Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

TFLite ConvTranspose3D implemented typo #68319

Open
RunnerZhong opened this issue May 21, 2024 · 1 comment
Open

TFLite ConvTranspose3D implemented typo #68319

RunnerZhong opened this issue May 21, 2024 · 1 comment
Assignees
Labels
comp:lite TF Lite related issues TF 2.16 type:bug Bug

Comments

@RunnerZhong
Copy link

Issue type

Bug

Have you reproduced the bug with TensorFlow Nightly?

Yes

Source

source

TensorFlow version

tf 2.16

Custom code

Yes

OS platform and distribution

Ubuntu 22.04

Mobile device

No response

Python version

No response

Bazel version

No response

GCC/compiler version

No response

CUDA/cuDNN version

No response

GPU model and memory

No response

Current behavior?

in tflite convtranspose3d optimized kernel, pad depth seems typo error
image

And also compared with same convtransposed3d with torch interface, tflite have mismatch

Standalone code to reproduce the issue

in tflite convtranspose3d optimized kernel, pad depth seems typo error
![image](https://github.com/tensorflow/tensorflow/assets/30307463/39bdf859-f236-48f9-b02a-c8231d11cb88)

here, seems 
const int spatial_dim_1_padding_after =
      params.padding_values.depth + params.padding_values.depth_offset;

Relevant log output

No response

@sawantkumar
Copy link

Hi @RunnerZhong ,

Let me verify the typo and I will get back to you.

@sawantkumar sawantkumar removed the WIP label Jun 11, 2024
@Venkat6871 Venkat6871 added comp:lite TF Lite related issues TF 2.16 labels Jun 14, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
comp:lite TF Lite related issues TF 2.16 type:bug Bug
Projects
None yet
Development

No branches or pull requests

3 participants