Giter Site home page Giter Site logo

segment-vasculature-5th-place's Introduction

Segment Vasculature 5th place solution

Gold solution for the Hacking the Human Vasculature in 3D competition.

The trained weights could be found in weights dir.

The proposed solution could be trained in about a week on a single RTX A6000 Ada.

  1. Install packages
pip install --upgrade pip
pip install -r requirements.dev.txt
pip install -r requirements.txt
  1. Download Kaggle data and place it into data/kaggle/.

  2. Download external kidney data + spleen data in 50um resolution, and place them into data/external

  3. Preprocess Kaggle + external data.

python segment_vasculature/preprocessing/create_3d_tensors.py
  1. Run MLFlow server
bash bash_scripts/run_mlflow_server.sh
  1. Export the project dir to enable relative imports
export PYTHONPATH="${PYTHONPATH}:${ABSOLUTE_PROJECT_PATH}"
  1. Export number of GPUs that are going to be used for training
export N_GPUS=1
  1. Train effnet_v2_m model for kidney1
bash bash_scripts/train_effnet_v2_m_kidney1.sh
  1. Calculate pseudo labels for kidney_2
bash bash_scripts/pseudo_label_kidney2.sh

Note: don't forget to insert correct path to weights in configs/callbacks/test.yaml

  1. Train effnet_v2_m model for kidney1 + kidney2
bash bash_scripts/train_effnet_v2_m_kidney1_2.sh
  1. Calculate pseudo labels for kidney_external
bash bash_scripts/pseudo_label_kidney_external.sh

Note: don't forget to insert correct path to weights in configs/callbacks/test.yaml

  1. Train effnet_v2_m model for kidney1 + kidney2 + kidney_external
bash bash_scripts/train_effnet_v2_m_kidney1_2_external.sh
  1. Calculate pseudo labels for spleen_external
bash bash_scripts/pseudo_label_spleen_external.sh

Note: don't forget to insert correct path to weights in configs/callbacks/test.yaml

  1. Create boundaries masks for all training data
python segment_vasculature/preprocessing/create_boundaries.py
  1. Train effnet_v2_m model for kidney1 + kidney2 + kidney_external + spleen_external
bash bash_scripts/train_effnet_v2_m_kidney1_2_external_spleen.sh
  1. Train dpn model for kidney1 + kidney2 + kidney_external + spleen_external
bash bash_scripts/train_dpn_kidney1_2_external_spleen.sh
  1. Train maxvit model for kidney1 + kidney2 + kidney_external + spleen_external
bash bash_scripts/train_maxvit_kidney1_2_external_spleen.sh

segment-vasculature-5th-place's People

Stargazers

 avatar  avatar  avatar  avatar

Watchers

 avatar

Forkers

hvt1609

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.