INT32 Failed on paddle.metric.accuracy

[mingxu1067]

bug描述 Describe the Bug

paddle.metric.accuracy fails when dtype of label is int32, even documents said it is allowed to be.

Reproduce Python Script:

import paddle

predictions = paddle.to_tensor([[0.2, 0.1, 0.4, 0.1, 0.1], [0.2, 0.3, 0.1, 0.15, 0.25]], dtype='float32')

print("Run with Label INT64")
label_int64 = paddle.to_tensor([[2], [0]], dtype="int64")
result = paddle.metric.accuracy(input=predictions, label=label_int64, k=1)
print(result)
print("INT64 Test Done\n")

print("Run with Label INT32")
label_int32 = paddle.to_tensor([[2], [0]], dtype="int32")
result = paddle.metric.accuracy(input=predictions, label=label_int32, k=1)
print(result)
print("INT32 Test Done\n")

Error Message:

• v2.3.0

• Docker image: paddlepaddle/paddle:2.3.0
• log

Run with Label INT64
Tensor(shape=[1], dtype=float32, place=Place(cpu), stop_gradient=True,
       [0.50000000])
INT64 Test Done

Run with Label INT32
Traceback (most recent call last):
  File "test.py", line 13, in <module>
    result = paddle.metric.accuracy(input=predictions, label=label_int32, k=1)
  File "/usr/local/lib/python3.7/dist-packages/paddle/metric/metrics.py", line 802, in accuracy
    total)
ValueError: (InvalidArgument) The type of data we are trying to retrieve does not match the type of data currently contained in the container.
  [Hint: Expected dtype() == paddle::experimental::CppTypeToDataType<T>::Type(), but received dtype():5 != paddle::experimental::CppTypeToDataType<T>::Type():7.] (at /paddle/paddle/phi/core/dense_tensor.cc:137)
  [operator < accuracy > error]

• v2.2.2

• Docker image: paddlepaddle/paddle:2.2.2
• log

Run with Label INT64
Tensor(shape=[1], dtype=float32, place=CPUPlace, stop_gradient=True,
       [0.50000000])
INT64 Test Done

Run with Label INT32
Traceback (most recent call last):
  File "test.py", line 13, in <module>
    result = paddle.metric.accuracy(input=predictions, label=label_int32, k=1)
  File "/usr/local/lib/python3.7/dist-packages/paddle/metric/metrics.py", line 802, in accuracy
    total)
ValueError: (InvalidArgument) Tensor holds the wrong type, it holds int, but desires to be int64_t.
  [Hint: Expected valid == true, but received valid:0 != true:1.] (at /paddle/paddle/fluid/framework/tensor_impl.h:33)
  [operator < accuracy > error]

其他补充信息 Additional Supplementary Information

No response

[paddle-bot-old]

您好，我们已经收到了您的问题，会安排技术人员尽快解答您的问题，请耐心等待。请您再次检查是否提供了清晰的问题描述、复现代码、环境&版本、报错信息等。同时，您也可以通过查看官网API文档、常见问题、历史Issue、AI社区来寻求解答。祝您生活愉快～

Hi! We've received your issue and please be patient to get responded. We will arrange technicians to answer your questions as soon as possible. Please make sure that you have posted enough message to demo your request. You may also check out the API，FAQ，Github Issue and AI community to get the answer.Have a nice day!

[wangxinxin08]

目前该API只支持INT64的输入，感谢您的反馈，我们评估下是否需要支持INT32的输入

[mingxu1067]

那官方的文件可以考慮先移除INT32的相關訊息，這樣使用者比較不易混淆

label (Tensor)-数据类型为int64，int32。输入为数据集的标签。shape为 [sample_number, 1] 。

[wangxinxin08]

收到，感谢反馈