Fashion-Product-masterCategory

Fashion-Product-masterCategory is a vision model fine-tuned from google/siglip2-base-patch16-224 using the SiglipForImageClassification architecture. It classifies fashion product images into high-level master categories.

Classification Report:
                precision    recall  f1-score   support

   Accessories     0.9611    0.9698    0.9654     11244
       Apparel     0.9855    0.9919    0.9887     21361
      Footwear     0.9952    0.9936    0.9944      9197
    Free Items     0.0000    0.0000    0.0000       105
          Home     0.0000    0.0000    0.0000         1
 Personal Care     0.9638    0.9219    0.9424      2139
Sporting Goods     1.0000    0.0400    0.0769        25

      accuracy                         0.9803     44072
     macro avg     0.7008    0.5596    0.5668     44072
  weighted avg     0.9779    0.9803    0.9788     44072

The model predicts one of the following master categories:

• 0: Accessories
• 1: Apparel
• 2: Footwear
• 3: Free Items
• 4: Home
• 5: Personal Care
• 6: Sporting Goods

---

Run with Transformers 🤗

!pip install -q transformers torch pillow gradio

import gradio as gr
from transformers import AutoImageProcessor, SiglipForImageClassification
from PIL import Image
import torch

# Load model and processor
model_name = "prithivMLmods/Fashion-Product-masterCategory"  # Replace with your actual model path
model = SiglipForImageClassification.from_pretrained(model_name)
processor = AutoImageProcessor.from_pretrained(model_name)

# Label mapping
id2label = {
    0: "Accessories",
    1: "Apparel",
    2: "Footwear",
    3: "Free Items",
    4: "Home",
    5: "Personal Care",
    6: "Sporting Goods"
}

def classify_master_category(image):
    """Predicts the master category of 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_master_category,
    inputs=gr.Image(type="numpy"),
    outputs=gr.Label(label="Master Category Prediction Scores"),
    title="Fashion-Product-masterCategory",
    description="Upload a fashion product image to predict its master category (Accessories, Apparel, Footwear, etc.)."
)

# Launch the app
if __name__ == "__main__":
    iface.launch()

---

Intended Use

This model can be applied to:

• E-commerce product categorization
• Automated tagging of product catalogs
• Enhancing search and filtering options
• Data annotation pipelines for fashion datasets