This repository hosts a suite of Python codes aimed at addressing the critical challenges faced by the Small-Scale fisheries in the Gulf of California, a pivotal region for Mexico's fisheries production and biodiversity. Amidst the backdrop of climate change, these tools leverage Deep Learning techniques to forecast fish catch volumes, contributing to sustainable resource management and informed decision-making.
The Gulf of California, a key ecosystem with significant contributions to Mexico's fisheries output, is experiencing shifts due to climate change, impacting fish distribution, species targeted, and overall catch volumes. This project aims to utilize Artificial Intelligence to understand and predict changes in catch volumes under varying climate scenarios, focusing on the multi-species small-scale fisheries in the region.
The goal is to employ deep learning models to explore the complex dynamics between oceanic conditions and catch volumes, thereby aiding in the sustainable management of fisheries in the Northern Gulf of California. This involves analyzing the intricate relationships and predicting future catch volumes amidst evolving climate patterns.
- data_preparation.py: Functions for data loading and preparation.
- forecasting.py: Contains forecasting logic and visualizations.
- model.py: Script for LSTM model creation, training, and evaluation.
- utils.py: Utility functions supporting data handling and visualization.
- config.py: Configuration settings for model and data preparation.
Instructions on how to set up the environment, load the model, and execute the forecasting scripts. Include steps for data preparation, model training (if applicable), and performing forecasts.
Guidelines for contributing to the project, including coding standards, pull request procedures, and other relevant information.
Specify the license under which the project is released.
Credit to researchers, data sources, and any other support received during the development of this forecasting tool.