“”"_impl.UnknownError: Failed to get convolution algorithm. I ran my same code, and got the exact same error : I used the ngc container : nvcr.io/nvidia/tensorflow:19.11-tf2-p圓" Session = InteractiveSession(config=config) _t_session(sess)įor TF2, using NGC container “ nvcr.io/nvidia/tensorflow:19.11-tf2-p圓”: python -m pip freeze | grep -i -e tensorflow -e kerasįrom 1 import ConfigProtoįrom 1 import InteractiveSessionĬonfig.gpu_options.per_process_gpu_memory_fraction = 0.2 Sess = tf.Session(config=tf.ConfigProto(gpu_options=gpu_options)) Gpu_options = tf.GPUOptions(allow_growth=True)
NVIDIA CUDA DOWNLOAD ANACONDA CODE
Try adding this code snippet at the top of your script.įor TF1, using NGC container “ nvcr.io/nvidia/tensorflow:19.10-p圓”: python -m pip freeze | grep -i -e tensorflow -e keras You can set the config to dynamically grow GPU memory as needed, and this way you shouldn’t run out unless the model actually requires more than you have. If you have other processes running using any GPU memory, that might make it run out.
Tensorflow by default allocates almost all of the GPU memory right at the start. I just ran your code and confirmed the model is only using about ~1GB of GPU memory. Model.add(MaxPooling2D(pool_size=(2, 2))) Y_test = _categorical(y_test, num_classes) Y_train = _categorical(y_train, num_classes) # Convert class vectors to binary class matrices. (x_train, y_train), (x_test, y_test) = cifar10.load_data() # The data, split between train and test sets: import kerasįrom import ImageDataGeneratorįrom keras.layers import Dense, Dropout, Activation, Flattenįrom keras.layers import Conv2D, MaxPooling2D The below code is working with CPU, but creates issues when run with GPU. Please find below code which i am trying to run with above mentioned versions of packages.
0 kommentar(er)
