tensorflow - Pre_trained model work well on ResNet, InceptionNet but unable to run on VGG16 and VGG19
问题描述
I got this trouble when applying object classification with some pre-trained model. This code works on ResNet and Inception, however it turned to have some problem with cudnn when I used VGG16 or VGG19.
I run my code in conda virtual environment which has tensorflow-gpu=2.2.0, cuda=10.1, cudnn=7.6.5.
My cudnn of my OS is 8.0.4. Could this be a problem??? I worked well for many models with this system but not this case.
Here is my code:
ap = argparse.ArgumentParser()
ap.add_argument("-i", "--image", required=True,
help="path to the input image")
ap.add_argument("-model", "--model", type=str, default="vgg16",
help="name of pre-trained network to use")
args = vars(ap.parse_args())
MODELS = {
"vgg16": VGG16,
"vgg19": VGG19,
"inception": InceptionV3,
"xception": Xception, # TensorFlow ONLY
"resnet": ResNet50
}
if args["model"] not in MODELS.keys():
raise AssertionError("The --model command line argument should "
"be a key in the `MODELS` dictionary")
inputShape = (224, 224)
preprocess = imagenet_utils.preprocess_input
if args["model"] in ("inception", "xception"):
inputShape = (299, 299)
preprocess = preprocess_input
Network = MODELS[args["model"]]
model = Network(weights="imagenet")
#model = Network()
model.summary()
image = load_img(args["image"], target_size=inputShape)
image = img_to_array(image)
image = np.expand_dims(image, axis=0)
image = preprocess(image)
preds = model.predict(image)
P = imagenet_utils.decode_predictions(preds)
for (i, (imagenetID, label, prob)) in enumerate(P[0]):
print("{}. {}: {:.2f}%".format(i + 1, label, prob * 100))
Here is the log message:
2020-11-08 11:14:31.324751: E tensorflow/stream_executor/cuda/cuda_dnn.cc:329] Could not create cudnn handle: CUDNN_STATUS_INTERNAL_ERROR
2020-11-08 11:14:31.334392: E tensorflow/stream_executor/cuda/cuda_dnn.cc:329] Could not create cudnn handle: CUDNN_STATUS_INTERNAL_ERROR
Traceback (most recent call last):
File "Classify_keras_applications.py", line 92, in <module>
preds = model.predict(image)
File "/home/phat/anaconda3/envs/DL/lib/python3.8/site-packages/tensorflow/python/keras/engine/training.py", line 88, in _method_wrapper
return method(self, *args, **kwargs)
File "/home/phat/anaconda3/envs/DL/lib/python3.8/site-packages/tensorflow/python/keras/engine/training.py", line 1268, in predict
tmp_batch_outputs = predict_function(iterator)
File "/home/phat/anaconda3/envs/DL/lib/python3.8/site-packages/tensorflow/python/eager/def_function.py", line 580, in __call__
result = self._call(*args, **kwds)
File "/home/phat/anaconda3/envs/DL/lib/python3.8/site-packages/tensorflow/python/eager/def_function.py", line 650, in _call
return self._concrete_stateful_fn._filtered_call(canon_args, canon_kwds) # pylint: disable=protected-access
File "/home/phat/anaconda3/envs/DL/lib/python3.8/site-packages/tensorflow/python/eager/function.py", line 1661, in _filtered_call
return self._call_flat(
File "/home/phat/anaconda3/envs/DL/lib/python3.8/site-packages/tensorflow/python/eager/function.py", line 1745, in _call_flat
return self._build_call_outputs(self._inference_function.call(
File "/home/phat/anaconda3/envs/DL/lib/python3.8/site-packages/tensorflow/python/eager/function.py", line 593, in call
outputs = execute.execute(
File "/home/phat/anaconda3/envs/DL/lib/python3.8/site-packages/tensorflow/python/eager/execute.py", line 59, in quick_execute
tensors = pywrap_tfe.TFE_Py_Execute(ctx._handle, device_name, op_name,
tensorflow.python.framework.errors_impl.UnknownError: Failed to get convolution algorithm. This is probably because cuDNN failed to initialize, so try looking to see if a warning log message was printed above.
[[node vgg19/block1_conv1/Conv2D (defined at Classify_keras_applications.py:92) ]] [Op:__inference_predict_function_763]
Function call stack:
predict_function
解决方案
Have you checked this issue: https://github.com/tensorflow/tensorflow/issues/34888
They mention to add this code on the top of your code:
import tensorflow as tf
gpus= tf.config.experimental.list_physical_devices('GPU')
tf.config.experimental.set_memory_growth(gpus[0], True)
This will not allocate all memory of your GPU at once, but it will increase as your model grow. BUT, I bet that VGGx doesn't fit into your GPU memory, and even with this extra code, I don't think it will fit anyway.
For reference, check this doc:
- VGG16: 528 MB
- VGG19: 549 MB
And:
- ResNet50: 98MB
- InceptionV3: 92MB
VGGx are 5 times bigger than the other ones
推荐阅读
- freeswitch - 使用数据库的 Freeswitch 用户目录
- github - 如何根据分支设置工作流环境变量
- angular - 'any[]' 类型的参数不能分配给'string' 类型的参数
- angular - 如何在.reduce中返回两个值?
- listview - 使用 searchview xamarin.android 使用适配器过滤列表视图
- python - 如何附加到正在运行的 Popen 进程
- linux - 使用 select() 和在循环中检查之间的区别
- python-3.x - 使用 API 网关非代理 lambda 和二进制输出设置自定义标头
- google-sheets - 没有计算正确的数字或刷新
- javascript - 以编程方式打开 ol-contextmenu