def init_inference(model_kwargs):
global model
global device
model = CustomModel(**model_kwargs)
model.eval()
if args.trt_module :
from torch2trt import TRTModule
if args.trt_conversion :
model.load_state_dict(torch.load(args.pretrained_model))
model = model.cuda()
x = torch.ones((1, 3, 240, 320)).cuda()
from torch2trt import torch2trt
model_trt = torch2trt(model, [x], max_batch_size=100, fp16_mode=True)
#model_trt = torch2trt(model, [x], max_batch_size=100)
torch.save(model_trt.state_dict(), args.trt_model)
exit()
model_trt = TRTModule()
#model_trt.load_state_dict(torch.load('road_following_model_trt_half.pth'))
model_trt.load_state_dict(torch.load(args.trt_model))
model = model_trt.to(device)
else :
model.load_state_dict(torch.load(args.pretrained_model))
model = model.to(device)
inputs, labels = data
inputs, labels = inputs.to(device), labels.to(device)
outputs = net(inputs)
optimizer.zero_grad() # reset gradients for a new calculation
loss_size = loss(outputs, labels)
loss_size.backward() # back-propagation of the loss
epoch_loss += loss_size.data
optimizer.step(
) # optimizer step based on the back-propagation results
net.eval() # freezes the layers from learning to evaluate the network
correct = 0
total = 0
for i, data in enumerate(img_loader, 0):
inputs, labels = data
inputs, labels = inputs.to(device), labels.to(device)
outputs = net(inputs)
_, predicted = torch.max(outputs.data, 1)
total += labels.size(0) # get the total number of images
correct += (
predicted == labels).sum().item() # get if the prediction is correct
print("Accuracy: {:2f}".format(100 * correct / total))