Mushroom Classification Model

Name: Christer Henrysson
Student Number: 101260693

This is a ResNet18-based CNN that classifies mushroom images and outputs the most likely top 3 species (with percentages of confidence).

The trained model was to large for github so you can find it here https://drive.google.com/file/d/1Pb7RKg5J9dacixvK6gWcTevNFaUaXM_r/view?usp=sharing 

How to get requirements:
Run "pip install -r requirements.txt"

How To Train:
If you want to train your own model
Download the dataset from https://www.kaggle.com/datasets/zlatan599/mushroom1
Insert dataset into project directory
Train by running "python model.py"

How To Run:
To get model I trained already go to the following link download the model and place it in the project directory https://drive.google.com/file/d/1Pb7RKg5J9dacixvK6gWcTevNFaUaXM_r/view?usp=sharing 
Insert desired mushroom images you want to predict into "images" folder (there is a example photo in there)
Run "python runner.py"
Read outputed text for predictions

How to evaluate:
Run "python evaluate.py"

Make sure:
CSV files (train.csv, val.csv, test.csv) are in the same poject directory
Images folder (/images) exists for predictions
Model file best_resnet18_mushroom.pth is generated before inference