Support for ComplexFloat

Python 3.9.7
Wandb 0.13.4
Pytorch 1.12.0+cu116

I am running a neural network over complex valued data and getting the following error:

Traceback (most recent call last):
  File "/home/aclifton/rf_fp/run_training_w_evaluate.py", line 523, in <module>
    run_training_pipeline(tmp_dict)
  File "/home/aclifton/rf_fp/run_training_w_evaluate.py", line 229, in run_training_pipeline
    outputs = rffp_model(**batch)
  File "/home/aclifton/anaconda3/envs/rffp/lib/python3.9/site-packages/torch/nn/modules/module.py", line 1151, in _call_impl
    hook_result = hook(self, input, result)
  File "/home/aclifton/anaconda3/envs/rffp/lib/python3.9/site-packages/wandb/wandb_torch.py", line 110, in <lambda>
    lambda mod, inp, outp: parameter_log_hook(
  File "/home/aclifton/anaconda3/envs/rffp/lib/python3.9/site-packages/wandb/wandb_torch.py", line 105, in parameter_log_hook
    self.log_tensor_stats(data.cpu(), "parameters/" + prefix + name)
  File "/home/aclifton/anaconda3/envs/rffp/lib/python3.9/site-packages/wandb/wandb_torch.py", line 221, in log_tensor_stats
    tmin = flat.min().item()
RuntimeError: "min_all" not implemented for 'ComplexFloat'

I’m not quite sure what to make of it and was wondering if anyone could offer some advice? Thanks in advance for your help!

For those interested, I have opened a github question here: