|
|
--- |
|
|
license: apache-2.0 |
|
|
language: |
|
|
- en |
|
|
base_model: |
|
|
- google/siglip2-base-patch16-224 |
|
|
pipeline_tag: image-classification |
|
|
library_name: transformers |
|
|
tags: |
|
|
- gender |
|
|
- fashion |
|
|
- product |
|
|
--- |
|
|
|
|
|
 |
|
|
|
|
|
# **Fashion-Product-Gender** |
|
|
|
|
|
> **Fashion-Product-Gender** is a vision model fine-tuned from **google/siglip2-base-patch16-224** using the **SiglipForImageClassification** architecture. It classifies fashion product images into one of five gender categories. |
|
|
|
|
|
```py |
|
|
Classification Report: |
|
|
precision recall f1-score support |
|
|
|
|
|
Boys 0.4127 0.0940 0.1531 830 |
|
|
Girls 0.5000 0.0061 0.0121 655 |
|
|
Men 0.7506 0.8393 0.7925 22104 |
|
|
Unisex 0.5714 0.0188 0.0364 2126 |
|
|
Women 0.7317 0.7609 0.7460 18357 |
|
|
|
|
|
accuracy 0.7407 44072 |
|
|
macro avg 0.5933 0.3438 0.3480 44072 |
|
|
weighted avg 0.7240 0.7407 0.7130 44072 |
|
|
``` |
|
|
|
|
|
The model predicts one of the following gender categories for fashion products: |
|
|
|
|
|
- **0:** Boys |
|
|
- **1:** Girls |
|
|
- **2:** Men |
|
|
- **3:** Unisex |
|
|
- **4:** Women |
|
|
|
|
|
--- |
|
|
|
|
|
# **Run with Transformers 🤗** |
|
|
|
|
|
```python |
|
|
!pip install -q transformers torch pillow gradio |
|
|
``` |
|
|
|
|
|
```python |
|
|
import gradio as gr |
|
|
from transformers import AutoImageProcessor, SiglipForImageClassification |
|
|
from PIL import Image |
|
|
import torch |
|
|
|
|
|
# Load model and processor |
|
|
model_name = "prithivMLmods/Fashion-Product-Gender" # Replace with your actual model path |
|
|
model = SiglipForImageClassification.from_pretrained(model_name) |
|
|
processor = AutoImageProcessor.from_pretrained(model_name) |
|
|
|
|
|
# Label mapping |
|
|
id2label = { |
|
|
0: "Boys", |
|
|
1: "Girls", |
|
|
2: "Men", |
|
|
3: "Unisex", |
|
|
4: "Women" |
|
|
} |
|
|
|
|
|
def classify_gender(image): |
|
|
"""Predicts the gender category for a fashion product.""" |
|
|
image = Image.fromarray(image).convert("RGB") |
|
|
inputs = processor(images=image, return_tensors="pt") |
|
|
|
|
|
with torch.no_grad(): |
|
|
outputs = model(**inputs) |
|
|
logits = outputs.logits |
|
|
probs = torch.nn.functional.softmax(logits, dim=1).squeeze().tolist() |
|
|
|
|
|
predictions = {id2label[i]: round(probs[i], 3) for i in range(len(probs))} |
|
|
return predictions |
|
|
|
|
|
# Gradio interface |
|
|
iface = gr.Interface( |
|
|
fn=classify_gender, |
|
|
inputs=gr.Image(type="numpy"), |
|
|
outputs=gr.Label(label="Gender Prediction Scores"), |
|
|
title="Fashion-Product-Gender", |
|
|
description="Upload a fashion product image to predict the target gender category (Boys, Girls, Men, Unisex, Women)." |
|
|
) |
|
|
|
|
|
# Launch the app |
|
|
if __name__ == "__main__": |
|
|
iface.launch() |
|
|
``` |
|
|
|
|
|
--- |
|
|
|
|
|
# **Intended Use** |
|
|
|
|
|
This model is best suited for: |
|
|
|
|
|
- **Fashion E-commerce tagging and search** |
|
|
- **Personalized recommendations based on gender** |
|
|
- **Catalog organization and gender-based filters** |
|
|
- **Retail analytics and demographic insights** |
