ben-hawks / ecoder

ECON-T autoencoder model

-# Ecoder ECON-T autoencoder model

Setup

On VM klijnsma-gpu3

Get data and untar

mkdir data; cd data
wget https://www.dropbox.com/s/502o1h5y0ukkasf/ecoder.tar.gz 
tar -xvzf ecoder.tar.gz
mv uscms/home/kkwok/eos/ecoder/* .

Setup environment using miniconda3

source install_miniconda3.sh #if your first time
source setup.sh #also if your first time
conda activate ecoder-env
pip install keras tensorflow numba numpy pandas matplotlib tensorflow_model_optimization pillow ot

Setup qkeras (h/t Thea!):

git clone https://github.com/google/qkeras.git
cd qkeras
python setup.py build
python setup.py install --user
cd ..

Setup on LPC

If you are working on the LPC cluster working node, use the following scripts to setup the environment

source LPC_envSetup.sh      ##do this for the first time
source lpc_env.sh           ##do this everytime

Juypter notebook demos

Following files illustrates prototypes of different autoencoder architectures

auto.ipynb - 1D deep NN autoencoder demo

auto_CNN.ipynb - 2D CNN autoencoder demo

Auto_qCNN.ipynb - 2D quantized CNN autoencoder, trained with qKeras demo

qkeras instructions: https://github.com/google/qkeras

Training scripts

Scripts to explore hyperparameters choices:

models.py - constructs and compile simple model architectures

denseCNN.py - model class for constructing conv2D-dense architectures

train.py - train(or load weights) and evaluate models

Example usage:

## edit parameters setting inside train.py
## train with 1 epoch to make sure model parameters are OK, output to a trainning folder
python train.py -i ~/eos/ecoder/pgun_pid1_pt200_200PU.csv  -o ./qjet_200PU/  --epoch 1
## train the weights with max 150 epoch 
python train.py -i ~/eos/ecoder/pgun_pid1_pt200_200PU.csv  -o ./qjet_200PU/  --epoch 150

## After producing a `.hdf5` file from trainning, you can re-run the model skipping the trainning phase.
## Do so by simply setting the model parameter 'ws' to `modelname.hdf5`