# Project Setup and Usage

## Environment Setup

This project uses conda for dependency management. To replicate the exact environment used for this project, follow these steps:

### Prerequisites
- [Anaconda](https://www.anaconda.com/products/distribution) or [Miniconda](https://docs.conda.io/en/latest/miniconda.html) installed on your system

### Installation Steps

1. **Clone the repository** (if applicable):
   ```bash
   git clone <repository-url>
   cd <repository-name>
   ```

2. **Create the conda environment** from the provided environment file:
   ```bash
   conda env create -f environment.yml
   ```

3. **Activate the environment**:
   ```bash
   conda activate <environment-name>
   ```

## Running the Code

The data used in the example (darcy) may be visualized and verified using data_vis.ipynb. 
The original datasets processed in the data_vis.ipynb file are taken from: https://www.kaggle.com/datasets/scaomath/pde-dataset 
Put these datasets in data-generation/darcy to run the example. 
The processing produces a 1d dataset of 8400 samples and a 2d dataset of 1200 samples. 
The example code for train_fno.py and the FNO implementation is only for 2d at this time. 

Once the environment is set up and activated, you can run the main training script:

```bash
python train_fno.py './experiments/code_debug/config.yml'
```

This command will execute the Fourier Neural Operator training with the configuration specified in the `config.yml` file.
To modify the hyperparameters and training specificatios, modify the 'config.yml' file. 


## Additional Notes

- The environment name will be taken from the `environment.yml` file
- All required dependencies and their versions are specified in the environment file to ensure reproducibility
- Make sure to activate the environment each time you work on this project