Swall0w / torchstat

Model analyzer in PyTorch

Swall0w/torchstat Issues

Numpy does not support dtype "long" anymore in the latest version
Updated 4 months ago3
fail to build for python 3.10
Updated 10 months ago2
What exactly does `query_granularity` parameter represent?
Updated a year ago
GFLOPs of ResNet50
Updated a year ago2
Need new Feature Got unsupported ScalarType QUInt8
Updated a year ago
AssertionError for the more-than-two-dimensional input of nn.Linear
Updated a year ago6
license in setup.py needs update
Updated a year ago
itemsize = input[0].detach().numpy().itemsize AttributeError: 'list' object has no attribute 'detach'
Updated a year ago1
which param represents GPU memory?
Updated a year ago
multiple inputs
Updated a year ago5
Bias not taken into account in compute_Linear_madd and compute_Linear_flops?
Updated a year ago
Some layer is not supported!
Updated a year ago2
Graph neural network support.
Updated 2 years ago
AttributeError: 'tuple' object has no attribute 'size'
Updated 2 years ago1
RuntimeError: Input type (torch.FloatTensor) and weight type (torch.cuda.FloatTensor) should be the same
Closed 5 years ago7
When does it support torch.nn.functional?
Updated 2 years ago
For 3D Networks?
Updated 2 years ago3
Not work for EfficientNet
Updated 2 years ago3
when i run the module，i got this error
Updated 2 years ago4
How does torchstat compute total memory?
Updated 3 years ago2
‘Total MAdd’ and 'Total Flops' results should exchange with each other
Updated 3 years ago12
not work in jupyter notebook?
Updated 3 years ago1
Spiking Neural Network support?
Updated 3 years ago
[BUG REPORT] total memory_wirte and total memory_read wrong.
Updated 3 years ago
something about memory
Updated 3 years ago
Why does torchstat calculate twice as many macs as flops?
Updated 3 years ago1
analysis nn.Linear() is error
Updated 3 years ago
Bug Report: assert module.__class__ in self._origin_call
Updated 3 years ago
please use rep:https://github.com/Lyken17/pytorch-OpCounter
Updated 4 years ago
about analysis of single layer
Updated 4 years ago
is cuda() type data are not added to the isinstance list?
Updated 4 years ago1
【bug】installing torchstat causes model(tensor) doesn't work!
Updated 4 years ago2
need to download the lastest torch?
Updated 4 years ago
stat doesn't generalize for non-image inputs
Updated 4 years ago4
The input shape
Updated 4 years ago2
How to use for my model with arguments?
Updated 5 years ago
How to use this tool if inputs are not only one image?
Updated 5 years ago5
ConvTranspose2d is not supported
Closed 5 years ago
AttributeError: 'torch.Size' object has no attribute 'numel'
Updated 5 years ago1
AttributeError: 'torch.Size' object has no attribute 'numel'
Closed 6 years ago8
I want to ask a naive question?
Updated 5 years ago
A possible solution to give gpu time duration statistics
Updated 6 years ago
[feature]: compute theoretical amount of memory read and write
Closed 6 years ago
PReLU and ELU was not add to the list into isinstance
Closed 6 years ago1