Skip to main contentCrate glowstick_burn
Source pub use tensor::Error;pub use tensor::Tensor;
- op
- tensor
- argmax
- Returns the indices of the maximum values along a specified dimension.
- cat
- Concatenates the given tensors along a specified dimension.
A dynamic dimension must be provided for the return type.
- expand
- Expands the lefthand tensor to the shape of the provided righthand tensor
or shape.
- flatten
- Flattens the given tensor from the specified start dimension to the end
dimension.
- gather
- Gathers the elements from a tensor at the provided indices along a specified dimension.
- log_softmax
- Applies the log softmax function to a tensor along the specified dimension.
- matmul
- Performs matrix multiplication of the lefthand tensor and righthand tensor(s).
- mean_dim
- Computes the mean of a tensor along a specified dimension, resulting in a tensor with size
U1 at that dimension. - narrow
- Narrows a tensor at the specified dimension from start index to length.
- reshape
- Reshapes a tensor to the specified dimensions.
- reshape_tys
- reshape_val
- softmax
- Applies the softmax function to a tensor along the specified dimension.
- sort_descending_with_indices
- Applies the sort-descending operation with indices to a tensor along the specified dimension.
- squeeze
- Squeezes the specified dimensions from a tensor.
- squeeze_next
- transpose
- Swaps the dimensions of a tensor.
- tril_mask
- Creates a lower triangular mask of the specified shape.
- unsqueeze
- Unsqueezes a tensor at the specified dimension(s).
- var_mean
- Computes the variance and mean of a tensor along the specified dimension.