def get_prediction(model, image_bytes):
try:
tensor = transform_image(image_bytes=image_bytes)
pred = model.forward(tensor)
except Exception:
return 0, "error"
return str(torch.argmax(pred.unsqueeze(0), dim=1).item())
def basic_iterative(model_inp,
image_bytes,
alpha,
epsilon,
num_iter,
device=torch.device('cpu')):
model = get_model(model_inp)
tensor = transform_image(image_bytes=image_bytes)
#outputs = model.forward(tensor)
original_output, perturbed_output, perturbed_image = GAE.basic_iterative(
model, tensor, alpha, epsilon, num_iter, device)
cpudev = torch.device('cpu')
original_output, perturbed_output = original_output.to(
cpudev), perturbed_output.to(cpudev)
perturbed_image, tensor = perturbed_image.to(cpudev), tensor.to(cpudev)
per_tensor = perturbed_image - tensor
or_percentage = torch.nn.functional.softmax(original_output,
dim=1)[0] * 100
_, or_indices = torch.sort(original_output, descending=True)
per_percentage = torch.nn.functional.softmax(perturbed_output,
dim=1)[0] * 100
_, per_indices = torch.sort(perturbed_output, descending=True)
top5_original = [(imagenet_class_index[str(idx.item())][1],
or_percentage[idx.item()].item())
for idx in or_indices[0][:5]]
top5_perturbed = [(imagenet_class_index[str(idx.item())][1],
per_percentage[idx.item()].item())
for idx in per_indices[0][:5]]
perturbed_image = rev_transform(perturbed_image[0])
original_image = rev_transform(tensor[0])
perturbation = rev_transform(per_tensor[0])
del model
return dict(top5_original), dict(
top5_perturbed), perturbed_image, original_image, perturbation
def get_prediction(image_bytes):
try:
tensor = transform_image(image_bytes=image_bytes)
outputs = generate_caption(model, tensor)
except Exception as err:
print(str(err))
return 0, 'error'
return outputs
def get_prediction(image_bytes):
try:
tensor = transform_image(image_bytes=image_bytes)
outputs = model.forward(tensor)
except Exception:
return 0, 'error'
_, y_hat = outputs.max(1)
predicted_idx = str(y_hat.item())
return imagenet_class_index[predicted_idx]
def get_prediction(image_bytes):
try:
tensor = transform_image(image_bytes=image_bytes)
outputs = model.forward(tensor)
except Exception:
return 0, 'error'
_, y_hat = outputs.max(1)
if str(y_hat.item()) == '1':
return "Malignant"
return "Benign"
def get_prediction(image_bytes):
transform_image(image_bytes=image_bytes)
#try:
tensor = transform_image(image_bytes=image_bytes)
model.eval()
with torch.no_grad():
output = model.forward(tensor)
#_, predicted = torch.max(output.data, 1)
probability = F.softmax(output.data, dim=1)
prob, clas = probability.topk(1)
#outputs = model.forward(tensor)
#print(prob,clas)
#except Exception:
# return 0, 'error'
#_, y_hat = outputs.max(1)
#predicted_idx = str(y_hat.item())
#print(clas)
return class_dict[clas[0][0].tolist()], round(prob[0][0].tolist(), 3)
def get_prediction(image_bytes):
try:
tensor = transform_image(image_bytes=image_bytes)
output = model(tensor)
except Exception:
return 0, 'error'
ps = torch.exp(output)
top_p, top_class = ps.topk(1, dim=1)
result = top_class.item()
top_p = round(top_p.item(),3)
return classes[result], top_p
def get_prediction(file):
#try:
input = transform_image(file)
outputs = model(input)
for element in outputs:
index = element.data.numpy().argmax()
#pred=imagenet_class_index[str(index)][1]
print(index)
if index % 2 == 0:
return "normal"
else:
return "abnormal"
def upload_file():
if request.method == 'POST':
if 'file' not in request.files:
return redirect(request.url)
file = request.files.get('file')
if not file:
return
img_bytes = file.read()
file_tensor = transform_image(image_bytes=img_bytes) #######
class_name = get_prediction(file_tensor)
return render_template('result.html', class_name=class_name)
return render_template('index.html')
def upload_file():
if request.method == 'POST':
if 'file' not in request.files:
return redirect(request.url)
file = request.files.get('file')
if not file:
return
img_bytes = file.read()
file_tensor = transform_image(image_bytes=img_bytes) ########
#file_share_ptr = secret_share(file_tensor,workers,hospital) # patient generates secret shares of the data and sends one share to the global agent.
class_name = get_prediction(file_tensor)
return render_template('result.html', class_name=class_name)
return render_template('index.html')
def get_detection(image_bytes):
try:
tensor = transform_image(image_bytes=image_bytes)
outputs = detection_model.forward(tensor)
except Exception:
return 0, 'error'
image = tensor.squeeze(0).mul(255).permute(1, 2, 0).byte().numpy()
x1, y1, x2, y2 = int(outputs[0]['boxes'][:, 0]), int(
outputs[0]['boxes'][:, 1]), int(outputs[0]['boxes'][:, 2]), int(
outputs[0]['boxes'][:, 3])
cv2.rectangle(image, (x1, y1), (x2, y2), (0, 255, 0), 1)
Image.fromarray(image).show() # 用PIL显示图像
return a
def lbfgs(model_inp,
image_bytes,
target,
c,
bin_search_steps,
max_iter,
const_upper,
device=torch.device('cpu')):
model = get_model(model_inp)
tensor = transform_image(image_bytes=image_bytes)
#outputs = model.forward(tensor)
target = torch.tensor([target], dtype=torch.long)
original_output, perturbed_output, perturbed_image = GAE.lbfgs(
model, tensor, target, c, bin_search_steps, max_iter, const_upper,
device)
cpudev = torch.device('cpu')
original_output, perturbed_output = original_output.to(
cpudev), perturbed_output.to(cpudev)
perturbed_image, tensor = perturbed_image.to(cpudev), tensor.to(cpudev)
per_tensor = perturbed_image - tensor
or_percentage = torch.nn.functional.softmax(original_output,
dim=1)[0] * 100
_, or_indices = torch.sort(original_output, descending=True)
per_percentage = torch.nn.functional.softmax(perturbed_output,
dim=1)[0] * 100
_, per_indices = torch.sort(perturbed_output, descending=True)
top5_original = [(imagenet_class_index[str(idx.item())][1],
or_percentage[idx.item()].item())
for idx in or_indices[0][:5]]
top5_perturbed = [(imagenet_class_index[str(idx.item())][1],
per_percentage[idx.item()].item())
for idx in per_indices[0][:5]]
perturbed_image = rev_transform(perturbed_image[0])
original_image = rev_transform(tensor[0])
perturbation = rev_transform(per_tensor[0])
del model
return dict(top5_original), dict(
top5_perturbed), perturbed_image, original_image, perturbation
def predict(image_bytes):
tensor = transform_image(image_bytes=image_bytes)
bboxes, labels, scores = get_predictions(tensor, model)
return bboxes, labels, scores
