search

OlafenwaMoses / ImageAI

A python library built to empower developers to build applications and systems with self-contained Computer Vision capabilities
https://www.genxr.co/#products
MIT License
8.58k stars 2.19k forks source link

Error: BatchNormalization in keras #749

Open M-Zubair10 opened 2 years ago

M-Zubair10 commented 2 years ago

I am a newbie and don't have a knowledge of tensorflow, keras I found imageAI as a powerful tool for my simple image processing in python The problem is after i install all of the dependencies and imageAI, I keep getting error no module named BatchNormalization From further research i change a little bit code of mode according to stackoverflow guide i.e, from keras.layers.normalization.batch_normalization import BatchNormalization Now keep getting this error, don't know what to do C:\Users\HP\PycharmProjects\Selenium-master\venv\Scripts\python.exe C:/Users/HP/PycharmProjects/Selenium-master/CaptchaSolutions/Hcaptcha/bin/detection.py 2022-03-24 20:37:30.803138: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'cudart64_110.dll'; dlerror: cudart64_110.dll not found 2022-03-24 20:37:30.803374: 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. 2022-03-24 20:37:34.824834: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'nvcuda.dll'; dlerror: nvcuda.dll not found 2022-03-24 20:37:34.825037: W tensorflow/stream_executor/cuda/cuda_driver.cc:269] failed call to cuInit: UNKNOWN ERROR (303) 2022-03-24 20:37:34.827954: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:169] retrieving CUDA diagnostic information for host: MuhammadZubair 2022-03-24 20:37:34.828200: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:176] hostname: MuhammadZubair 2022-03-24 20:37:34.828693: I tensorflow/core/platform/cpu_feature_guard.cc:151] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations: AVX AVX2 To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags. Traceback (most recent call last): File "C:\Users\HP\PycharmProjects\Selenium-master\CaptchaSolutions\Hcaptcha\bin\detection.py", line 22, in detect('Sample/img1.png') File "C:\Users\HP\PycharmProjects\Selenium-master\CaptchaSolutions\Hcaptcha\bin\detection.py", line 11, in detect detector.loadModel(detection_speed=speed) File "C:\Users\HP\PycharmProjects\Selenium-master\venv\lib\site-packages\imageai\Detection__init__.py", line 190, in loadModel model = resnet50_retinanet(num_classes=80) File "C:\Users\HP\PycharmProjects\Selenium-master\venv\lib\site-packages\imageai\Detection\keras_retinanet\models\resnet.py", line 86, in resnet50_retinanet return resnet_retinanet(num_classes=num_classes, backbone='resnet50', inputs=inputs, *kwargs) File "C:\Users\HP\PycharmProjects\Selenium-master\venv\lib\site-packages\imageai\Detection\keras_retinanet\models\resnet.py", line 69, in resnet_retinanet resnet = keras_resnet.models.ResNet50(inputs, include_top=False, freeze_bn=True) File "C:\Users\HP\PycharmProjects\Selenium-master\venv\lib\site-packages\imageai\Detection\keras_resnet\models_2d.py", line 190, in ResNet50 return ResNet(inputs, blocks, numerical_names=numerical_names, block=keras_resnet.blocks.bottleneck_2d, include_top=include_top, classes=classes, args, **kwargs) File "C:\Users\HP\PycharmProjects\Selenium-master\venv\lib\site-packages\imageai\Detection\keras_resnet\models_2d.py", line 68, in ResNet x = keras_resnet.layers.BatchNormalization(axis=axis, epsilon=1e-5, freeze=freeze_bn, name="bn_conv1")(x) File "C:\Users\HP\PycharmProjects\Selenium-master\venv\lib\site-packages\keras\utils\traceback_utils.py", line 67, in error_handler raise e.with_traceback(filtered_tb) from None File "C:\Users\HP\PycharmProjects\Selenium-master\venv\lib\site-packages\tensorflow\python\autograph\impl\api.py", line 692, in wrapper raise e.ag_error_metadata.to_exception(e) TypeError: Exception encountered when calling layer "bn_conv1" (type BatchNormalization).

in user code:

File "C:\Users\HP\PycharmProjects\Selenium-master\venv\lib\site-packages\imageai\Detection\keras_resnet\layers\_batch_normalization.py", line 17, in call  *
    return super(BatchNormalization, self).call(training=(not self.freeze), *args, **kwargs)

TypeError: dict() got multiple values for keyword argument 'training'

Call arguments received: • args=('tf.Tensor(shape=(None, None, None, 64), dtype=float32)',) • kwargs={'training': 'None'}

Process finished with exit code 1

Any help would be appreciated

celt1125 commented 2 years ago

Hi, I got the same issue and finally solved it. Hope this may help you. I tried install imageai by "pip install imageai --upgrade". This gives me version 2.1.5, which is outdated. In this package, the import "from keras.layers.normalization import BatchNormalization" is not compatible for the new version keras. Next, I tried clone this package and put the "imageai" directory in the same folder with my python code. Then everything works fine. (I have tried to install version 2.1.6 by "pip install imageai==2.1.6", but there are lots of package to be resolved. So maybe clone this package is the best way to do what you want)