Ye Huang

Changing class name/structure changes program functionality (using TensorFlow)

2

For example: In my project (https://github.com/edwardyehuang/CAR/blob/master/carnet.py), line 163 (instantiate SegManaged). If I create a wrapper class that inherited SegManaged and place it on line 163, e.g. ``` class gbb(SegManaged): pass...

From the source code, I find there are codes that calling the "CleanUp" for each child nodes. So do I still need to explicitly call the "Dispose" for node?

MobileNetv4

9

**Short Description** **Papers** MobileNetV4 - Universal Models for the Mobile Ecosystem **Existing Implementations** https://github.com/tensorflow/models/blob/master/official/vision/modeling/backbones/mobilenet.py **Other Information**

Unstable numeric output for downstream task (moat 4 w/o pos)

1

ckpts : moat4 w/o pos The output from moat4 can easily result in the following layers (e.g., 3x3 conv) having a NaN output. The same issue, at least, does not...

About `window_size` in MOAT

3

I found `window_size` is None in MOAT https://github.com/google-research/deeplab2/blob/eb66c852f86c2add70cf7067dd5430ddb2df3b5f/model/pixel_encoder/moat.py#L347-L360 https://github.com/google-research/deeplab2/blob/eb66c852f86c2add70cf7067dd5430ddb2df3b5f/model/pixel_encoder/moat.py#L405 Is only global attention used for segmentation tasks?

[BUG][all_reduce] INVALID_ARGUMENT: You must feed a value for placeholder tensor

26

``` File "/usr/local/lib/python3.11/dist-packages/keras/src/utils/traceback_utils.py", line 118, in error_handler File "/usr/local/lib/python3.11/dist-packages/keras/src/backend/tensorflow/trainer.py", line 323, in fit File "/usr/local/lib/python3.11/dist-packages/keras/src/backend/tensorflow/trainer.py", line 117, in one_step_on_iterator 3 root error(s) found. (0) INVALID_ARGUMENT: You must feed a value...