Machine Learning Research Wizard
Project description
MLWiz: the Machine Learning Research Wizard
Documentation
MLWiz is a Python library that fosters machine learning research by reducing the boilerplate code to run reproducible experiments. It provides automatic management of data splitting, loading and common experimental settings. It especially handles both model selection and risk assessment procedures, by trying many different configurations in parallel (CPU or GPU). It is a generalized version of PyDGN that can handle different kinds of data and models (vectors, images, time-series, graphs).
Installation:
Requires at least Python 3.10. Simply run
pip install mlwiz
Quickstart:
Build dataset and data splits
mlwiz-data --config-file examples/DATA_CONFIGS/config_NCI1.yml [--debug]
Launch experiments
mlwiz-exp --config-file examples/MODEL_CONFIGS/config_SupToyDGN.yml [--debug]
Stop experiments
Use CTRL-C
, then type ray stop --force
to stop all ray processes you have launched.
Using the Trained Models
It's very easy to load the model from the experiments (see also the Tutorial):
from mlwiz.evaluation.util import *
config = retrieve_best_configuration('RESULTS/supervised_grid_search_toy_NCI1/MODEL_ASSESSMENT/OUTER_FOLD_1/MODEL_SELECTION/')
splits_filepath = 'examples/DATA_SPLITS/CHEMICAL/NCI1/NCI1_outer10_inner1.splits'
device = 'cpu'
# instantiate dataset
dataset = instantiate_dataset_from_config(config)
# instantiate model
model = instantiate_model_from_config(config, dataset, config_type="supervised_config")
# load model's checkpoint, assuming the best configuration has been loaded
checkpoint_location = 'RESULTS/supervised_grid_search_toy_NCI1/MODEL_ASSESSMENT/OUTER_FOLD_1/final_run1/best_checkpoint.pth'
load_checkpoint(checkpoint_location, model, device=device)
# you can now call the forward method of your model
y, embeddings = model(dataset[0])
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.