ailab-bio/PROTAC-Splitter-EncoderDecoder-lr_cosine-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: 0.3986
• Poi Has Attachment Point(s): 0.9602
• Num Fragments: 3.0004
• All Ligands Equal: 0.5961
• Linker Equal: 0.8502
• Heavy Atoms Difference: 4.5700
• E3 Has Attachment Point(s): 0.9887
• Reassembly Nostereo: 0.6338
• Poi Graph Edit Distance: inf
• E3 Valid: 0.9887
• E3 Heavy Atoms Difference: 0.1957
• Poi Valid: 0.9602
• Tanimoto Similarity: 0.0
• Linker Tanimoto Similarity: 0.0
• Linker Has Attachment Point(s): 0.9970
• E3 Tanimoto Similarity: 0.0
• Poi Heavy Atoms Difference: 1.1950
• Has All Attachment Points: 0.9904
• Linker Heavy Atoms Difference: 0.1979
• Poi Equal: 0.7910
• Linker Graph Edit Distance Norm: inf
• E3 Equal: 0.8273
• Poi Graph Edit Distance Norm: inf
• Poi Tanimoto Similarity: 0.0
• E3 Heavy Atoms Difference Norm: 0.0011
• E3 Graph Edit Distance: inf
• Heavy Atoms Difference Norm: 0.0620
• Linker Heavy Atoms Difference Norm: 0.0009
• Reassembly: 0.6025
• Has Three Substructures: 0.9995
• Linker Graph Edit Distance: 30099150141643059856874246094752676646564926231321411573514240.0000
• Linker Valid: 0.9970
• Valid: 0.9474
• E3 Graph Edit Distance Norm: inf
• Poi Heavy Atoms Difference Norm: 0.0361

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
• lr_scheduler_warmup_steps: 100
• 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
0.012	0.4932	5000	0.4824	0.7897	inf	inf	0.9972	0.1648	-0.0036	0.0	0.9972	0.9873	0.9995	4.9586	0.0650	0.6946	25672804532577903995569209904347871257364201785538851047997440.0000	inf	0.9974	0.5344	0.0103	0.0	0.9974	0.3055	3.0005	0.7354	inf	inf	0.9479	1.4326	0.0421	0.0	0.9479	0.4899	0.5198	0.0	0.9441
0.0064	0.7398	7500	0.5282	0.8017	inf	inf	0.9944	0.3429	0.0053	0.0	0.9944	0.9873	0.9985	4.5195	0.0593	0.7533	40722379603399433924006332951724209580646664901199556834754560.0000	0.0452	0.9959	0.2564	-0.0000	0.0	0.9959	0.3156	3.0003	0.7551	inf	inf	0.9532	1.3395	0.0396	0.0	0.9532	0.5359	0.5689	0.0	0.9477
0.0051	0.9864	10000	0.5390	0.8036	inf	inf	0.9966	0.2976	0.0033	0.0	0.9966	0.9905	0.9992	4.1049	0.0536	0.7668	28328611898016997512352231618590754490884636453008387363307520.0000	inf	0.9972	0.3031	0.0034	0.0	0.9972	0.3184	3.0003	0.7620	inf	inf	0.9589	1.1856	0.0352	0.0	0.9589	0.5459	0.5799	0.0	0.9547
0.0003	6.9047	70000	0.3939	0.9555	3.0004	0.5924	0.8415	4.5431	0.9915	0.6323	inf	0.9915	0.1461	0.9555	0.0	0.0	0.9978	0.0	1.2542	0.9910	0.1698	0.7887	inf	0.8258	inf	0.0	-0.0020	inf	0.0607	-0.0016	0.5999	0.9996	22131728045325776452029795540104264829431683329922529862090752.0000	0.9978	0.9464	inf	0.0349
0.0003	7.1513	72500	0.3921	0.9602	3.0002	0.5919	0.8427	4.5754	0.9887	0.6308	inf	0.9887	0.3155	0.9602	0.0	0.0	0.9963	0.0	1.1927	0.9888	0.2604	0.7898	inf	0.8252	inf	0.0	0.0047	inf	0.0611	0.0043	0.5992	0.9991	37181303116147309234962303999400365269286085344573508414341120.0000	0.9963	0.9479	inf	0.0355
0.0003	7.3979	75000	0.3946	0.9585	2.9998	0.5922	0.8430	4.6840	0.9876	0.6307	inf	0.9876	0.2445	0.9585	0.0	0.0	0.9972	0.0	1.2902	0.9894	0.2403	0.7896	inf	0.8269	inf	0.0	0.0029	inf	0.0631	0.0042	0.5991	0.9988	28328611898016997512352231618590754490884636453008387363307520.0000	0.9972	0.9464	inf	0.0394
