Deep dreamer download
verbose: If this flag is present, the current octave and iteration will be printed in the terminal. print_model: If this flag is present, all the inputs and layers of the model will be printed to the terminal. Default is tensorflow_inception_graph.pb.
tile_size: The desired size of tiles to use. channel: The desired channel of the target layer. output_image: The name of your output image. input_image: The input image for performing DeepDream on. Usage:īasic usage: python pb_dreamer.py -input_image input.pngĪdvanced usage: python pb_dreamer.py -input_image input.png -output_image output.png -octaves 4 -layer mixed4d_3x3_bottleneck_pre_relu -channel 139 -iter 10 -tile_size 512 -model /home/ubuntu/Protobuf-Dreamer/model/tensorflow_inception_graph.pb Parameters: Run the following to download and setup the default model:įor a full list of layers, see here.
Refer to the Tensorflow installation and setup guide for more info: Setup:
#Deep dreamer download install#
Sudo pip install tensorflow or sudo pip install tensorflow-gpu Any "blurriness" caused by resizing a smaller image to a larger size, should disappear after the DeepDream process. In order to control the desired output size, resize your image prior to running pb_dreamer.py. This project was inspired by jnordberg's DreamCanvas project. The tiling code is based on the Tensorflow DeepDream example code. The CPU is slower, but this project should allow anyone to create an image of any size. Tensorflow should be compiled for either the CPU, or GPU depending on what your prefer. A tiled DeepDream project for creating any size of image, on both CPU and GPU.