app.py

import os
import gradio as gr
import torch
from torchvision import models, transforms
from PIL import Image

# -- get torch and cuda version
#TORCH_VERSION = ".".join(torch.__version__.split(".")[:2])
#CUDA_VERSION = torch.__version__.split("+")[-1]

# -- install pre-build detectron2
os.system('pip install git+https://github.com/facebookresearch/detectron2.git')
os.system('pip install pyyaml==5.1')

import detectron2
from detectron2.utils.logger import setup_logger # ????

from detectron2 import model_zoo
from detectron2.engine import DefaultPredictor
from detectron2.config import get_cfg
'''
# ????
setup_logger()

# -- load rcnn model
cfg = get_cfg()
# add project-specific config (e.g., TensorMask) here if you're not running a model in detectron2's core library
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_zoo.get_checkpoint_url("COCO-InstanceSegmentation/mask_rcnn_R_50_FPN_3x.yaml")
predictor = DefaultPredictor(cfg)

os.system(wget http://images.cocodataset.org/val2017/000000439715.jpg -q -O input.jpg)
im = cv2.imread("./input.jpg")
cv2_imshow(im)

outputs = predictor(im)

print(outputs["instances"].pred_classes)
print(outputs["instances"].pred_boxes)
'''
# -- load Mask R-CNN model for segmentation
DesignModernityModel = torch.load("DesignModernityModel.pt")

#INPUT_FEATURES = DesignModernityModel.fc.in_features
#linear = nn.linear(INPUT_FEATURES, 5)

DesignModernityModel.eval() # set state of the model to inference

LABELS = ['2000-2003', '2006-2008', '2009-2011', '2012-2014', '2015-2018']
n_labels = len(LABELS)

MEAN = [0.485, 0.456, 0.406]
STD = [0.229, 0.224, 0.225]

carTransforms = transforms.Compose([transforms.Resize(224),
                                    transforms.ToTensor(),
                                    transforms.Normalize(mean=MEAN, std=STD)])

def classifyCar(im):
  im = Image.fromarray(im.astype('uint8'), 'RGB')
  im = carTransforms(im).unsqueeze(0)  # transform and add batch dimension
  with torch.no_grad():
    scores = torch.nn.functional.softmax(DesignModernityModel(im)[0])
  return {LABELS[i]: float(scores[i]) for i in range(n_labels)}

#examples = [[example_img.jpg], [example_img2.jpg]]  # must be uploaded in repo

# create interface for model
interface = gr.Interface(classifyCar, inputs='image', outputs='label', cache_examples=False, title='VW Up or Fiat 500')
interface.launch()