Add new notebook and remove useless modules

CITATION.cff

@@ -1,11 +1,11 @@
-cff-version: 1.0.1
+cff-version: 1.0.2
 message: "If you use this software, please cite it as below."
 authors:
 - family-names: "Scriva"
   given-names: "Giuseppe"
   orcid: "https://orcid.org/0000-0001-5617-7436"
 title: "Accelerating equilibrium spin-glass simulations using quantum data and deep learning"
-version: 1.0.1
+version: 1.0.2
 doi: 10.5281/zenodo.7118502
-date-released: 2022-09-28
+date-released: 2022-10-26
 url: "https://github.com/gscriva/n-mcmc"

README.md

@@ -6,7 +6,7 @@
 <a href="https://pytorchlightning.ai/"><img alt="Lightning" src="https://img.shields.io/badge/-Lightning-792ee5?logo=pytorchlightning&logoColor=white"></a>
 <a href="https://hydra.cc/"><img alt="Config: Hydra" src="https://img.shields.io/badge/Config-Hydra-89b8cd"></a>
 <a href="https://github.com/ashleve/lightning-hydra-template"><img alt="Template" src="https://img.shields.io/badge/-Lightning--Hydra--Template-017F2F?style=flat&logo=github&labelColor=gray"></a><br>
-[![Paper](http://img.shields.io/badge/paper-arxiv.1001.2234-B31B1B.svg)](https://www.nature.com/articles/nature14539)
+[![Paper](http://img.shields.io/badge/paper-arxiv.1001.2234-B31B1B.svg)](https://arxiv.org/abs/2210.11288v1)
 [![DOI](https://zenodo.org/badge/542095061.svg)](https://zenodo.org/badge/latestdoi/542095061)
 
 </div>
@@ -18,7 +18,7 @@ Adiabatic quantum computers are being used to tackle the optimization task, corr
 Hastings algorithm. In particular, we explore hybrid schemes by combining neural and single spin-flip proposals, as well as D-Wave and classical Monte Carlo training data. The hybrid algorithm
 outperforms the single spin-flip Metropolis-Hastings algorithm and it is competitive with parallel tempering in terms of correlation times, with the significant benefit of a faster equilibration.
 
-For a visual summary (with some results) you can have a look to the notebook [`accelerating_monte_carlo`](notebooks/accelerating_monte_carlo.ipynb) without re-running anything. If you want to reproduce the same plots of the article, move the notebook in the parent directory `n-mcmc`, dowload the data and install the dependecies before run it.    
+For a visual summary (with some results) you can have a look to the notebook [`article_figures`](article_figures.ipynb) without re-running anything. If you want to reproduce the same plots of the article, dowload the data and install the dependecies before run it.    
 
 ## How to run
 Install dependencies
@@ -33,7 +33,31 @@ bash bash/setup_conda.sh
 # install requirements
 pip install -r requirements.txt
 ```
-Get the data from the Zenodo directory #TODO and unzip them in [data/](data/)
+Get the data from the Zenodo directory [10.5281/zenodo.7250436](https://doi.org/10.5281/zenodo.7250436) and move them in [data/](data/), the directory must be organized as follow:
+```
+data
+  ├── couplings
+  |     ├── 100.txt
+  |     .
+  |     .
+  |     └── 484-z8.txt
+  ├── data_for_fig
+  |     ├── data_fig1.csv
+  |     .
+  |     .
+  |     .
+  |     └── data_fig7.csv
+  └── datasets
+        ├── 100-1mus
+        |    ├── train_1us.npy
+        |    └── train_1us.npy
+        .
+        .
+        .
+        └── 484-z8-1mus
+            ├── ...
+            └── ...
+```
 
 Train model with default configuration
 ```yaml
@@ -52,7 +76,7 @@ python run.py experiment=484spin-3nn.yaml
 
 You can generate from the trained model with 
 ```yaml
-python predict.py --ckpt-path=logs/the/trained/model.ckpt --model=made 
+python predict.py --ckpt-path=logs/the/trained/model/path.ckpt --model=made 
 ```
 
 <br>