Skip to content

Latest commit

 

History

History
34 lines (23 loc) · 1.55 KB

README.md

File metadata and controls

34 lines (23 loc) · 1.55 KB
Raw

How to run the LIDC-IDRI experiment

cd experiments/lidc/

  1. Download and prepare data Download LIDC data from this link. Create data directory and move the downloaded data to ./data. Unzip data.

    mkdir ./data
mv <downloaded data> ./data
unzip <downloaded data>

  2. Run training scripts for segmentation, run

    python train_segmentation.py

    for classification, run

    python train_classification.py

The default model is ACS ResNet18 p.. To change ACSConv to Conv3d / Conv2_5d or random initialization, or to change to other backbones (DenseNet or VGG16), modify LIDCSegConfig and LIDCClassConfig in lidc_config.py, as instructed by the comments aside. For ResNet18, the pretraining methods of Conv3d include i3d, video and mednet. Note that to use pretrained models of video and mednet, you need to download and unzip checkpoints from these two links (video, mednet), and then copy the checkpoint file locations to the corresponding variables of LIDCEnv in lidc_config.py.