Skip to content

Latest commit

 

History

History
46 lines (37 loc) · 2.3 KB

README.md

File metadata and controls

46 lines (37 loc) · 2.3 KB

DeepAL modification to study Head and Neck cancer

Description

This is a modification of DeepAL, the original project can be found here.

This work is focused on the study of Head and Neck cancer. The file VAE_HeadAndNeck.ipynb is included, made in Google Colab, which details the reduction of a dataset with dimensionality 564 x 60498.

Other necessary files

There are files that must be downloaded for trying some experiments:

  1. This one is needed to run the VAE_HeadAndNeck.ipynb file. It is a dataset with negative (non-cancer) samples. The size is 13.9MB.
  2. This dataset is also needed for the VAE_HeadAndNeck.ipynb file. It is a dataset with positive (Head and Neck cancer) samples. The size is 162MB.
  3. This is the reduced dataset after using the Variational Autoencoder. It is necessary for executing Deep Active Learning experiments.
  4. This is an artificial dataset, with the same dimensionality as the previous one. The values are randomly generated. It can be used to check how the algorithm behaves.

Install

DeepAL can be tried in your computer or uploading the files to Google Drive and then using Google Colab to run the demo. If using your computer, you will need these prerequisites:

  • numpy 1.21.2
  • scipy 1.7.1
  • pytorch 1.10.0
  • torchvision 0.11.1
  • scikit-learn 1.0.1
  • tqdm 4.62.3
  • ipdb 0.13.9

You can also install a conda environment: conda env create -f environment.yml

Demo

To try with the Head and Neck cancer dataset:

python demo.py --dataset_name HeadAndNeck --n_init_labeled 10 --n_query 6 --n_round 10 --strategy_name RandomSampling

Citing

Please do not forget to cite the original project, this is just a modification to study a concrete case:

@article{Huang2021deepal,
    author    = {Kuan-Hao Huang},
    title     = {DeepAL: Deep Active Learning in Python},
    journal   = {arXiv preprint arXiv:2111.15258},
    year      = {2021},
}