ChildGrowthMonitor's ML Common code
Project description
Child Growth Monitor Machine Learning
Child Growth Monitor (CGM) is a
game-changing app to detect malnutrition. If you have questions about the project, reach out to info@childgrowthmonitor.org
.
This is the Machine Learnine repository associated with the CGM project.
Introduction
This project uses machine learning to identify malnutrition from 3D scans of children under 5 years of age. This one-minute video explains.
Getting started
Requirements
Our development environment is Microsoft Azure ML
You will need:
- Python 3
- TensorFlow version 2
- other libraries
To install, run:
pip install -r requirements.txt
For installing point cloud libraries, refer to README_installation_details_pcl.md.
Dataset access
If you have access to scan data, you can use: src/data_utils
to understand and visualize the data.
Data access is provided on as-needed basis following signature of the Welthungerhilfe Data Privacy & Commitment to Maintain Data Secrecy Agreement. If you need data access (e.g. to train your machine learning models), please contact Markus Matiaschek for details.
Repository structure
The source code is in src/
.
We make heavy use of AzureML. For AzureML, all code for an experiment run needs to reside in one directory. Example: All code for one specific training, e.g. a ResNet training, needs to be in this training directory.
However, many of our trainings (and also evaluation runs) share large portions of code.
In order to reduce code duplication, we copy shared(a.k.a. common) utility code with copy_dir()
from src/common/
into the training/evaluation directory.
This way, during the experiment run, the code is in the directory and can be used during the run.
Contributing
Please see CONTRIBUTING.md for details.
Run linting / tests
# Make sure to be in the root dir of this repository
flake8 src/
pytest
Versioning
Our releases use semantic versioning. You can find a chronologically ordered list of notable changes in CHANGELOG.md.
License
This project is licensed under the GNU General Public License v3.0. See LICENSE for details and refer to NOTICE for additional licensing notes and use of third-party components.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for cgm_ml_common-3.0.0a2-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 7839608b60f4f978817a42c57a8ca9284d26b53dbe6a7a1d8db02f70429cf071 |
|
MD5 | 2bf1d62764651ff10d629566dc3b8357 |
|
BLAKE2b-256 | e4f18e887e02a82a66de3d93311f1d995d8d13ee779bc10420d042199c2df256 |