import json
import gradio as gr
import yolov5
from PIL import Image
from huggingface_hub import hf_hub_download

app_title = "NFL Object Detection"
models_ids = ['keremberke/yolov5n-nfl', 'keremberke/yolov5s-nfl', 'keremberke/yolov5m-nfl']
article = f"<p style='text-align: center'> <a href='https://huggingface.co/{models_ids[-1]}'>huggingface.co/{models_ids[-1]}</a> | <a href='https://huggingface.co/keremberke/nfl-object-detection'>huggingface.co/keremberke/nfl-object-detection</a> | <a href='https://github.com/keremberke/awesome-yolov5-models'>awesome-yolov5-models</a> </p>"

current_model_id = models_ids[-1]
model = yolov5.load(current_model_id)

examples = [['test_images/57638_001089_Endzone_frame262_jpg.rf.4a34e04af4f7b46c8dd9454e34740317.jpg', 0.25, 'keremberke/yolov5m-nfl'], ['test_images/57660_001234_Endzone_frame0845_jpg.rf.745d52b49774ae36d821d752435c8481.jpg', 0.25, 'keremberke/yolov5m-nfl'], ['test_images/57848_002061_Sideline_frame0529_jpg.rf.3747d55691fd4bd0ca5f26e713531f6e.jpg', 0.25, 'keremberke/yolov5m-nfl'], ['test_images/57873_003005_Endzone_frame356_jpg.rf.5f45decabc82c2f9c102bfe4200ece25.jpg', 0.25, 'keremberke/yolov5m-nfl'], ['test_images/57928_002004_Sideline_frame0820_jpg.rf.31445da39fd67e4455b8107cbe7918f5.jpg', 0.25, 'keremberke/yolov5m-nfl'], ['test_images/58037_001432_Sideline_frame386_jpg.rf.0e72f6bd6a685a8149467eeb50184c56.jpg', 0.25, 'keremberke/yolov5m-nfl']]


def predict(image, threshold=0.25, model_id=None):
    # update model if required
    global current_model_id
    global model
    if model_id != current_model_id:
        model = yolov5.load(model_id)
        current_model_id = model_id

    # get model input size
    config_path = hf_hub_download(repo_id=model_id, filename="config.json")
    with open(config_path, "r") as f:
        config = json.load(f)
    input_size = config["input_size"]

    # perform inference
    model.conf = threshold
    results = model(image, size=input_size)
    numpy_image = results.render()[0]
    output_image = Image.fromarray(numpy_image)
    return output_image


gr.Interface(
    title=app_title,
    description="Created by 'keremberke'",
    article=article,
    fn=predict,
    inputs=[
        gr.Image(type="pil"),
        gr.Slider(maximum=1, step=0.01, value=0.25),
        gr.Dropdown(models_ids, value=models_ids[-1]),
    ],
    outputs=gr.Image(type="pil"),
    examples=examples,
    cache_examples=True if examples else False,
).launch(enable_queue=True)