ailab-bio/PROTAC-Splitter-Trial-11-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: 0.0002
• Linker Equal: 0.9887
• Num Fragments: 3.0
• E3 Heavy Atoms Difference: 0.0042
• Poi Heavy Atoms Difference Norm: 0.0007
• Reassembly: 0.9995
• Linker Tanimoto Similarity: 0.0
• E3 Tanimoto Similarity: 0.0
• E3 Has Attachment Point(s): 1.0
• E3 Graph Edit Distance: inf
• Poi Equal: 0.9933
• All Ligands Equal: 0.9884
• Poi Valid: 1.0
• Tanimoto Similarity: 0.0
• Poi Heavy Atoms Difference: 0.0225
• E3 Graph Edit Distance Norm: inf
• E3 Equal: 0.9948
• Linker Heavy Atoms Difference Norm: -0.0033
• Poi Graph Edit Distance: inf
• Linker Has Attachment Point(s): 1.0
• E3 Valid: 1.0
• Linker Valid: 1.0
• Heavy Atoms Difference Norm: -0.0000
• Has Three Substructures: 1.0
• Has All Attachment Points: 1.0
• Poi Graph Edit Distance Norm: inf
• Poi Has Attachment Point(s): 1.0
• Linker Graph Edit Distance Norm: inf
• E3 Heavy Atoms Difference Norm: 0.0002
• Reassembly Nostereo: 0.9995
• Valid: 1.0
• Heavy Atoms Difference: -0.0017
• Linker Heavy Atoms Difference: -0.0284
• Linker Graph Edit Distance: inf
• Poi Tanimoto Similarity: 0.0

