mlzoomcamp-2022-capstone-project-2

This is the repository for the second capstone project of the Machine Learning Zoomcamp course 2022.

Project description

In this project, I have worked on a Kaggle image classification contest organized by DataTalks.Club.

The goal of the competition is to classify images of kitchenware into 6 classes. The dataset contains 9367 images of kitchenware, each belonging to one of 6 classes. The classes are:

cups
glasses
plates
spoons
forks
knives

The dataset is split into a training set and a test set. The training set contains 5559 images, while the test set contains 3808 images.

Project structure

The project is structured as follows along with files and folders:

kitchenware-classification/images/ - contains the images named as .jpg (NOTE: Make sure to download the data from Kaggle and put images folder here.)
kitchenware-classification/data/ - contains the train.csv and test.csv files.
kitchenware-classification/models/ - contains the trained models.
kitchenware-classification/notebooks.ipynb - Jupyter notebook which have the EDA along with the model training code and the submission code.
kitchenware-classification/kaggel-submission.csv - submission file
Pipfile - pipenv file
Pipfile.lock - pipenv file
Dockerfile - docker file for the API and streamlit app
main.py - API file which will run the model and return the prediction for web image.
streamlit.py - streamlit app which will run the model and return the prediction for local image.
README.md - Readme file

Training model:

Build model with train script:

NOTE: Make sure to download the data from Kaggle and put images folder under kitchenware-classification folder.

pipenv run python train.py - this will train the model and save it in the models folder.

Test with Docker:

Run Docker which will run the API and the streamlit app together:

docker run -p 8000:8000 -p 8501:8501 -it mahesh00000/mlzoomcamp-2022-capstone-project-2:latest

API:

curl -X 'POST' \
  'http://localhost:8000/predict' \
  -H 'accept: application/json' \
  -H 'Content-Type: application/json' \
  -d '{
  "image": "https://vistaalegre.com/eu/content/images/thumbs/0039972_us-perle-cn-garfo-de-mesa-perle.jpeg"
}'

Streamlit app: Access the streamlit app on http://localhost:8501/ in your browser.

Run API and streamlit app locally:

Run & Test the API with web image:

pipenv run python -m uvicorn main:app

curl -X 'POST' \
  'http://localhost:8000/predict' \
  -H 'accept: application/json' \
  -H 'Content-Type: application/json' \
  -d '{
  "image": "https://vistaalegre.com/eu/content/images/thumbs/0039972_us-perle-cn-garfo-de-mesa-perle.jpeg"
}'

Run & Test with streamlit app:

pipenv run streamlit run app.py - this will run the streamlit app locally on port 8501. Access the streamlit app on http://localhost:8501/ in your browser.

msinha251/mlzoomcamp-2022-capstone-project-2