Deploy pytorch model on webcam - image-processing

Question

I am trying to deploy PyTorch classifier on webcam, but always getting errors, mostly "AttributeError: 'collections.OrderedDict' object has no attribute 'load_state_dict'". The classifier is a binary classifier. Saved the model as .pt file.
Hope for your support to resolve the issue.
Here are the codes I am using:
import numpy as np
import torch
import torch.nn
import torchvision
from torch.autograd import Variable
from torchvision import transforms
import PIL
import cv2
#This is the Label
Labels = { 0 : 'Perfect',
1 : 'Defected'
}
# Let's preprocess the inputted frame
data_transforms = torchvision.transforms.Compose([
torchvision.transforms.Resize(size=(224, 224)),
torchvision.transforms.RandomHorizontalFlip(),
torchvision.transforms.ToTensor(),
torchvision.transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
])
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu") ##Assigning the Device which will do the calculation
model = torch.load("defect_classifier.pt") #Load model to CPU
model.load_state_dict(torch.load("defect_classifier.pt"))
model = model.to(device) #set where to run the model and matrix calculation
model.eval() #set the device to eval() mode for testing
#Set the Webcam
def Webcam_720p():
cap.set(3,1280)
cap.set(4,720)
def argmax(prediction):
prediction = prediction.cpu()
prediction = prediction.detach().numpy()
top_1 = np.argmax(prediction, axis=1)
score = np.amax(prediction)
score = '{:6f}'.format(score)
prediction = top_1[0]
result = Labels[prediction]
return result,score
def preprocess(image):
image = PIL.Image.fromarray(image) #Webcam frames are numpy array format
#Therefore transform back to PIL image
print(image)
image = data_transforms(image)
image = image.float()
#image = Variable(image, requires_autograd=True)
image = image.cuda()
image = image.unsqueeze(0) #I don't know for sure but Resnet-50 model seems to only
#accpets 4-D Vector Tensor so we need to squeeze another
return image #dimension out of our 3-D vector Tensor
#Let's start the real-time classification process!
cap = cv2.VideoCapture(0) #Set the webcam
Webcam_720p()
fps = 0
show_score = 0
show_res = 'Nothing'
sequence = 0
while True:
ret, frame = cap.read() #Capture each frame
if fps == 4:
image = frame[100:450,150:570]
image_data = preprocess(image)
print(image_data)
prediction = model(image_data)
result,score = argmax(prediction)
fps = 0
if result >= 0.5:
show_res = result
show_score= score
else:
show_res = "Nothing"
show_score = score
fps += 1
cv2.putText(frame, '%s' %(show_res),(950,250), cv2.FONT_HERSHEY_SIMPLEX, 2, (255,255,255), 3)
cv2.putText(frame, '(score = %.5f)' %(show_score), (950,300), cv2.FONT_HERSHEY_SIMPLEX, 1,(255,255,255),2)
cv2.rectangle(frame,(400,150),(900,550), (250,0,0), 2)
cv2.imshow("ASL SIGN DETECTER", frame)
if cv2.waitKey(1) & 0xFF == ord('q'):
break
cap.release()
cv2.destroyWindow("ASL SIGN DETECTER")

Answer 1

Dont use this line
model = torch.load("defect_classifier.pt")
instead use model = Your_model_class() , since your model is object of your model class