sjSDM-py 0.0.3

jSDM package

s-jSDM - Fast and accurate Joint Species Distribution Modeling

About the method

The method is described in the preprint Pichler & Hartig (2020) A new method for faster and more accurate inference of species associations from novel community data, https://arxiv.org/abs/2003.05331. The code for producing the results in this paper is available under the subfolder publications in this repo.

The method itself is wrapped into an R package, available under subfolder sjSDM. You can also use it stand-alone under Python (see instructions below). Note: for both the R and the python package, python >= 3.6 and pytorch must be installed (more details below).

Install instructions

Dependencies:

• PyTorch >= 1.4, see PyTorch for install instructions.

pip install sjSDM_py

Example

import sjSDM_py as fa
import numpy as np
import torch
Env = np.random.randn(100, 5)
Occ = np.random.binomial(1, 0.5, [100, 10])

model = fa.Model_sjSDM(device=torch.device("cpu"), dtype=torch.float32)
model.add_env(5, 10)
model.build(5, optimizer=fa.optimizer_adamax(0.001),scheduler=False)
model.fit(Env, Occ, batch_size = 20, epochs = 100)
# print(model.weights)
# print(model.covariance)