Welcome to DeepMTP’s documentation!
DeepMTP is a Python package implementing a flexible two-branch neural network architecture that makes it compatible with the majority of multi-target prediction problems.
Documentation
- Release:
0.0.24.dev21+g9f9c19a65
- Date:
Aug 03, 2026
Contents:
- Installation
- Migrating from DeepMTP 0.0.22 to 0.0.23
- Loading a dataset
- Configuration options
- Architecture visualization
- Reproducible checkpoints
- Weights & Biases tracking
- Multiclass classification
- Mixed tabular inputs
- Sparse high-dimensional inputs
- Token-sequence inputs
- Foundation-model sequence encoders
- Supported model contract
- Running reference compatibility checks
- Tokenizing raw values
- Length policy
- Configuring the branch
- Pooling
- Training modes
- LoRA configuration
- Caching frozen entity representations
- Compact adapter checkpoints
- Composite inputs
- Reproducibility and checkpoints
- Security and offline use
- Molecular graph inputs
- Composite branch inputs
- Hyperparameter Optimization
- Demo notebooks
- Project policies
- API reference
- Credits