Molecular Encoders
Collection
Collection of pre-trained encoder models trained on large molecules databases. • 3 items • Updated • 1
# Load model directly
from transformers import AutoTokenizer, AutoModelForMaskedLM
tokenizer = AutoTokenizer.from_pretrained("knowledgator/SMILES-DeBERTa-base")
model = AutoModelForMaskedLM.from_pretrained("knowledgator/SMILES-DeBERTa-base")Quick Links
SMILES-DeBERTa-base
SMILES-DeBERTa-base was designed to be used as an encoder of SMILES sequences in a general organic chemistry context.
Model Details
Model Description
SMILES-DeBERTa-base is based on the DeBERTa-V2 model with optimizations in implementing SMILES tokenizer for the encoder.
- Developed by: Knowladgator Engineering
- Language(s) (NLP): SMILES
- License: Apache License 2.0
Citation
Coming soon.
Model Card Authors
Model Card Contact
- Downloads last month
- 15
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("fill-mask", model="knowledgator/SMILES-DeBERTa-base")