Netflix Movie-TV Show Classifier
Model Description
This project uses a fine-tuned DistilBERT model to classify Netflix content as either a Movie or a TV Show.
Base Model
distilbert-base-uncased
Dataset
Netflix Titles Dataset (netflix_titles.csv)
The dataset contains information about Netflix movies and TV shows, including:
- Title
- Description
- Genre
- Director
- Cast
- Country
- Release Year
- Rating
- Type
Model Input
The model uses:
Title + Description + Genre
Example:
"The Lost Kingdom. A young warrior travels across an ancient kingdom to rescue his family. Action, Adventure."
Output Classes
- Movie
- TV Show
Fine-Tuning
The pre-trained DistilBERT model was fine-tuned using the Netflix dataset for binary text classification.
Training settings:
- Epochs: 3
- Learning Rate: 2e-5
- Batch Size: 16
- Maximum Sequence Length: 256
- Optimizer: AdamW
Prediction Example
Input:
Title: The Lost Kingdom
Description: A young warrior travels across an ancient kingdom to rescue his family and defeat a powerful enemy.
Genre: Action, Adventure
Output:
Prediction: Movie
Evaluation
The model is evaluated using:
- Accuracy
- Precision
- Recall
- F1-score
Technology Used
- Python
- PyTorch
- Hugging Face Transformers
- DistilBERT
- Pandas
- Scikit-learn
- Google Colab
Purpose
The purpose of this project is to demonstrate how a pre-trained transformer model can be fine-tuned on a custom Netflix dataset for text classification.
Model tree for FabiInSpringday/Netflix-Movie-TVShow-Classifier
Base model
distilbert/distilbert-base-uncased