app.py

import gradio as gr

from pathlib import Path
import openvino as ov
import mmcv
import numpy as np

with open("/model_config/car_classes.txt", "r") as f:
    car_classes = f.readlines()
CAR_CLASSES = [c.replace(",\n", "").replace("'","") for c in car_classes]

def pred_class(image_path):

    base_artifacts_dir = Path("./")

    model_name = "end2end"
    model_xml_name = f'{model_name}.xml'
    model_bin_name = f'{model_name}.bin'

    model_xml_path = base_artifacts_dir / model_xml_name
    device="CPU"
    
    core = ov.Core()
    model = core.read_model(model=model_xml_path)
    compiled_model = core.compile_model(model=model, device_name=device)

    output_layer = compiled_model.output(0)

    image = mmcv.imread(image_path)

    # Resize to MobileNet image shape.
    input_image = mmcv.imresize(image, (224, 224))
    normalized_image = mmcv.imnormalize(input_image, mean=np.array([123.675, 116.28, 103.53]), std=np.array([58.395, 57.12, 57.375]))

    # Reshape to model input shape.
    input_image = np.expand_dims(np.transpose(normalized_image, (2,0,1)), 0)
    result_infer = compiled_model([input_image])[output_layer]
    result_index = np.argmax(result_infer)
    
    return CAR_CLASSES[result_index]

inputs_image = [
    gr.components.Image(type="filepath", label="Input Image"),
]
outputs_text = [
    gr.components.Textbox(label="Output Text"),
]
interface_image = gr.Interface(
    fn=pred_class,
    inputs=inputs_image,
    outputs=outputs_text,
    title="Car Classification",
    cache_examples=False,
)
def greet(name):
    return "Hello " + name + "!!"

gr.TabbedInterface(
    [interface_image],
    tab_names=['Image inference']
).queue().launch()