Amazon Fine Food Reviews is sentiment analysis problem where we classify each review as positive and negative using machine learning and deeplearning techniques. Finally we will deploy our best model using Flask.
https://www.kaggle.com/snap/amazon-fine-food-reviews
Given a review, determine whether the review is positive (Rating of 4 or 5) or negative (rating of 1 or 2).
I have written a blog explaining the approach i used to solve this sentiment analysis problem from basic EDA to deployment. You can read my medium blog for that.
We got a better generalized model using LSTM with 2 layers.
- First install the requirements
pip install -r requirements.txt
-
Download the weights from drive and move the contents to 'weights' folder.
-
Start our applicaion by running
python app.py
note: After cloning the repository if you donot need the notebooks, you are free to delete Notebook folder. You can also test our application using tmp.py
python tmp.py