Doesn't work with CPU only instance

[mhoangvslev]

I ran the train script without --cuda flag and got the following error. I make a colab notebook that runs well on GPU instance but I am trying to accelerate with a CPU instance on GCE.

Loading checkpoint model models/id.pth
Shuffling batches for the following epochs
Selected optimization level O0:  Pure FP32 training.

Defaults for this optimization level are:
enabled                : True
opt_level              : O0
cast_model_type        : torch.float32
patch_torch_functions  : False
keep_batchnorm_fp32    : None
master_weights         : False
loss_scale             : 1.0
Processing user overrides (additional kwargs that are not None)...
After processing overrides, optimization options are:
enabled                : True
opt_level              : O0
cast_model_type        : torch.float32
patch_torch_functions  : False
keep_batchnorm_fp32    : None
master_weights         : False
loss_scale             : 1.0
Traceback (most recent call last):
  File "train.py", line 159, in <module>
    model, optimizer = amp.initialize(model, optimizer, opt_level='O0')
  File "/home/minhhoangdang/miniconda3/envs/a2s/lib/python3.6/site-packages/apex/amp/frontend.py", line 358, in initialize
    return _initialize(models, optimizers, _amp_state.opt_properties, num_losses, cast_model_outputs)
  File "/home/minhhoangdang/miniconda3/envs/a2s/lib/python3.6/site-packages/apex/amp/_initialize.py", line 171, in _initialize
    check_params_fp32(models)
  File "/home/minhhoangdang/miniconda3/envs/a2s/lib/python3.6/site-packages/apex/amp/_initialize.py", line 93, in check_params_fp32
    name, param.type()))
  File "/home/minhhoangdang/miniconda3/envs/a2s/lib/python3.6/site-packages/apex/amp/_amp_state.py", line 32, in warn_or_err
    raise RuntimeError(msg)
RuntimeError: Found param conv.seq_module.0.weight with type torch.FloatTensor, expected torch.cuda.FloatTensor.
When using amp.initialize, you need to provide a model with parameters
located on a CUDA device before passing it no matter what optimization level
you chose. Use model.to('cuda') to use the default device.