def test_d4_image2label():
x = torch.tensor([[1, 2, 3, 4], [5, 6, 7, 8], [9, 0, 1, 2], [3, 4, 5, 6]]).unsqueeze(0).unsqueeze(0).float()
model = SumAll()
output = tta.d4_image2label(model, x)
expected = int(x.sum())
assert int(output) == expected
def predict(self, image_path):
'''
模型预测返回结果
:param input: 评估传入样例 {"image_path":"image\/172691.jpg"}
:return: 模型预测成功之后返回给系统样例 {"label":"Loxura_atymnus"}
'''
outputs = []
for submodel, transform in zip(self.models, self.val_transforms):
img = Image.open(image_path).convert('RGB')
img = transform(img)
img = img.unsqueeze(0)
img = img.cuda()
with torch.no_grad():
output = tta.d4_image2label(submodel, img)
outputs.append(output)
final = torch.mean(torch.stack(outputs, 0), 0)
_,pred = torch.max(final,1)
pred_name = self.id_name_dict[pred.detach().cpu().item()]
return {"label": pred_name}