---
datasets:
- andrewt28/MolLM-dataset
---
MolLM: A Unified Language Model to Integrate Biomedical Text with 2D and 3D Molecular Representations

Model checkpoints for [MolLM model](https://github.com/gersteinlab/MolLM).

The checkpoints for each downstream task can be found as follows (along with a summary of the contents of the top-level of each archive):
- `GraphTextRetrieval`
    - `GraphTextRetrieval-model.zip`
        - `bert_pretrained`
        - `all_checkpoints`
        - `data`
        - `finetune_save`
- `MoleculeCaption`
    - `MoleculeCaption-model.zip`
        - `data`
        - `text2mol-data`
        - `M3_checkpoints`
        - `scibert`
        - `molt5-base-smiles2caption`
        - `molt5-small-smiles2caption`
        - `molt5-large-smiles2caption`
- `MoleculeEditing`
    - `MoleculeEditing-model.zip`
        - `bert_pretrained`
        - `checkpoints`
        - `embedding_data`
        - `model-epoch=507.ckpt`
- `MoleculeGeneration`
    - `MoleculeGeneration-model.zip`
        - `results`
        - `model-epoch=465.ckpt`
        - `qm9_64gnn_128-64lin_1-1mask_0d6noise_convlu1`
- `MoleculePrediction`
    - `MoleculePrediction-model.zip`
        - `all_checkpoints`
        - `dataset`
        - `model-epoch=507.ckpt`