import os 
os.system('pip install detectron2 -f https://dl.fbaipublicfiles.com/detectron2/wheels/cu102/torch1.9/index.html')

import gradio as gr
import cv2
os.system('pip install torch==1.9.0 torchvision==0.10.0')

# check pytorch installation: 
import torch, torchvision
print(torch.__version__, torch.cuda.is_available())
assert torch.__version__.startswith("1.9")   # please manually install torch 1.9 if Colab changes its default version
# Some basic setup:
# Setup detectron2 logger
import detectron2
from detectron2.utils.logger import setup_logger
# import some common libraries
import numpy as np
import os, json, random
# import some common detectron2 utilities
from detectron2 import model_zoo
from detectron2.engine import DefaultPredictor
from detectron2.config import get_cfg
from detectron2.utils.visualizer import Visualizer, ColorMode
from detectron2.data import MetadataCatalog, DatasetCatalog
from PIL import Image
from pathlib import Path
from detectron2.data.datasets import register_coco_instances
from matplotlib import pyplot as plt


cfg = get_cfg()
cfg.MODEL.DEVICE='cpu'
# add project-specific config (e.g., TensorMask) here if you're not running a model in detectron2's core library
cfg.INPUT.MASK_FORMAT='bitmask'
cfg.MODEL.ROI_HEADS.NUM_CLASSES = 3 
cfg.TEST.DETECTIONS_PER_IMAGE = 1000
cfg.merge_from_file(model_zoo.get_config_file("COCO-InstanceSegmentation/mask_rcnn_R_50_FPN_3x.yaml"))
cfg.MODEL.ROI_HEADS.SCORE_THRESH_TEST = 0.5  # set threshold for this model
# Find a model from detectron2's model zoo. You can use the https://dl.fbaipublicfiles... url as well
cfg.MODEL.WEIGHTS = "model_final.pth"

predictor = DefaultPredictor(cfg)


def inference(img):
    # im = cv2.imread(img.name)
    im = cv2.imread(img)
    outputs = predictor(im)

    take = outputs['instances'].scores >= 0.5 #Threshold
    pred_masks = outputs['instances'].pred_masks[take].cpu().numpy()

    mask = np.stack(pred_masks)
    mask = np.any(mask == 1, axis=0)

    p = plt.imshow(im,cmap='gray')
    p1 = plt.imshow(mask, alpha=0.4)

    return plt



title = "Sartorius Cell Instance Segmentation"
description = "Sartorius Cell Instance Segmentation Demo: Current Kaggle competition - kaggle.com/c/sartorius-cell-instance-segmentation"
article = "<p style='text-align: center'><a href='https://ai.facebook.com/blog/-detectron2-a-pytorch-based-modular-object-detection-library-/' target='_blank'>Detectron2: A PyTorch-based modular object detection library</a> | <a href='https://github.com/facebookresearch/detectron2' target='_blank'>Github Repo</a></p>"
examples = [['0030fd0e6378.png']]  
gr.Interface(inference, inputs=gr.inputs.Image(type="filepath"), outputs=gr.outputs.Image('plot')  ,enable_queue=True, title=title,
    description=description,
    article=article,
    examples=examples).launch(debug=False)