File size: 2,825 Bytes
a1cb511 a255492 0e3c9d4 a1cb511 a255492 0e3c9d4 a1cb511 a255492 a1cb511 0e3c9d4 a1cb511 a255492 a1cb511 0e3c9d4 a1cb511 a255492 a1cb511 0e3c9d4 a1cb511 a255492 0e3c9d4 a255492 0e3c9d4 e661df8 0e3c9d4 a255492 b574a9e | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 | import gradio as gr
import pandas as pd
from catboost import CatBoostRegressor
import os
# 1. Load the model
# Make sure 'catboost_skincare_model.cbm' is uploaded to the same Space
model = CatBoostRegressor()
if os.path.exists("catboost_skincare_model.cbm"):
model.load_model("catboost_skincare_model.cbm")
else:
print("Model file not found!")
def predict_rating(brand, product, cat1, cat2, origin, age_range, skin_type, concern, ing1, ing2, ing3):
# 2. Feature Engineering: Age Mapping
age_map = {
'18 and Under': 'Teens',
'19 - 24': 'Young Adult',
'25 - 29': 'Young Adult',
'30 - 34': 'Adult',
'35 - 39': 'Adult',
'40 - 44': 'Mature',
'45 and Above': 'Mature'
}
age_segment = age_map.get(age_range, 'Young Adult')
# 3. Feature Engineering: Ingredient Flags
all_ings = f"{ing1} {ing2} {ing3}".lower()
def check_ing(term):
return "1" if term in all_ings else "0"
# 4. Create DataFrame for prediction (Must match training column order)
input_df = pd.DataFrame([{
'brand_name': brand,
'product_name': product,
'1st_category': cat1,
'2nd_category': cat2,
'brand_country_origin': origin,
'age': age_range,
'skin_type': skin_type,
'skin_concern': concern,
'ingredient_1': ing1,
'ingredient_2': ing2,
'ingredient_3': ing3,
# Engineered features the model expects
'age_segment': age_segment,
'has_niacinamide': check_ing('niacinamide'),
'has_retinol': check_ing('retinol'),
'has_salicylic_acid': check_ing('salicylic'),
'has_hyaluronic_acid': check_ing('hyaluronic'),
'has_vitamin_c': check_ing('vitamin c'),
'skin_profile': f"{skin_type}_{concern}",
'product_skin_fit': f"{cat2}_{skin_type}"
}])
# 5. Predict
prediction = model.predict(input_df)
# Return rounded rating
return round(float(prediction[0]), 2)
# 6. Define the Gradio Interface
inputs = [
gr.Textbox(label="Brand Name"),
gr.Textbox(label="Product Name"),
gr.Textbox(label="1st Category"),
gr.Textbox(label="2nd Category"),
gr.Textbox(label="Country of Origin"),
gr.Dropdown(choices=['18 and Under', '19 - 24', '25 - 29', '30 - 34', '35 - 39', '40 - 44', '45 and Above'], label="Age Range"),
gr.Textbox(label="Skin Type"),
gr.Textbox(label="Skin Concern"),
gr.Textbox(label="Ingredient 1"),
gr.Textbox(label="Ingredient 2"),
gr.Textbox(label="Ingredient 3")
]
demo = gr.Interface(
fn=predict_rating,
inputs=inputs,
outputs=gr.Number(label="Predicted Rating"),
title="Skincare Rating Predictor API",
api_name="predict_rating"
)
if __name__ == "__main__":
demo.launch(show_error=True, share=False) |