0
stars
0
forks
source link
renormalization
- Requirements
- For overview of code + some plots, check out
main.ipynb
- Generate Ising model data:
- run
python generate_data.py
- Place correlated .npy file in correct temperature directory
- Create uncorrelated samples in
supervised_convnet/generate_uncorrelated_data.py
- Set
data
variable to be path of the Ising model data (of each temperature)
- Place uncorrelated .npy file in correct temperature directory
- Train neural network to distinguish between correlated/uncorrelated samples in temperature directory e.g.
supervised_convnet/t_1/train.py
- Make sure both correlated and uncorrelated .npy file in
supervised_convnet/t_1
- Neural network architecture in
supervised_convnet/supervised_convnet.py
- Also contains
IsingDataset
class for pytorch data set loading