neural-dream
neural-dream copied to clipboard
PyTorch implementation of DeepDream algorithm
Hi, I would like to dream over a model created by finetuning pretrained BVLC with your other repo 'dream-creator' but I get this error "ValueError: Model architecture not recognized." Any...
Hi, I'm building a model based on pt_inception5h.pth with the help of the arg "-model_file" but when I run this in "https://github.com/ProGamerGov/neural-dream/tree/dream-creator-support" I get: ``` `Successfully loaded Dream-Creator model: models/1model5h_v1120.pth...
Hey @ProGamerGov! 👋 I really like your implementation! It works so well ant the results are impressive! This pull request makes it possible to run your model inside a Docker...
Hi @ProGamerGov, I've been digging around quite a lot lately to understand the significance of `-use_fft`. I'm trying to implement it myself on my own feature visualization library. What I...
ran it few times it gave beautiful outputs like these:  Then ran it after some time with same settings but now it is giving something like this:  Any...
Like splitting it into frames and then dreaming on each frame? i know that will be slow. But on the whole internet i only found your code that exposes all...
` File "D:\Cancer\neural-dream\neural_dream\models\googlenet\bvlc_googlenet.py", line 352, in forward pool5_7x7_s1 = F.avg_pool2d(inception_5b_output, kernel_size=(7, 7), stride=(1, 1), padding=(0,), ceil_mode=False, count_include_pad=False) RuntimeError: Given input size: (1024x4x4). Calculated output size: (1024x-2x-2). Output size is too...
https://pytorch.org/docs/stable/notes/mps.html
what its the way to import more than one pic with the same config (models and layers) like this -content_image examples/inputs/ "all picture on folder".jpg -output_image ex/ou/"1...2...3....png" -num_iterations 5 -optimizer...
Hello, i know this is an old project, but can you please add a feature for Guide Image ? Here is example of how it's used: https://github.com/google/deepdream/blob/master/dream.ipynb Under "Controlling dreams"..