Distinguish between "rendered" and "unrendered" Tensors.
There are now three types of `Tensor`:
- `Tensor Value a`: rendered value
- `Tensor Ref a`: rendered reference
- `Tensor Build a` : unrendered value
The extra bookkeeping makes it easier to track (and enforce) which tensors are
rendered or not. For examples where this has been confusing in the past, see
With this change, pure ops look similar to before, returning `Tensor Build`
instead of `Tensor Value`. "Stateful" (monadic) ops are unchanged. For
example:
add :: OneOf [..] t => Tensor v'1 t -> Tensor v'2 t -> Tensor Build t
assign :: (MonadBuild m, TensorType t)
=> Tensor Ref t -> Tensor v'2 t -> m (Tensor Ref t)
The `gradients` function now requires that the variables over which it's
differentiating are pre-rendered:
gradients :: (..., Rendered v2) => Tensor v1 a -> [Tensor v2 a]
-> m [Tensor Value a]
(`Rendered v2` means that `v2` is either a `Ref` or a `Value`.)
Additionally, the implementation of `gradients` now takes care to render every
intermediate value when performing the reverse accumulation. I suspect this
fixes an exponential blowup for complicated expressions.
Add support for logging to tensorboard
Based on @gnezdo's internal version with some differences:
* Uses a pure haskell implementation of EventWriter instead of FFI.
* Special `buildAnd*` functions were dropped in favor of using
`mergeAllSummaries :: Build SummaryTensor` with the normal
`build` function.
