'Yogi' object has no attribute 'Yogi'

[sailfish009]

Hi, if calling yogi from pytorch-optimizer has some bug ( runtime error, Yogi object has no attribute Yogi).
so at this moment, i am calling yogi.py directly.

#import torch_optimizer as optim      # in second iteration (for statement), it getting error. 
from yogi import Yogi                        # import from yogi.py file (include types.py definition)

for fold, (train_idx, val_idx) in enumerate(...):
    model = Net(...)
    # optim = optim.Yogi(model, parameters(), lr=1e-2, betas=(0.9, 0.999), eps=1e-3, initial_accumulator=1e-6, weight_decay=0)
    optim = Yogi(model, parameters(), lr=1e-2, betas=(0.9, 0.999), eps=1e-3, initial_accumulator=1e-6, weight_decay=0)
    ...

[jettify]

I just tried locally:

In [1]: import torch_optimizer as optim

In [2]: optim.Yogi.__doc__
Out[2]: 'Implements Yogi Optimizer Algorithm.\n    It has been proposed in `Adaptive methods for Nonconvex Optimization`__.\n\n    Arguments:\n        params: iterable of parameters to optimize or dicts defining\n            parameter groups\n        lr: learning rate (default: 1e-2)\n        betas: coefficients used for computing\n            running averages of gradient and its square (default: (0.9, 0.999))\n        eps: term added to the denominator to improve\n            numerical stability (default: 1e-8)\n        initial_accumulator: initial values for first and\n            second moments (default: 1e-6)\n        weight_decay: weight decay (L2 penalty) (default: 0)\n\n    Example:\n        >>> import torch_optimizer as optim\n        >>> optimizer = optim.Yogi(model.parameters(), lr=0.01)\n        >>> optimizer.zero_grad()\n        >>> loss_fn(model(input), target).backward()\n        >>> optimizer.step()\n\n    __ https://papers.nips.cc/paper/8186-adaptive-methods-for-nonconvex-optimization  # noqa\n\n    Note:\n        Reference code: https://github.com/4rtemi5/Yogi-Optimizer_Keras\n    '

working as expected. Also modified MNIST example https://github.com/jettify/pytorch-optimizer/blob/master/examples/mnist.py working as expected. (not counting tests)

If you think problem on library side, can you please create small repro script with so I can debug locally?

[sailfish009]

Here is the kernel what i was using yogi optimizer with it, so much time consuming, i didn't checked it again.
https://www.kaggle.com/nroman/melanoma-pytorch-starter-efficientnet

[jettify]

Unfortunately can not run it myself:

[sailfish009]

i had tested that python script in my local pc, but had to download dataset(~100GB)

[jettify]

Closing for now, feel free to reopen if you have additional information how to reproduce this issue.