This project shows how to apply transfer learning with MXNet Gluon, on the husky or akita
classification problem.
- install MXNet:
pip install mxnet
- install Jupyter Notebook:
pip install jupyter
data
directory:img
: image data to be used in generatingrec
filetrain
,sample
,test
,validation
: rec files for training
prepare-dataset.ipynb
: preprocess and prepare datasets for trainingtrain.ipynb
: build model and define training processgenerate_rec.sh
: script to generate rec filesim2rec.py
: script provided by MXNet tools
This project provides the simple guidance to apply transfer learning.
Users are expected to gather their own dataset and follow the notebooks.
For more details, please refer to this post