Skip to content

Latest commit

 

History

History
36 lines (24 loc) · 792 Bytes

README.md

File metadata and controls

36 lines (24 loc) · 792 Bytes
Raw

Recognizing off-sample mass spectrometry images with Convolutional Neural Networks

Requirements

  • Ubuntu >= 14.04
  • Conda >= 4.5.11
  • Ideally Nvidia GPU

Setup

  • Clone repository
  • Create conda environment
cd CNN
conda env create -n offsample-fastai --file fastai-env.yml
conda activate offsample-fastai
  • If Jupyter is already installed add a new kernel
python -m ipykernel install --user --name offsample-fastai --display-name "offsample-fastai"
  • Otherwise install Jupyter into the environment.

Run

  • Start Jupyter and open train_fastai_on_gs.ipynb

Pretrained Model

A model pretrained with fastai library is provided on the release page under model.fai name