Calculate the ACC error during train the model

[VivizSun]

When I am training the model, there is an error in calculating ACC.
Could you provide a way for me to solve it?

Error message:Tensor holds the wrong type, it holds int, but desires to be int64_t.

pred and label type

[xperzy]

Hello, Thanks for the issue, we don't see this problem in our env, could you please provide your paddle version ?
A quick fix is to add a type conversion in the loop where the data is loaded from dataloader, such as:

for batch_id, data in enumerate(dataloader):
    # get data
    images = data[0]
    label = data[1]
    # add data conversion to int64
    label = label.astype('int64')

P.S.,
From the screenshot I saw the main_single_gpu.py may not be the updated code in our implementation, could you please refer to our new impl https://github.com/BR-IDL/PaddleViT/blob/develop/image_classification/SwinTransformer/main_multi_gpu.py

[VivizSun]

Hi xperzy,
My env are as shown below.
Package Version

---

appdirs 1.4.4
astor 0.8.1
certifi 2021.10.8
charset-normalizer 2.0.12
cityscapesScripts 2.2.0
colorama 0.4.4
coloredlogs 15.0.1
cycler 0.11.0
decorator 5.1.1
fonttools 4.33.3
humanfriendly 10.0
idna 3.3
kiwisolver 1.4.2
lmdb 1.3.0
matplotlib 3.5.1
numpy 1.19.3
opencv-python 4.5.5.64
packaging 21.3
paddlepaddle-gpu 2.2.2.post112
Pillow 9.1.0
pip 21.2.4
protobuf 3.20.1
pyparsing 3.0.8
pyquaternion 0.9.9
pyreadline 2.1
python-dateutil 2.8.2
PyYAML 6.0
requests 2.27.1
setuptools 61.2.0
six 1.16.0
tqdm 4.64.0
typing 3.7.4.3
typing_extensions 4.2.0
urllib3 1.26.9
wheel 0.37.1
wincertstore 0.2
yacs 0.1.8

---

P.S.,
I use the SwinTransformer model for image classification.

[xperzy]

@VivizSun Is your problem sovled by using : label = label.astype('int64') ?

[VivizSun]

@xperzy Yes, thank you very much.