Ioannis Siglidis
Ioannis Siglidis
You can use weight information if you add it as float values in the agacency matrix.
@giannisnik
I have no objection in adding this, but please be specific on what breaks. If y is set and not used it may be deceiving for what our method does...
Please make your example reproducible. What is the type of adj_m? What does it look like if you print it?
Can you please provide us with the minimum reproducing example possible? This could also help you possibly debug the code or highlight a problem of this package, if there is...
Can you please provide a reproducible example and detail what your use case is according to the library? We have run multiple experiments in the past with MUTAG and graphlet...
Hi @ragnarlevi. Nice question. So Grakel supports basic parallelism using joblib whenever it's possible. On the other hand for some kernels are written only using numpy vectorized operations, which follow...
Hi, thanks a lot for this really precise answer! For that, I have two small questions: - In the first case did you need to organise training data in a...
This is my script: ``` #!/bin/bash CUDA_VISIBLE_DEVICES=0,1,2,3 python train.py \ --dataset I256_hdf5 --parallel --shuffle --num_workers 8 --batch_size 64 \ --num_G_accumulations 4 --num_D_accumulations 4 \ --num_D_steps 1 --G_lr 1e-4 --D_lr 4e-4...
You have not understood my question. Training a GAN with 512x512 leads to OOM which of course depends on my system. However the official paper claims that it can sample...