OVERVIEW 📚

• This project uses a Bidirectional LSTM model and numerous machine learning models such as Naive Bayes, Random Forest, and KNN to classify a person's gender (M/F) based on his/her first name.
• The dataset used consists of 95,025 first names and their corresponding gender.
• Bidirectional LSTM performed best (95% accuracy on train data, 91% accuracy on validation data, 91% accuracy on test data).
• Key Tools & Technologies Used - TensorFlow, Keras, scikit-learn, Flask, HTML, CSS.

---

DEMO 🎥

---

INSTALLATION AND USAGE 🔌

1. Open Anaconda command prompt
2. Create new anaconda environment

conda create -n "gc_project" python==3.7

3. Activate anaconda environment

conda activate "gc_project"

4. Open the project

git clone https://github.com/AparGarg99/Gender-Classifier.git
cd Gender-Classifier

5. Install the required dependencies

pip install -r requirements.txt

6. Create models directory

mkdir models

7. Launch Jupyter Notebook

jupyter notebook

8. Execute 1_BiLSTM_Modeling.ipynb notebook to train and save model in models directory

9. Run flask web app using anaconda prompt

python app.py

10. Open a browser and type in the URL displayed in anaconda prompt. e.g. http://127.0.0.1:5000/

Don't forget to give a ⭐ if you like this project !!