Data

Tools

Indexes

Applications

Experiments

Open Source Science

https://github.com/gscriva/dlai-project

https://github.com/gscriva/dlai-project

Science Score: 23.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • ○
    CITATION.cff file
  • ○
    codemeta.json file
  • ○
    .zenodo.json file
  • ✓
    DOI references
    Found 1 DOI reference(s) in README
  • ✓
    Academic publication links
    Links to: nature.com, aps.org
  • ○
    Academic email domains
  • ○
    Institutional organization owner
  • ○
    JOSS paper metadata
  • ○
    Scientific vocabulary similarity
    Low similarity (7.6%) to scientific vocabulary
Last synced: 5 months ago · JSON representation

Repository

Basic Info
  • Host: GitHub
  • Owner: gscriva
  • Language: Jupyter Notebook
  • Default Branch: main
  • Homepage:
  • Size: 3.41 MB
Statistics
  • Stars: 0
  • Watchers: 1
  • Forks: 0
  • Open Issues: 0
  • Releases: 0
Created about 5 years ago · Last pushed over 4 years ago

https://github.com/gscriva/dlai-project/blob/main/ 

# A Scalable Model for Predicting Ground State Energy

## Problem 
Predict the lowest energy level of a one-dimensional ultracold atom subject to an external optical speckle field (disorder).
Our aim is to solve the Schrdinger equation and thus find the the ground state energy. 


## Goal
We want to reproduce the results achieved in [Scientific reports 9.1 (2019): 1-12.](https://www.nature.com/articles/s41598-019-42125-w), but using a scalable version of the network, as did in [Physical Review E 102.3 (2020): 033301](https://journals.aps.org/pre/abstract/10.1103/PhysRevE.102.033301).


## Usage

### Requirements
Install all the requirements using `pip install -r requirements.txt`, we recommend to create an isolated environment before.

### Data structure
The input data must be organized as follow:
```
data
   train_data_L14.npz
   test_data_L14.npz
   train_data_L28.npz
   test_dataL_28.npz
   ...
```

They are not present in the git repo, but they are available in my own Drive.

## Train
To train with the simplest MLP model one has to run

```python main.py --train --data_dir data/train_dataL14.npz --input_size 15```.

Several commands are avialable to switch model, change MLP hidden structure, etc. Just run `python main.py --help`.

Owner

  • Name: Giuseppe Scriva
  • Login: gscriva
  • Kind: user
  • Location: Zurich
  • Company: University of Camerino

Just another infinite monkey trying to write his Divine Comedy.

GitHub Events

Total
Last Year

Dependencies

requirements.txt pypi
  • ignite ==0.4.2
  • numpy ==1.20.1
  • torch ==1.7.1
  • torchvision ==0.8.2
  • wandb ==0.10.17