Dynamical Model Extension of the Supirfactor Model
Project description
supirfactor-dynamical
This is a PyTorch model package for creating dynamical, biophysical models of transcriptional output and regulation.
Installation
Install this package using the standard python package manager python -m pip install supirfactor_dynamical
.
It depends on PyTorch and the standard python scientific computing
packages (e.g. scipy, numpy, pandas).
Usage
from supirfactor_dynamical import (
SupirFactorBiophysical
)
# Construct model object
model = SupirFactorBiophysical(
prior_network, # Prior knowledge connectivity network [Genes x TFs]
output_activation='softplus' # Use softplus activation for transcriptional model output
)
# Set prediction parameter
model.set_time_parameters(
n_additional_predictions=10 # Make forward predictions in time during training
)
# Train model
model.train_model(
training_dataloader, # Training data in a torch DataLoader
500 # Epochs
)
# Save model
model.save("supirfactor_dynamical.h5")
Examples containing data loading, hyperparameter searching, and result testing are located in ./scripts/
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Close
Hashes for supirfactor_dynamical-1.0.0.tar.gz
Algorithm | Hash digest | |
---|---|---|
SHA256 | f215184fa406f4251135404d75f418f3d7aab72c431b29b423dc75b57bcd9a28 |
|
MD5 | 23b64feabfb3bc11fe1be1cb328fe705 |
|
BLAKE2b-256 | 6546a698b54da42b57d399a3c1c173dbdea40a90e2efcd004c521de5cd7a4b13 |
Close
Hashes for supirfactor_dynamical-1.0.0-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 60b46fcb57d41b20335e0b1848a4829996c18351c99492815209a124c357838f |
|
MD5 | b8a08dbf28d87c8d74eb9d8000edbd77 |
|
BLAKE2b-256 | e42dbc497d7ccee20c1bd6e87f14adc7763f614c9b6ac67341c9c84cfcbd0b65 |