Working object detection API installation

TensorFlow General Discussion
,
1

I am trying to train an object detection model using Tensorflow 2.16.1 in a Kaggle notebook. Here is my installation procedure:

!git clone https://github.com/tensorflow/models.git
%cd /kaggle/working/models/research
!protoc ./object_detection/protos/*.proto --python_out=.
!cp ./object_detection/packages/tf2/setup.py .
!python -m pip install --user
!python object_detection/builders/model_builder_tf2_test.py

When using this installation out of the box, the last line fails with
AttributeError: module 'keras._tf_keras.keras.layers' has no attribute 'experimental'

which has to do with an incompatible keras version.

I have also tried other Tensorflow versions to fix the problem. However either installation fails instantly or I am getting the keras error message.

Has anyone a working Tensorflow 2 object detection API installation?

1 Like
3

Ok, I am now one step further thanks to this post:
http://discuss.ai.google.dev/t/keras-experimental-error/23397/9

I had to add these two lines on top of the installation:

import os
os.environ["TF_USE_LEGACY_KERAS"] = "1"

Then the AttributeError disappears.

Now when trying to train my model with model_main_tf2.py, I am getting this error:

ImportError: cannot import name 'estimator' from 'tensorflow.compat.v1' (/root/.local/lib/python3.10/site-packages/tensorflow/_api/v2/compat/v1/__init__.py)

1 Like
4

This post suggests to go back to Tensorflow 2.15.0:
https://github.com/tensorflow/tensorflow/issues/63618#issuecomment-2002577864

So I am adding these lines at the bottom of the installation:

!pip uninstall tensorflow --y
!pip install tensorflow==2.15.0

But now I am running into

AttributeError: module 'tensorflow.python.ops.control_flow_ops' has no attribute 'case'

5

Next attempt is to go further back to Tensorflow 2.13.0 as suggested here:
http://discuss.ai.google.dev/t/attributeerror-module-tensorflow-python-ops-control-flow-ops-has-no-attribute-case/21377/7

But now my GPUs are not detected anymore (which worked well with the later Tensorflow versions). So running the following lines

import tensorflow as tf
print("Num GPUs Available: ", len(tf.config.list_physical_devices('GPU')))

returns 0 intead of 2.

So it seems I am stuck. What a mess!

7

i have totally no idea of how to object detection, but i solved that problem by running tensorflow on docker

9

I managed to find a working installation but I had to do some code modifications to achieve this. Here is my configuration:

import os
os.environ["TF_USE_LEGACY_KERAS"] = "1"

!git clone https://github.com/tensorflow/models.git
%cd /kaggle/working/Tensorflow/models/research
!protoc ./object_detection/protos/*.proto --python_out=.
!cp ./object_detection/packages/tf2/setup.py .
!python -m pip install --user .

!pip install tensorflow==2.15.0

I also had to modify tfexample_decoder.py as described in this post:

Now object detection is running fine with GPU.

I don’t know whether the issue has to do with the Kaggle environment but it would be great to use object detection out of the box with the latest Tensorflow version.

1 Like
11

No, I am having the exact same problem as you are right now, and I’ve been stuck for days trying to figure out and I’m not using kaggle at all. I am becoming insane and I hope this post finally helps me to solve it.

13

Hi johann, can you describe in more detail where you are stuck?

15

Hello, how do you modify tfexample_decoder.py in kaggle?

16

It is decribed in this post:

17

I can use that solution in google colab, but I have another issue that force me to use Kaggle Notebooks. And I dont know how to find it on kaggle

18

I have put the patch version of tfexample_decoder.py into a Kaggle dataset:

Then in my notebook I overwrite the installed version with the patched one like this:

PATCH_DIR = "/kaggle/input/detect-music-symbols-patches"
!cp -f {PATCH_DIR}/tfexample_decoder.py /root/.local/lib/python3.10/site-packages/tf_slim/data

WIth this patch version, model_builder_tf2_test.py runs fine.

Hope this helps…

20

A big thank you to you, Dhoernel, this has really helped me