Typhon Experiments

This repository contains the code of our new Typhon framework as well as the code for our experiments.

To run the code, a Conda enviroment must first be setup using

conda env create -f environment.yml

after cloning this repository.

One can then activate the environment using

conda activate typhon

The example experiment can finally be run using the following command:

python3 experiments/prostate_0.py

Hyperparameters and many other options can be modified in the aformentioned file.

N.B. that you can do a shorter run by simply adding your OS name by going into the experiment.py file at line 86.

Name		Name	Last commit message	Last commit date
Latest commit History 115 Commits
DATASET_BUS_png		DATASET_BUS_png
Small		Small
Small_npy		Small_npy
architectures		architectures
datasets_autoencoding		datasets_autoencoding
dsets_preparation		dsets_preparation
experiments		experiments
experiments_atari		experiments_atari
helpers		helpers
old_files_20230421		old_files_20230421
results_atari_140		results_atari_140
.gitignore		.gitignore
LICENCE.txt		LICENCE.txt
Metrics_multiple_exp.ipynb		Metrics_multiple_exp.ipynb
Pipfile		Pipfile
Plot_red_mask.ipynb		Plot_red_mask.ipynb
Plot_with_confidence_intervall.ipynb		Plot_with_confidence_intervall.ipynb
Plot_with_confidence_intervall_baseline.ipynb		Plot_with_confidence_intervall_baseline.ipynb
README.md		README.md
adapt_bootstrap.ipynb		adapt_bootstrap.ipynb
architecture_loader.py		architecture_loader.py
brg_rgb_converter.ipynb		brg_rgb_converter.ipynb
environment.yml		environment.yml
experiment.py		experiment.py
plot_detected_features.ipynb		plot_detected_features.ipynb
plot_dms.ipynb		plot_dms.ipynb
plot_segmentation_example.ipynb		plot_segmentation_example.ipynb
results-Copy1.ipynb		results-Copy1.ipynb
results.ipynb		results.ipynb
results_atari.ipynb		results_atari.ipynb
test.ipynb		test.ipynb
test.py		test.py
test_mode.ipynb		test_mode.ipynb
training_plus_spec.ipynb		training_plus_spec.ipynb
typhon.py		typhon.py
typhon_model.py		typhon_model.py
utils.py		utils.py

Provide feedback