# tensorplay.autograd.forward_ad API

Source: https://www.tensorplay.cn/docs/api/tensorplay.autograd.forward_ad.html

## Functions 4

[#](#api-tensorplay.autograd.forward_ad.enter_dual_level)

### enter_dual_level

function[Full reference ↗](/docs/generated/tensorplay.autograd.forward_ad.enter_dual_level.html)

```python
tensorplay.autograd.forward_ad.enter_dual_level() → int
```

Enters a new forward grad level and returns its index.

This level can be used to make and unpack dual Tensors to compute
forward gradients.  Levels cannot nest: entering while another level is
active raises, so all forward AD computation happens inside one level.

[#](#api-tensorplay.autograd.forward_ad.exit_dual_level)

### exit_dual_level

function[Full reference ↗](/docs/generated/tensorplay.autograd.forward_ad.exit_dual_level.html)

```python
tensorplay.autograd.forward_ad.exit_dual_level(*, level: int | None = None) → None
```

Exits a forward grad level.

This deletes all the tangents associated with this level.  Only exiting
the most recently entered level is allowed.

[#](#api-tensorplay.autograd.forward_ad.make_dual)

### make_dual

function[Full reference ↗](/docs/generated/tensorplay.autograd.forward_ad.make_dual.html)

```python
tensorplay.autograd.forward_ad.make_dual(tensor: Tensor, tangent: Tensor, *, level: int | None = None) → Tensor
```

Associates a tensor value with its tangent to create a “dual tensor”.

The result is a new tensor aliased to tensor with tangent
attached as its forward gradient.  The tangent can be recovered with
[unpack_dual()](/docs/generated/tensorplay.autograd.forward_ad.unpack_dual.html#tensorplay.autograd.forward_ad.unpack_dual).

Given a function f whose jacobian is J, this computes the
Jacobian-vector product (jvp) between J and a given vector v:

```
>>> # xdoctest: +SKIP("Undefined variables")
>>> with dual_level():
...     inp = make_dual(x, v)
...     out = f(inp)
...     y, jvp = unpack_dual(out)
```

[#](#api-tensorplay.autograd.forward_ad.unpack_dual)

### unpack_dual

function[Full reference ↗](/docs/generated/tensorplay.autograd.forward_ad.unpack_dual.html)

```python
tensorplay.autograd.forward_ad.unpack_dual(tensor: Tensor, *, level: int | None = None) → UnpackedDualTensor
```

Unpacks a dual tensor into its primal value and forward gradient.

Returns (primal, tangent) where primal is a view of tensor’s
primal and tangent is tensor’s tangent (None when tensor
carries no tangent at level).