models icon indicating copy to clipboard operation
models copied to clipboard

ImportError: cannot import name 'tensor' from 'tensorflow.python.framework ' error fromTensorflow object detection api model_builder_tf2_test.py

Open OscarB11 opened this issue 1 year ago • 3 comments

ImportError: cannot import name 'tensor' from 'tensorflow.python.framework ' error fromTensorflow object detection api model_builder_tf2_test.py

I get this error when running the model_builder_tf2_test.py file after installing the object detection API

I got the file to pass with tensorflow 2.15 but when I tried again in different environment using tensorflow 2.10 for the GPU support this error was shown

How do I get it to pass with tensorflow 2.10?

full error message below

2024-04-22 20:39:25.792930: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'cudart64_110.dll'; dlerror: cudart64_110.dll not found 2024-04-22 20:39:25.793198: I tensorflow/stream_executor/cuda/cudart_stub.cc:29] Ignore above cudart dlerror if you do not have a GPU set up on your machine. Traceback (most recent call last): File "c:\Users\besto\Documents\Local vscode\Tensorflow Object Detection GPU\TFODCourse\Tensorflow\models\research\object_detection\builders\model_builder_tf2_test.py", line 24, in from object_detection.builders import model_builder File "c:\Users\besto\Documents\Local vscode\Tensorflow Object Detection GPU\TFODCourse\tfodgpu\lib\site-packages\object_detection-0.1-py3.9.egg\object_detection\builders\model_builder.py", line 70, in from object_detection.models import ssd_efficientnet_bifpn_feature_extractor as ssd_efficientnet_bifpn File "c:\Users\besto\Documents\Local vscode\Tensorflow Object Detection GPU\TFODCourse\tfodgpu\lib\site-packages\object_detection-0.1-py3.9.egg\object_detection\models\ssd_efficientnet_bifpn_feature_extractor.py", line 35, in from official.legacy.image_classification.efficientnet import efficientnet_model File "c:\Users\besto\Documents\Local vscode\Tensorflow Object Detection GPU\TFODCourse\tfodgpu\lib\site-packages\tf_models_official-2.16.0-py3.9.egg\official\legacy\image_classification\efficientnet\efficientnet_model.py", line 30, in import tensorflow as tf, tf_keras File "c:\Users\besto\Documents\Local vscode\Tensorflow Object Detection GPU\TFODCourse\tfodgpu\lib\site-packages\tf_keras_init_.py", line 3, in from tf_keras import internal File "c:\Users\besto\Documents\Local vscode\Tensorflow Object Detection GPU\TFODCourse\tfodgpu\lib\site-packages\tf_keras_internal__init_.py", line 3, in from tf_keras.internal import backend File "c:\Users\besto\Documents\Local vscode\Tensorflow Object Detection GPU\TFODCourse\tfodgpu\lib\site-packages\tf_keras_internal_\backend_init_.py", line 3, in from tf_keras.src.backend import initialize_variables as initialize_variables File "c:\Users\besto\Documents\Local vscode\Tensorflow Object Detection GPU\TFODCourse\tfodgpu\lib\site-packages\tf_keras\src_init.py", line 21, in from tf_keras.src import applications File "c:\Users\besto\Documents\Local vscode\Tensorflow Object Detection GPU\TFODCourse\tfodgpu\lib\site-packages\tf_keras\src\applications_init_.py", line 18, in from tf_keras.src.applications.convnext import ConvNeXtBase File "c:\Users\besto\Documents\Local vscode\Tensorflow Object Detection GPU\TFODCourse\tfodgpu\lib\site-packages\tf_keras\src\applications\convnext.py", line 28, in from tf_keras.src import backend File "c:\Users\besto\Documents\Local vscode\Tensorflow Object Detection GPU\TFODCourse\tfodgpu\lib\site-packages\tf_keras\src\backend.py", line 35, in from tf_keras.src.engine import keras_tensor File "c:\Users\besto\Documents\Local vscode\Tensorflow Object Detection GPU\TFODCourse\tfodgpu\lib\site-packages\tf_keras\src\engine\keras_tensor.py", line 19, in from tf_keras.src.utils import object_identity File "c:\Users\besto\Documents\Local vscode\Tensorflow Object Detection GPU\TFODCourse\tfodgpu\lib\site-packages\tf_keras\src\utils_init_.py", line 53, in from tf_keras.src.utils.feature_space import FeatureSpace File "c:\Users\besto\Documents\Local vscode\Tensorflow Object Detection GPU\TFODCourse\tfodgpu\lib\site-packages\tf_keras\src\utils\feature_space.py", line 20, in from tf_keras.src.engine import base_layer File "c:\Users\besto\Documents\Local vscode\Tensorflow Object Detection GPU\TFODCourse\tfodgpu\lib\site-packages\tf_keras\src\engine\base_layer.py", line 35, in from tf_keras.src.dtensor import lazy_variable File "c:\Users\besto\Documents\Local vscode\Tensorflow Object Detection GPU\TFODCourse\tfodgpu\lib\site-packages\tf_keras\src\dtensor\lazy_variable.py", line 23, in from tensorflow.python.framework import tensor ImportError: cannot import name 'tensor' from 'tensorflow.python.framework' (c:\Users\besto\Documents\Local vscode\Tensorflow Object Detection GPU\TFODCourse\tfodgpu\lib\site-packages\tensorflow\python\framework_init_.py)

OscarB11 avatar Apr 22 '24 20:04 OscarB11

hi, I also had the same problem, could you please tell me how your problem was solved:)

Hathamwang avatar May 10 '24 02:05 Hathamwang

I didnt exactly solve it but I did find a workaround. Using tensorflow version 2.13 fixed the error but this version doesn't support the GPU.

OscarB11 avatar May 10 '24 23:05 OscarB11

I faced the same issue using tensorflow version 2.10, the latest version that supports GPU accessorization in Windows native.

Chathura-Ranasinghe avatar Jun 01 '24 12:06 Chathura-Ranasinghe