https://github.com/xprithvi/covid-survival-model

This artificial neural network (ANN) was trained on patients with COVID-19 based on pre-conditions such as age, obesity, hypertension, etc. The model was created with PyTorch on Juypter Notebook.
https://github.com/xprithvi/covid-survival-model

classification covid-19 juypter-notebook machine-learning neural-network python pytorch

Last synced: about 2 months ago
JSON representation

This artificial neural network (ANN) was trained on patients with COVID-19 based on pre-conditions such as age, obesity, hypertension, etc. The model was created with PyTorch on Juypter Notebook.

• Host: GitHub
• URL: https://github.com/xprithvi/covid-survival-model
• Owner: xPrithvi
• Created: 2023-09-13T10:50:23.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2023-09-15T13:56:46.000Z (over 1 year ago)
• Last Synced: 2023-09-16T06:12:54.329Z (over 1 year ago)
• Topics: classification, covid-19, juypter-notebook, machine-learning, neural-network, python, pytorch
• Language: Jupyter Notebook
• Homepage:
• Size: 503 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# COVID Survival Model

This artificial neural network (ANN) was trained on patients with COVID-19 based on pre-conditions such as age, obesity, hypertension, etc.

The model was created with PyTorch on Juypter Notebook and was trained on the following [dataset](https://www.kaggle.com/datasets/meirnizri/covid19-dataset) 

which was adapted from this [database](https://datos.gob.mx/busca/dataset/informacion-referente-a-casos-covid-19-en-mexico) released by the government of Mexico.

In particular, "Model-SIM3T" was trained with a 80/20 split concerning the training and test datasets. These datasets were created by selecting all patients 

entries marked as deceased from the original dataset and then randomly sampling the same number patients which were not marked as deceased. In total, the model 

was trained on approximately 120,000 examples and was tested on almost 30,000 examples with an accuracy of 91%. Feature importance analysis via a gradient-based method

revealed the following,

![Feature Importance](https://github.com/xPrithvi/COVID-Survival-Model/assets/34770840/ebc36167-a976-4904-b69e-dd9cdc5643aa)

**NOTE: This project was created for learning purposes and not practical application.**