app.py

# -*- coding: utf-8 -*-
"""tabular_gradio.ipynb

Automatically generated by Colab.

Original file is located at
    https://colab.research.google.com/drive/1AOluTRpRSCrd9WaBE5wvo6sojvJ3My6F
"""

# app.py
import os
import shutil
import zipfile
import pathlib
import pandas as pd
import gradio as gr
from huggingface_hub import hf_hub_download
from autogluon.tabular import TabularPredictor

MODEL_REPO_ID = "its-zion-18/flowers-tabular-autolguon-predictor"
ZIP_FILENAME  = "autogluon_predictor_dir.zip"
CACHE_DIR = pathlib.Path("hf_assets")
EXTRACT_DIR = CACHE_DIR / "predictor_native"

FEATURE_COLS = [
    "flower_diameter_cm",
    "petal_length_cm",
    "petal_width_cm",
    "petal_count",
    "stem_height_cm",
]
COLOR_LABELS = ["Red", "Orange", "Pink", "White", "Yellow", "Purple"]

def _prepare_predictor_dir() -> str:
    CACHE_DIR.mkdir(parents=True, exist_ok=True)
    local_zip = hf_hub_download(
        repo_id=MODEL_REPO_ID,
        filename=ZIP_FILENAME,
        repo_type="model",
        local_dir=str(CACHE_DIR),
        local_dir_use_symlinks=False,
    )
    if EXTRACT_DIR.exists():
        shutil.rmtree(EXTRACT_DIR)
    EXTRACT_DIR.mkdir(parents=True, exist_ok=True)
    with zipfile.ZipFile(local_zip, "r") as zf:
        zf.extractall(str(EXTRACT_DIR))
    contents = list(EXTRACT_DIR.iterdir())
    predictor_root = contents[0] if (len(contents) == 1 and contents[0].is_dir()) else EXTRACT_DIR
    return str(predictor_root)

PREDICTOR_DIR = _prepare_predictor_dir()
PREDICTOR = TabularPredictor.load(PREDICTOR_DIR, require_py_version_match=False)

def do_predict(flower_diameter_cm, petal_length_cm, petal_width_cm, petal_count, stem_height_cm):
    X = pd.DataFrame([{
        FEATURE_COLS[0]: float(flower_diameter_cm),
        FEATURE_COLS[1]: float(petal_length_cm),
        FEATURE_COLS[2]: float(petal_width_cm),
        FEATURE_COLS[3]: int(petal_count),
        FEATURE_COLS[4]: float(stem_height_cm),
    }], columns=FEATURE_COLS)

    try:
        pred = PREDICTOR.predict(X).iloc[0]
    except Exception as e:
        return f"**Prediction failed:** {e}", {}

    proba_dict = {}
    try:
        proba = PREDICTOR.predict_proba(X)
        if isinstance(proba, pd.Series):
            proba = proba.to_frame().T
        row0 = proba.iloc[0]
        proba_dict = {str(k): float(v) for k, v in row0.items()}
        proba_dict = dict(sorted(proba_dict.items(), key=lambda kv: kv[1], reverse=True))
    except Exception:
        pass

    md = f"**Predicted color:** {pred}"
    if pred in proba_dict:
        md += f"\n\n**Confidence:** {proba_dict[pred]*100:.2f}%"
    return md, proba_dict

EXAMPLES = [
    [3.4, 1.3, 1.0,  7, 68.7],
    [5.7, 3.2, 0.9,  5, 21.2],
    [9.7, 4.6, 0.6, 21, 72.6],
    [7.8, 2.8, 0.7, 30, 25.5],
    [9.0, 4.5, 1.8, 10, 13.1],
    [3.5, 4.6, 1.6, 17, 13.4],
]

with gr.Blocks() as demo:
    gr.Markdown("# Flower Color Classifier (Tabular)")
    gr.Markdown("Adjust the sliders to set flower measurements. The model predicts the flower's color and shows class probabilities.")

    with gr.Row():
        flower_diameter_cm = gr.Slider(3.0, 10.0, step=0.1, value=6.0, label=FEATURE_COLS[0])
        petal_length_cm    = gr.Slider(1.0,  5.0, step=0.1, value=3.0, label=FEATURE_COLS[1])
        petal_width_cm     = gr.Slider(0.5,  2.0, step=0.1, value=1.0, label=FEATURE_COLS[2])

    with gr.Row():
        petal_count        = gr.Slider(1,   30,  step=1,   value=12,  label=FEATURE_COLS[3])
        stem_height_cm     = gr.Slider(10.0, 80.0, step=0.1, value=40.0, label=FEATURE_COLS[4])

    pred_md = gr.Markdown()
    proba_pretty = gr.Label(num_top_classes=6, label="Class probabilities")

    inputs = [flower_diameter_cm, petal_length_cm, petal_width_cm, petal_count, stem_height_cm]
    for comp in inputs:
        comp.change(fn=do_predict, inputs=inputs, outputs=[pred_md, proba_pretty])

    gr.Examples(examples=EXAMPLES, inputs=inputs, label="Representative examples", examples_per_page=5, cache_examples=False)

if __name__ == "__main__":
    demo.launch()

"""UPLOADING TO HUGGINGFACE"""