Installation

[bondrewd]

Hello!

First of all, thank you very much for making this work available!

I'm trying to use the code but having some trouble making it work.

First, I tried to install it by following the instructions on my MacOS machine, but then I found out that the environment needed to be Linux.

Then, I tried to install it on a Linux machine, but then found out that it requires a specific CUDA:

Traceback (most recent call last):
  File "run_eval.py", line 262, in <module>
    gamma=gamma, train_lr=train_lr, adamw=adamw, rescale=rescale, round2=False, decode_termini=decode_termini)
  File "run_eval.py", line 194, in iterative_inference
    gen_xyzs = run_inference(trainer, data_name=test_iter_prefix, folder_name=test_save_dir, batch_size=5000)
  File "/home/diego/bkp.bmp/DiAMoNDBack/scripts/utils.py", line 442, in run_inference
    trainer.op_number, batch_size=n, samples = next(sample_dl).cuda()[:n, :]), batches))
  File "/home/diego/bkp.bmp/DiAMoNDBack/scripts/utils.py", line 442, in <lambda>
    trainer.op_number, batch_size=n, samples = next(sample_dl).cuda()[:n, :]), batches))
  File "/home/diego/.local/conda/envs/diffback/lib/python3.7/site-packages/torch/autograd/grad_mode.py", line 26, in decorate_context
    return func(*args, **kwargs)
  File "../denoising_diffusion_pytorch/denoising_diffusion_pytorch_backmap_combined.py", line 702, in sample
    return self.p_sample_loop((batch_size, 1, op_number), samples)
  File "/home/diego/.local/conda/envs/diffback/lib/python3.7/site-packages/torch/autograd/grad_mode.py", line 26, in decorate_context
    return func(*args, **kwargs)
  File "../denoising_diffusion_pytorch/denoising_diffusion_pytorch_backmap_combined.py", line 655, in p_sample_loop
    state = torch.randn(shape, device=device)
RuntimeError: CUDA error: no kernel image is available for execution on the device

I wonder if you could update the conda environment script, make available a CPU version, or at least write down a list of minimum requirements that are needed to try/test/reproduce the work.

Thank you!