 0.0 0.9868 3.0002 0.6001 0.7882 inf inf 0.9981 0.0018 0.9525 0.9414 0.0426 0.0 0.0 19475920679886688644237544649700905829055126460433539077767168.0000 0.8470 0.0655 0.0028 4.8411 inf 0.1896 0.6345 inf 0.0 0.9995 0.8283 1.3738 0.5931
0.0006 8.3843 85000 0.3532 0.9969 0.9912 inf 0.9912 0.1049 0.9551 0.0 0.9896 2.9999 0.5975 0.7880 inf inf 0.9969 0.0045 0.9551 0.9444 0.0424 0.0 0.0 inf 0.8409 0.0589 -0.0023 4.4111 inf 0.2537 0.6296 inf 0.0 0.9997 0.8255 1.3678 0.5902
0.0015 8.6309 87500 0.3422 0.9975 0.9941 inf 0.9941 0.1578 0.9388 0.0 0.9927 2.9999 0.5950 0.7813 inf inf 0.9975 0.0093 0.9388 0.9317 0.0561 0.0 0.0 24787535410764872823308202666266910179524056896382338942894080.0000 0.8415 0.0740 -0.0025 5.6405 inf 0.3237 0.6294 inf 0.0 0.9999 0.8237 1.7777 0.5883
0.0004 8.8775 90000 0.3554 0.9982 0.9916 inf 0.9916 0.1501 0.9603 0.0 0.9942 2.9996 0.6016 0.7898 inf inf 0.9982 0.0017 0.9603 0.9511 0.0345 0.0 0.0 17705382436260626299715530173538983673374836682120514867560448.0000 0.8493 0.0532 -0.0026 4.1205 inf 0.1871 0.6375 inf 0.0 0.9996 0.8269 1.1763 0.5943
0.0004 9.1241 92500 0.3515 0.9974 0.9851 inf 0.9851 0.4341 0.9510 0.0 0.9898 2.9996 0.6034 0.7930 inf inf 0.9974 0.0031 0.9510 0.9348 0.0433 0.0 0.0 inf 0.8492 0.0691 0.0076 5.2700 inf 0.2049 0.6379 inf 0.0 0.9996 0.8267 1.4062 0.5962
0.0006 9.3707 95000 0.3472 0.9978 0.9886 inf 0.9886 0.3374 0.9533 0.0 0.9892 2.9997 0.5980 0.7876 inf inf 0.9978 0.0030 0.9533 0.9408 0.0424 0.0 0.0 inf 0.8475 0.0656 0.0104 4.8067 inf 0.2247 0.6291 inf 0.0 0.9997 0.8252 1.2974 0.5910
0.0009 9.6173 97500 0.3462 0.9970 0.9955 inf 0.9955 0.2014 0.9459 0.0 0.9920 2.9999 0.5945 0.7863 inf inf 0.9970 -0.0004 0.9459 0.9390 0.0480 0.0 0.0 30099150141643059856874246094752676646564926231321411573514240.0000 0.8381 0.0651 0.0015 4.9979 inf 0.1987 0.6289 inf 0.0 0.9999 0.8247 1.5948 0.5873
0.0004 9.8639 100000 0.3584 0.9983 0.9934 inf 0.9934 0.1883 0.9567 0.0 0.9939 2.9997 0.6015 0.7927 inf inf 0.9983 0.0036 0.9567 0.9493 0.0441 0.0 0.0 inf 0.8512 0.0570 0.0026 4.2382 inf 0.2012 0.6357 inf 0.0 0.9997 0.8282 1.3882 0.5946

Framework versions

  • Transformers 4.44.2
  • Pytorch 2.4.1+cu121
  • Datasets 3.0.0
  • Tokenizers 0.19.1
Downloads last month
6
Safetensors
Model size
0.1B params
Tensor type
F32
ยท
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support

Model tree for ailab-bio/PROTAC-Splitter-EncoderDecoder-lr_cosine_restarts-rand-smiles

Finetuned
(20)
this model

Evaluation results