ailab-bio/PROTAC-Splitter-EncoderDecoder-lr_cosine-rand-smiles
016
<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
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
- trainbatchsize: 128
- evalbatchsize: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lrschedulertype: cosine
- lrschedulerwarmup_steps: 100
- training_steps: 100000
- mixedprecisiontraining: Native AMP
Training results
Framework versions
- Transformers 4.44.2
- Pytorch 2.4.1+cu121
- Datasets 3.0.0
- Tokenizers 0.19.1
