app.py

import copy
import enum
import pandas as pd
from typing import List, Optional

import requests
import streamlit as st


http_session = requests.Session()

@enum.unique
class NeuralCategoryClassifierModel(enum.Enum):
    keras_2_0 = "keras-2.0"
    keras_sota_3_0 = "keras-sota-3-0"
    keras_ingredient_ocr_3_0 = "keras-ingredient-ocr-3.0"
    keras_baseline_3_0 = "keras-baseline-3.0"
    keras_original_3_0 = "keras-original-3.0"
    keras_product_name_only_3_0 = "keras-product-name-only-3.0"


LOCAL_DB = False

if LOCAL_DB:
    ROBOTOFF_BASE_URL = "http://localhost:5500/api/v1"
else:
    ROBOTOFF_BASE_URL = "https://robotoff.openfoodfacts.org/api/v1"

PREDICTION_URL = ROBOTOFF_BASE_URL + "/predict/category"


@st.cache
def get_predictions(barcode: str, model_name: str, threshold: Optional[float] = None):
    data = {"barcode": barcode, "predictors": ["neural"], "neural_model_name": model_name}
    if threshold is not None:
        data["threshold"] = threshold

    r = requests.post(PREDICTION_URL, json=data)
    r.raise_for_status()
    return r.json()["neural"]

def display_predictions(
    barcode: str,
    model_names: List[str],
    threshold: Optional[float] = None,
):
    debug = None
    for model_name in model_names:
        response = get_predictions(barcode, model_name, threshold)
        response = copy.deepcopy(response) 
        if model_name != NeuralCategoryClassifierModel.keras_2_0.name and "debug" in response:
            if debug is None:
                debug = response["debug"]
            response.pop("debug")
        st.markdown(f"**{model_name}**")
        st.write(pd.DataFrame(response["predictions"]))
    
    if debug is not None:
        st.markdown("**v3 debug information**")
        st.write(debug)



st.sidebar.title("Category Prediction Demo")
query_params = st.experimental_get_query_params()

default_barcode = query_params["barcode"][0] if "barcode" in query_params else ""
barcode = st.sidebar.text_input(
    "Product barcode", default_barcode
)
threshold = st.sidebar.number_input("Threshold", format="%f", value=0.5) or None
model_names = st.multiselect(
    "Name of the model",
    [x.name for x in NeuralCategoryClassifierModel],
    default=[x.name for x in NeuralCategoryClassifierModel],
)

if barcode:
    barcode = barcode.strip()
    display_predictions(
        barcode=barcode,
        threshold=threshold,
        model_names=model_names,
    )