Hugging Face's logo

neelnsoni13
/
genderverification

Joblib
Model card Files
xet
Community
neelnsoni13 commited on
Commit
bfacd17
·
verified ·
1 Parent(s): 4b61423

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +139 -3
README.md CHANGED
@@ -1,3 +1,139 @@
1
- ---
2
- license: mit
3
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: mit
3
+ ---
4
+ <html>
5
+
6
+ # Gender Prediction from Names using Neural Network
7
+
8
+ ### 🚀 **Project Overview**
9
+ This project uses a **Neural Network** model trained on **TF-IDF vectors** of names to predict the gender (Male/Female). The model is deployed using **Streamlit**, making it easy to interact and predict the gender from a user-inputted name.
10
+
11
+ ---
12
+
13
+ ## 📸 **Application Screenshot**
14
+ <a href="https://ibb.co/7JMfmKPp"><img src="https://i.ibb.co/Y7nCshmd/Screenshot-2025-02-11-222451.png" alt="Screenshot-2025-02-11-222451" border="0" /></a>
15
+ ---
16
+
17
+ ## 🛠 **How It Works (End-to-End)**
18
+
19
+ ### 1. **Data Preparation**
20
+ - The dataset `gender.xlsx` contains names and their corresponding genders (Male/Female).
21
+ - The `Gender` column is mapped to numerical values:
22
+ - **Male (M)** is mapped to `1`
23
+ - **Female (F)** is mapped to `0`
24
+
25
+ ### 2. **Feature Extraction (TF-IDF Vectorization)**
26
+ - The names are converted to **TF-IDF vectors** using character n-grams (1 to 3 characters).
27
+ - This helps the model learn important patterns in names.
28
+
29
+ ### 3. **Model Training**
30
+ - A **Neural Network** is built using **Keras Sequential API**:
31
+ - Dense layers with **ReLU activation**
32
+ - **Batch Normalization** and **Dropout layers** to prevent overfitting
33
+ - Output layer with **Sigmoid activation** for binary classification
34
+ - The model is trained with **callbacks** like early stopping and learning rate reduction.
35
+
36
+ ### 4. **Saving the Model and Vectorizer**
37
+ - The trained model is saved as `gender_prediction_model_Improve.h5`
38
+ - The TF-IDF vectorizer is saved as `tfidf_vectorizer_Improve.joblib`
39
+
40
+ ### 5. **Streamlit Application**
41
+ - Loads the pre-trained model and vectorizer.
42
+ - Accepts user input (name) and predicts gender.
43
+ - Displays the predicted gender in a clean UI.
44
+
45
+ ---
46
+
47
+ ## 📝 **Project File Structure**
48
+ ```
49
+ .
50
+ ├── TrainImprove.py # Training script for the model
51
+ ├── ml-st1.py # Streamlit app for gender prediction
52
+ ├── gender.xlsx # Dataset with names and gender
53
+ ├── gender_prediction_model_Improve.h5 # Saved Keras model
54
+ ├── tfidf_vectorizer_Improve.joblib # Saved TF-IDF vectorizer
55
+ └── screenshot.png # Screenshot of the app UI
56
+ ```
57
+
58
+ ---
59
+
60
+ ## 🚀 **How to Run the Project**
61
+
62
+ ### 1. **Clone the Repository**
63
+ ```bash
64
+ $ git clone <repository-url>
65
+ $ cd <repository-folder>
66
+ ```
67
+
68
+ ### 2. **Install Dependencies**
69
+ ```bash
70
+ $ pip install -r requirements.txt
71
+ ```
72
+
73
+ ### 3. **Train the Model (Optional)**
74
+ If you want to retrain the model, run the training script:
75
+ ```bash
76
+ $ python TrainImprove.py
77
+ ```
78
+
79
+ ### 4. **Run the Streamlit Application**
80
+ ```bash
81
+ $ streamlit run final.py
82
+ ```
83
+
84
+ ### 5. **Access the App**
85
+ Open your browser and go to: [http://localhost:8501](http://localhost:8501)
86
+
87
+ ---
88
+
89
+ ## 💡 **How the Code Works**
90
+
91
+ ### **Training (TrainImprove.py)**
92
+ 1. **Data Loading:** Reads the dataset from `gender.xlsx`.
93
+ 2. **Preprocessing:** Converts names to TF-IDF vectors.
94
+ 3. **Model Building:** Defines a neural network with regularization.
95
+ 4. **Model Training:** Trains the model with early stopping.
96
+ 5. **Saving Artifacts:** Stores the trained model (`.h5`) and vectorizer (`.joblib`).
97
+
98
+ ### **Application (final.py)**
99
+ 1. **Load Model and Vectorizer:** Loads the pre-trained model and TF-IDF vectorizer.
100
+ 2. **User Input:** Accepts a name input from the user.
101
+ 3. **Prediction:** Transforms the name using TF-IDF and makes a prediction.
102
+ 4. **Output:** Displays the predicted gender (Male/Female) in the app.
103
+
104
+ ---
105
+
106
+ ## 📦 **Dependencies**
107
+ - Python 3.x
108
+ - TensorFlow
109
+ - Scikit-learn
110
+ - Pandas
111
+ - Streamlit
112
+
113
+ Install them using:
114
+ ```bash
115
+ $ pip install tensorflow scikit-learn pandas streamlit joblib
116
+ ```
117
+
118
+ ---
119
+
120
+ ## 🎨 **Future Enhancements**
121
+ - Improve the UI design.
122
+ - Include more diverse datasets for better generalization.
123
+ - Add confidence scores for predictions.
124
+ - Deploy the app online for public access.
125
+
126
+ ---
127
+
128
+ ## 🤝 **Contributing**
129
+ Feel free to fork the project and submit a pull request for improvements.
130
+
131
+ ---
132
+
133
+ ## 📜 **License**
134
+ This project is licensed under the MIT License.
135
+
136
+
137
+
138
+
139
+ </html>