Published March 10, 2022
| Version v1
Model
Open
Unified Deep Learning Model for Multitask Reaction Predictions with Explanation
Creators
Description
The pretrained model weights and dataset used in T5Chem
Pre-trainied Models
All pre-trained models are trained on 97 million PubChem Molecules with BERT-like self-supervised mask-filling scheme. They will need to be loaded as "--pretrain" models for T5Chem and fine-tuned for any down-streaming tasks. A ready-to-go fine-tuned model for multi-task training (on USPTO_500_MT) is also available.
- simple_pretrain.tar.bz2: Character-level pubchem pretrained model
- USPTO_MT_model.tar.bz2: Multi-task USPTO_500_MT (Trained)
Note: For simple_pretrain.tar.bz2, finetune is REQUIRED!
Datasets
| Name | Size | Task type |
|---|
| sample | ~60,000 (1.9 MB) | Multi-task (small sample dataset) |
| USPTO_TPL | 445,115 (9.1 MB) | Classification (Reaction Type) |
| USPTO_MIT | 479,035 (18.4 MB) | Forward Prediction |
| USPTO_50k | 50,037 (892 KB) | Retrosynthesis |
| C-N coupling | 3,955 (774 KB) | Regression (Reaction Yield) |
| USPTO_500_MT | 143,535 (54.5 MB) | Multi-task |
Files
Files
(201.6 MB)
| Name | Size | Download all |
|---|---|---|
|
md5:41abdeb4e6438146f8dbeed77229b631
|
792.2 kB | Download |
|
md5:5448ea681c715cbc83b7651cbf059944
|
2.0 MB | Download |
|
md5:0c792d4cb198e433291d9b8a22c3ab21
|
55.9 MB | Download |
|
md5:a46765a8b16965961c4d8f06eb0a92fd
|
57.3 MB | Download |
|
md5:44a5f3ae08fe55933404c9398be22f5b
|
913.1 kB | Download |
|
md5:7f263ce46fc2272271f35d20c87980b1
|
19.3 MB | Download |
|
md5:124c50a1fb7a7d55582dcfc11b3223bd
|
55.9 MB | Download |
|
md5:8c5b0df7355b3a68b1a0bb8ccbbde897
|
9.5 MB | Download |
Additional details
Software
- Repository URL
- https://github.com/HelloJocelynLu/t5chem
- Programming language
- Python
- Development Status
- Active