This Project support sentiment classification for binary as well as multi class classification
The project has the following structure:
Sentiment_Classifier
|---
sentiment_analysis
|----
|---data
|---preprocessor
|---feature_extracction
|---model
|---saved_instance
|---notebook
|---config.py
|---__main__.py
|---app.py
|---requirement.txt
|---README.md
data : consist of datafile and dataloader module.
- the datafile consist of the dataset .
preprocessor : module for text pre-processing .
feature_extraction : module for extracting feature from the processed data.
model : module responsible for training model on the extracted feature.
saved_instance : stores the trained model and feature vectorizer on the disk .
notebook : For detailed understanding of techniques use in training model check this one out.
config.py : This config file is needed to be initialize before running __main__.py .
__main__.py : This file is needed to be run to train the model .
app.py : Will open an endpoint for client .