ROKO
Pytorch function summary 본문
This page is pytorch function summary got used to in paper.
Docs will be updated forever.
TORCH.TOPK
torch.topk(input, k, dim=None, largest=True, sorted=True, *, out=None)
- Returns the k largest elements of the given input tensor along a given dimension.
Parameters:
- input (Tensor) – the input tensor.
- k (int) – the k in “top-k”
- dim (int, optional) – the dimension to sort along
- largest (bool, optional) – controls whether to return largest or smallest elements
- sorted (bool, optional) – controls whether to return the elements in sorted order
torch.scatter(input, dim, index, src) → Tensor
- Out-of-place version of torch.Tensor.scatter_()
TORCH.TENSOR.ITEM
Tensor.item() → number
- contverts Tensor(one element) to plain python number
- Use when you need python number during training(GPU), take value to CPU
- Python number can live only in CPU
TORCH.TENSOR.TOLIST
Tensor.tolist() → list or number
- convert number or list to plain python number
- higjer version of TORCH.TENSOR.ITEM
TORCH.MM
torch.mm(input, mat2, *, out=None) -> Tensor
- Performs a matrix multiplication of the matrices input and mat2.
TORCH.ADDMM
torch.addmm(input, mat1, mat2, *, beta=1, alpha=1, out=None) → Tensor
- \(out = \beta input + \alpha(mat1_i@mat2_i)\)
- If mat1 is a \((n \times m)\) tensor, mat2 is a \((m \times p)\) tensor, then input must be broadcastable with a \((n \times p)\)out will be a \((n \times p)\) tensor tensor and
TORCH.TENSOR.DIM
Tensor.dim() → int
- Returns the number of dimensions of self tensor
TORCH.EQUAL
torch.equal(input, other) → bool
- If two tensor have same size and elements ,then return true, else false
TORCH.CHUNK
torch.chunk(input, chunks, dim=0) → List of Tensors
- split input tensor to the number of chunks(int)
- e.g. torch.chunk([1,2,3,4],2,dim=0) = [[1,2],[3,4]]
TORCH.NN.PARMETER.PARAMETER
torch.nn.parameter.Parameter(data=None, requires_grad=True)
- subclass of tensor
- parameters : data(Tensor), requres_grad(Bool)
TORCH.NN.MODULE.REGISTER_PARAMETER
register_parameter(name, param)
- adds parameter to module
parameters:
- name(str)
- param(Parameter or None) if None cuda ignore this parameters, is not contained operations
'Develop' 카테고리의 다른 글
How to read NVIDIA spec? (2) | 2023.01.11 |
---|---|
How to use tqdm? (0) | 2023.01.11 |
zsh: command not found: conda (0) | 2023.01.05 |
AttributeError: module 'scipy.sparse' has no attribute 'coo_array' (0) | 2023.01.04 |