app.py

from gradio.outputs import Label
from icevision.all import *
from icevision.models.checkpoint import *
import PIL
import gradio as gr
import os

# Load model
checkpoint_path = "models/model_checkpoint.pth"
checkpoint_and_model = model_from_checkpoint(checkpoint_path)
model = checkpoint_and_model["model"]
model_type = checkpoint_and_model["model_type"]
class_map = checkpoint_and_model["class_map"]

# Transforms
img_size = checkpoint_and_model["img_size"]
valid_tfms = tfms.A.Adapter([*tfms.A.resize_and_pad(img_size), tfms.A.Normalize()])

examples = [['sample_images/IMG_20191212_151351.jpg'],['sample_images/IMG_20191212_153420.jpg'],['sample_images/IMG_20191212_154100.jpg']]

def show_preds(input_image):
    img = PIL.Image.fromarray(input_image, "RGB")
    pred_dict  = model_type.end2end_detect(img, valid_tfms, model, class_map=class_map, detection_threshold=0.5,
                                           display_label=False, display_bbox=True, return_img=True, 
                                           font_size=16, label_color="#FF59D6")
                                           
    return pred_dict["img"], len(pred_dict["detection"]["bboxes"])


gr_interface = gr.Interface(
    fn=show_preds,
    inputs=["image"],
    outputs=[gr.outputs.Image(type="pil", label="RetinaNet Inference"), gr.outputs.Textbox(type="number", label="Microalgae Count")],
    title="Microalgae Detector with RetinaNet",
    description="This RetinaNet model counts microalgaes on a given image. Upload an image or click an example image below to use.",
    article="<p style='text-align: center'><a href='https://dicksonneoh.com/' target='_blank'>Blog post</a></p>",
    examples=examples,
)
gr_interface.launch()