*Animal Classifier app using Pytorch framework *

Instructios to run this

* Install Anaconda
* Create a conda environment
* Install pyrorch CPU or GPU version from https://pytorch.org/
* Run main.py
* Do not run train if you dont have a GPU for training
* Extract and move contents from data.zip to data folder
* Download Animal_Classifier_80-10-10.pth and remane it to Animal_Classifier.pth then move it to Model_Save folder

Note

* For Training testing and Inference you need data.zip and conda environment with Dependencies from requirements.txt
* For Inference only you need  Animal_Classifier_80-10-10.pth and conda environment ith Dependencies from requirements.txt

• Contain a full DL structure from training to testing to inference

• The codes are created in a modular manner which helps if you want to create a new project

• Created based on plug and play modules

• Available animals are dog, horse, elephant, butterfly, chicken, cat, cow, sheep, spider, squirrel

• Used dataset is Animals-10 clasification dataset from kaggle

• Dataset link https://www.kaggle.com/alessiocorrado99/animals10/notebooks

• The Dataset from Kaggle has been altered to satisfy the needs of thsis project

• If you have suggestion to improve the code feel free to let me know

• If you like the project please prove a '*' to this repo

• Follow me for more projects like this

Downloads

• Augumented Dataset https://drive.google.com/file/d/1_5m9TTmr3NeClO75MtPzVixeShG1Qp2_/view?usp=sharing
• Model trained by me if you want to test any images in the listed 10 classes https://drive.google.com/file/d/1kXXzu_biwAQRnRmw9PSq9qLrucD9wvDS/view?usp=sharing

Current build issues

• Train and test function only works on Linux Machines for now (if you found it working on a Windows machine please let me know)