The feature_adversarial_with_topology_signatures from emmanguyen

feature_adversarial_with_topology_signatures's People

Contributors

Watchers

feature_adversarial_with_topology_signatures's Issues

Test deployment with CUDA in MILA infrastructure

Hi @anoperson,

Please help me verify that a deployment refactor_nips_chirs_et_al.py is enabled with GPUs of MILA.

If possible, could you please let me know about the technical information of machine you have tried. I notice that your first running tests with test_npht_small_train_32dirs.h5 have much higher accuracies rather than my CPU machines and previous trials with K80.

Thanks,

Emma

Test posting an issue on Github and linking to Trello Board

/home/emma/anaconda3/envs/dev_py35_refactor_topo_signatures_cnn/lib/python3.5/site-packages/h5py/__init__.py:36: FutureWarning: Conversion of the second argument of issubdtype from `float` to `np.floating` is deprecated. In future, it will be treated as `np.float64 == np.dtype(float).type`.
  from ._conv import register_converters as _register_converters
/home/emma/Research/GAN/refactor_topo_signatures_cnn/pershombox/_software_backends/resource_handler.py:92: UserWarning: The following backends are not properly configured
hera_wasserstein_dist
perseus
Using stuff dependent on those backends will cause runtime errors.
You can get all errors by calling pershombox.get_backend_cfg_errors().

  warnings.warn(error_text, UserWarning)
Traceback (most recent call last):
  File "play_pytorch_cnn_with_signatures.py", line 12, in <module>
    from src.small_norb.experiments import experiment
  File "/home/emma/Research/GAN/refactor_topo_signatures_cnn/src/small_norb/experiments.py", line 8, in <module>
    from ..sharedCode.experiments import train_test_from_dataset, \
  File "/home/emma/Research/GAN/refactor_topo_signatures_cnn/src/sharedCode/experiments.py", line 15, in <module>
    from chofer_torchex.nn import SLayer
  File "/home/emma/Research/GAN/refactor_topo_signatures_cnn/chofer_torchex/nn/__init__.py", line 1, in <module>
    from .slayer import SLayer
  File "/home/emma/Research/GAN/refactor_topo_signatures_cnn/chofer_torchex/nn/slayer.py", line 3, in <module>
    from torch.tensor import _TensorBase
ImportError: cannot import name '_TensorBase'

Recommend Projects