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: 500000
• 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
0.0006	6.5543	70000	0.9869	0.9939	inf	inf	1.0	0.0065	0.0003	0.0	1.0	1.0	1.0	-0.0006	-0.0000	0.9875	0.1116	inf	1.0	-0.0244	-0.0027	0.0	1.0	0.0003	3.0	0.9932	inf	inf	1.0	0.0174	0.0005	0.0	1.0	0.9986	0.9989	0.0	1.0
0.0005	7.4906	80000	0.9872	0.9948	inf	inf	1.0	0.0057	0.0003	0.0	1.0	1.0	1.0	0.0021	0.0000	0.9875	0.1162	inf	1.0	-0.0149	-0.0025	0.0	1.0	0.0003	3.0	0.9931	inf	inf	1.0	0.0113	0.0002	0.0	1.0	0.9988	0.9989	0.0	1.0
0.0004	8.4270	90000	0.9875	0.9944	inf	inf	1.0	0.0049	0.0002	0.0	1.0	1.0	1.0	-0.0007	-0.0000	0.9880	inf	inf	1.0	-0.0115	-0.0024	0.0	1.0	0.0003	3.0	0.9932	inf	inf	1.0	0.0059	-0.0000	0.0	1.0	0.9992	0.9992	0.0	1.0
0.0004	9.3633	100000	0.9876	0.9950	inf	inf	1.0	0.0045	0.0002	0.0	1.0	1.0	1.0	0.0004	0.0000	0.9877	0.1182	inf	1.0	-0.0237	-0.0027	0.0	1.0	0.0002	3.0	0.9933	inf	inf	1.0	0.0196	0.0006	0.0	1.0	0.9992	0.9993	0.0	1.0
0.0004	16.3858	350000	0.0002	0.9882	3.0	0.0044	0.0003	0.9991	0.0	0.0	1.0	inf	0.9926	0.9877	1.0	0.0	0.0138	inf	0.9953	-0.0029	inf	1.0	1.0	1.0	-0.0000	1.0	1.0	inf	1.0	inf	0.0002	0.9991	1.0	-0.0012	-0.0194	inf	0.0
0.0004	16.8539	360000	0.0002	0.9880	3.0	0.0052	0.0008	0.9994	0.0	0.0	1.0	inf	0.9931	0.9878	1.0	0.0	0.0243	inf	0.9949	-0.0032	inf	1.0	1.0	1.0	-0.0000	1.0	1.0	inf	1.0	inf	0.0002	0.9994	1.0	-0.0006	-0.0301	0.1106	0.0
0.0004	17.3221	370000	0.0002	0.9882	3.0	0.0042	0.0005	0.9996	0.0	0.0	1.0	inf	0.9928	0.9878	1.0	0.0	0.0181	inf	0.9946	-0.0031	inf	1.0	1.0	1.0	0.0001	1.0	1.0	inf	1.0	inf	0.0002	0.9996	1.0	0.0049	-0.0232	0.1202	0.0
0.0003	17.7903	380000	0.0002	0.9882	3.0	0.0037	0.0008	0.9994	0.0	0.0	1.0	inf	0.9931	0.9880	1.0	0.0	0.0244	inf	0.9949	-0.0029	inf	1.0	1.0	1.0	0.0000	1.0	1.0	inf	1.0	inf	0.0002	0.9994	1.0	0.0002	-0.0279	0.1036	0.0
0.0003	18.2584	390000	0.0002	0.9876	3.0	0.0046	0.0006	0.9994	0.0	0.0	1.0	inf	0.9928	0.9874	1.0	0.0	0.0217	inf	0.9947	-0.0035	inf	1.0	1.0	1.0	-0.0000	1.0	1.0	inf	1.0	inf	0.0002	0.9995	1.0	-0.0006	-0.0268	inf	0.0
0.0004	18.7266	400000	0.0003	0.9877	3.0	0.0065	0.0007	0.9991	0.0	0.0	1.0	0.0390	0.9929	0.9875	1.0	0.0	0.0237	inf	0.9948	-0.0035	inf	0.9999	1.0	0.9999	0.0001	1.0	1.0	inf	1.0	0.0023	0.0003	0.9992	0.9999	0.0068	-0.0278	716948666475480386420205582353984848982468782380308030816256.0000	0.0
0.0004	19.1948	410000	0.0002	0.9872	3.0	0.0039	0.0004	0.9994	0.0	0.0	1.0	inf	0.9923	0.9870	1.0	0.0	0.0177	inf	0.9947	-0.0032	inf	1.0	1.0	1.0	-0.0000	1.0	1.0	inf	1.0	inf	0.0002	0.9995	1.0	-0.0018	-0.0234	inf	0.0
0.0003	19.6629	420000	0.0002	0.9879	3.0	0.0022	0.0007	0.9995	0.0	0.0	1.0	inf	0.9930	0.9877	1.0	0.0	0.0234	inf	0.9947	-0.0030	inf	1.0	1.0	1.0	0.0001	1.0	1.0	inf	1.0	inf	0.0001	0.9996	1.0	0.0054	-0.0258	0.1158	0.0
0.0003	20.1311	430000	0.0002	0.9885	3.0	0.0056	-0.0001	0.9995	0.0	0.0	1.0	inf	0.9933	0.9884	1.0	0.0	0.0049	inf	0.9951	-0.0021	inf	1.0	1.0	1.0	0.0000	1.0	1.0	inf	1.0	inf	0.0003	0.9995	1.0	0.0002	-0.0103	inf	0.0
0.0003	20.5993	440000	0.0002	0.9888	3.0	0.0042	0.0008	0.9996	0.0	0.0	1.0	inf	0.9939	0.9887	1.0	0.0	0.0253	inf	0.9950	-0.0033	inf	1.0	1.0	1.0	-0.0000	1.0	1.0	inf	1.0	inf	0.0002	0.9996	1.0	-0.0006	-0.0300	0.1093	0.0
0.0003	21.0674	450000	0.0002	0.9881	3.0	0.0052	0.0007	0.9995	0.0	0.0	1.0	inf	0.9931	0.9880	1.0	0.0	0.0218	inf	0.9950	-0.0030	inf	1.0	1.0	1.0	-0.0000	1.0	1.0	inf	1.0	inf	0.0002	0.9995	1.0	-0.0010	-0.0280	inf	0.0
0.0004	21.5356	460000	0.0002	0.9885	3.0	0.0026	0.0006	0.9994	0.0	0.0	1.0	inf	0.9936	0.9883	1.0	0.0	0.0201	inf	0.9949	-0.0029	inf	1.0	1.0	1.0	-0.0000	1.0	1.0	inf	1.0	inf	0.0001	0.9994	1.0	-0.0009	-0.0236	0.1040	0.0
0.0003	22.0037	470000	0.0002	0.9889	3.0	0.0050	0.0006	0.9996	0.0	0.0	1.0	inf	0.9939	0.9888	1.0	0.0	0.0207	inf	0.9951	-0.0029	inf	1.0	1.0	1.0	-0.0000	1.0	1.0	inf	1.0	inf	0.0002	0.9996	1.0	-0.0009	-0.0266	inf	0.0
0.0003	22.4719	480000	0.0002	0.9875	3.0	0.0070	0.0000	0.9991	0.0	0.0	0.9999	inf	0.9925	0.9870	1.0	0.0	0.0075	inf	0.9948	-0.0027	inf	1.0	0.9999	1.0	0.0001	1.0	1.0	inf	1.0	inf	0.0003	0.9991	0.9999	0.0095	-0.0140	0.1202	0.0
0.0003	22.9401	490000	0.0002	0.9892	3.0	0.0040	0.0006	0.9994	0.0	0.0	1.0	inf	0.9938	0.9889	1.0	0.0	0.0206	inf	0.9951	-0.0029	inf	1.0	1.0	1.0	-0.0000	1.0	1.0	inf	1.0	inf	0.0002	0.9994	1.0	-0.0001	-0.0247	inf	0.0
0.0003	23.4082	500000	0.0002	0.9887	3.0	0.0042	0.0007	0.9995	0.0	0.0	1.0	inf	0.9933	0.9884	1.0	0.0	0.0225	inf	0.9948	-0.0033	inf	1.0	1.0	1.0	-0.0000	1.0	1.0	inf	1.0	inf	0.0002	0.9995	1.0	-0.0017	-0.0284	inf	0.0

Framework versions

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