0.0003	7.6445	77500	0.3915	0.9580	3.0004	0.5952	0.8483	4.5722	0.9893	0.6334	inf	0.9893	0.2443	0.9580	0.0	0.0	0.9977	0.0	1.2063	0.9888	0.2202	0.7890	inf	0.8266	inf	0.0	0.0019	inf	0.0619	0.0031	0.6016	0.9993	23016997167138810478786188190104988023843767118069314732687360.0000	0.9977	0.9475	inf	0.0368
0.0003	7.8911	80000	0.3949	0.9589	3.0004	0.5950	0.8472	4.5701	0.9879	0.6330	inf	0.9879	0.1097	0.9589	0.0	0.0	0.9973	0.0	1.2103	0.9905	0.1845	0.7888	0.0285	0.8258	inf	0.0	-0.0029	inf	0.0615	-0.0007	0.6022	0.9989	26558073654390935167830217142428832335204346674695363153100800.0000	0.9973	0.9466	inf	0.0350
0.0002	8.1377	82500	0.3974	0.9626	3.0002	0.5951	0.8476	4.1669	0.9908	0.6333	inf	0.9908	0.0201	0.9626	0.0	0.0	0.9973	0.0	1.1490	0.9909	0.1837	0.7907	inf	0.8269	inf	0.0	-0.0051	inf	0.0565	0.0004	0.6021	0.9995	inf	0.9973	0.9523	inf	0.0342
0.0002	8.3843	85000	0.3964	0.9615	3.0006	0.5960	0.8492	4.2838	0.9900	0.6346	inf	0.9900	0.1197	0.9615	0.0	0.0	0.9971	0.0	1.1409	0.9894	0.1703	0.7904	inf	0.8274	inf	0.0	-0.0009	inf	0.0581	-0.0010	0.6024	0.9994	29213881019830025830117853444751953452152842443174626702917632.0000	0.9971	0.9502	inf	0.0336
0.0002	8.6309	87500	0.3990	0.9604	3.0001	0.5953	0.8501	4.5262	0.9888	0.6329	inf	0.9888	0.1731	0.9604	0.0	0.0	0.9973	0.0	1.1703	0.9884	0.2064	0.7899	inf	0.8263	inf	0.0	0.0008	inf	0.0616	0.0013	0.6017	0.9996	inf	0.9973	0.9479	inf	0.0349
0.0002	8.8775	90000	0.3977	0.9608	3.0003	0.5956	0.8498	4.7022	0.9862	0.6330	inf	0.9862	0.2705	0.9608	0.0	0.0	0.9970	0.0	1.1535	0.9884	0.1989	0.7905	0.0278	0.8270	inf	0.0	0.0049	inf	0.0641	0.0014	0.6022	0.9996	30099150141643059856874246094752676646564926231321411573514240.0000	0.9970	0.9456	inf	0.0346
0.0002	9.1241	92500	0.3981	0.9606	3.0003	0.5966	0.8514	4.5609	0.9883	0.6340	inf	0.9883	0.2075	0.9606	0.0	0.0	0.9968	0.0	1.1666	0.9898	0.1784	0.7913	0.0278	0.8275	inf	0.0	0.0023	inf	0.0620	-0.0004	0.6030	0.9996	31869688385269122201396260570914598802245216009634435783720960.0000	0.9968	0.9471	inf	0.0352
0.0002	9.3707	95000	0.3984	0.9596	3.0004	0.5964	0.8510	4.6478	0.9884	0.6343	inf	0.9884	0.1992	0.9596	0.0	0.0	0.9965	0.0	1.2270	0.9898	0.2018	0.7906	inf	0.8270	inf	0.0	0.0013	inf	0.0630	0.0011	0.6030	0.9995	34525495750708218572674667697077244152337589576094244864524288.0000	0.9965	0.9464	inf	0.0375
0.0002	9.6173	97500	0.3987	0.9603	3.0004	0.5961	0.8502	4.5589	0.9888	0.6339	inf	0.9888	0.1958	0.9603	0.0	0.0	0.9969	0.0	1.1934	0.9904	0.1982	0.7909	inf	0.8273	inf	0.0	0.0010	inf	0.0619	0.0008	0.6026	0.9995	30984419263456093883630638744753399840977010019468196444110848.0000	0.9969	0.9476	inf	0.0361
0.0002	9.8639	100000	0.3986	0.9602	3.0004	0.5961	0.8502	4.5700	0.9887	0.6338	inf	0.9887	0.1957	0.9602	0.0	0.0	0.9970	0.0	1.1950	0.9904	0.1979	0.7910	inf	0.8273	inf	0.0	0.0011	inf	0.0620	0.0009	0.6025	0.9995	30099150141643059856874246094752676646564926231321411573514240.0000	0.9970	0.9474	inf	0.0361

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.1+cu121
• Datasets 3.0.0
• Tokenizers 0.19.1