### 3 releases

0.1.5-alpha.0 | Mar 25, 2023 |
---|---|

0.1.4-alpha.0 | Mar 3, 2023 |

0.1.3-alpha.0 | Mar 2, 2023 |

**BSD-3-Clause**

700KB

5K
SLoC

The

crate provides an operator
for accumulating values in networked digital
signal processing and deep learning
computations. This crate defines the
`caffe2op-accum`

which performs accumulation on the
input tensors according to a specified axis. The
resulting tensor is returned as the output. The
`AccumulateOp`

token is used to refer to the output
tensor.`Accumulated`

**Note: This crate is currently being translated from C++ to Rust, and some function bodies may still be in the process of translation.**

The operation performed by

can be
represented mathematically as follows:`AccumulateOp`

If the input tensor is of shape

and the accumulation axis is `[`a`,` b`,` c`,` `...``,` m`,` n`]`

, then the
output tensor will be of shape `k`

where `[`a`,` b`,` c`,` `...``,` l`,` n`]`

is the size of the tensor along the
accumulation axis. The elements of the output
tensor are calculated as follows:`l`

`output``[`i1`,` i2`,` `...``,` il`,` j`]`
`=` `sum``(`input`[`i1`,` i2`,` `...``,` i_`{`k`-``1``}``,` l`,` i_`{`k`+``1``}``,` `...``,` i_`{`m`}``,` j`]``)`

where the sum is over all

along the
accumulation axis.`l`

The

crate also provides tokens
such as `caffe2op-accum`

, `Accumulation`

,
`accumulates`

, `accumulations`

, `depends`

, `fiddles`

,
and `interim`

, which are used within the
implementation of the accumulation operation.`reshaped`

#### Dependencies

~**35MB**

~383K SLoC