package

ops

Implements various ops used when building a graph.

Although the following modules provide a lot of the ops you want when building a graph, you can also use functions in Graph to add constant values, such as constant(), vector(), and scalar().

The Symbol type (returned by all ops) also implements various dunder methods to support operations between symbols, such as + add, * multiply, and @ matmul, plus convenience methods such as reshape() and swapaxes().

Modules

• ​casting: Ops that modify the shape or data type of a symbolic tensor.
• ​complex: Ops that work with complex types.
• ​convolution: Ops commonly used in convolutional networks.
• ​custom_ops: Helpers for building custom ops.
• ​elementwise: Ops that perform element-wise computations/comparisons.
• ​linalg: Ops that perform linear algebra.
• ​lists: Ops that create lists.
• ​quantized_ops: Optimized quantized operators.
• ​reduction: Ops that accumulate or reduce a tensor along an axis.
• ​slicing: Ops that slice, index, stack, concat etc.