Skip to content

spha-code/YouTube-Comments-Sentiment-Analysis-MLOps

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

59 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

YouTube Comments Sentiment Analysis

1. Data Collection and Dataset Construction

The dataset Kanye West YouTube Comments Sentiment Analysis was created by collecting YouTube comments using the YouTube Data API. This process is handled by the following script:

📄 00_Python_YouTube_Comments_Scraper.py

This script retrieves comments from selected YouTube videos and formats them for further analysis.

The resulting dataset is available on Kaggle: Kanye West YouTube Comments Sentiment Analysis

🗃️ https://www.kaggle.com/datasets/sphacode/kanye-west-youtube-comments-sentiment-analysis


2. Data Preprocessing and EDA (Exported Data Analysis)

https://github.com/spha-code/YouTube-Comments-Sentiment-Analysis-MLOps/blob/main/02_experiment_1_MLflow_Baseline_Model.ipynb


3. Starting with a Baseline Model

https://github.com/spha-code/YouTube-Comments-Sentiment-Analysis-MLOps/blob/main/02_MLflow_2_baseline_model.ipynb


4. Setup MLflow server locally for Experiment Tracking

RUN THIS COMMAND IN TERMINAL:

mlflow server --backend-store-uri sqlite:///mlflow.db --default-artifact-root ./mlruns --host 127.0.0.1 --port 5000

In file:

import mlflow

Set local MLflow tracking URI

mlflow.set_tracking_uri("http://127.0.0.1:5000")

Set or create an experiment - not using the default one

mlflow.set_experiment("RF Baseline")

5. Improve Baseline Model

 - TFIDF
 - Max Feature
 - Handling Imbalanced Data
 - Hyperparameter tuning
 - Multiple Model
 - Stacking Model

6. ML Pipeline using DVC

 git init, dvc init
 
 - Data Ingestion
 - Data PreProcessing
 - Model Building
 - Model Evaluation with MLflow
 - Model Register in MLflow

7. Add to Model Registry

8. Implement Chrome plugin

9. Prepare CI/CD workflow

https://github.com/spha-code/YouTube-Comments-Sentiment-Analysis-MLOps/blob/main/.github/workflows/cicd.yaml

10. Dockerfile

https://github.com/spha-code/YouTube-Comments-Sentiment-Analysis-MLOps/blob/main/Docker/Dockerfile

11. Deployment - AWS

12. Github upload

About

youtube_sentiment_analysis

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages