From f5bcc102686e443e46147b400cec32136a609f75 Mon Sep 17 00:00:00 2001 From: Matthew Honnibal Date: Fri, 21 Aug 2020 15:34:54 +0200 Subject: [PATCH] Update architectures --- website/docs/api/architectures.md | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/website/docs/api/architectures.md b/website/docs/api/architectures.md index 993f25bbc..c74737b66 100644 --- a/website/docs/api/architectures.md +++ b/website/docs/api/architectures.md @@ -315,9 +315,9 @@ to pass in a `get_spans` function that will divide up the `Doc` objects before passing them through the transformer. Your spans are allowed to overlap or exclude tokens. -This layer outputs a `FullTransformerBatch` dataclass. In order to plug the -layer into most architectures, you'll probably need to map the raw transformer -output to token-aligned vectors using a layer such as `trfs2arrays`. +This layer is usually used directly by the `Transformer` component, which +allows you to share the transformer weights across your pipeline. For a layer +that's configured for use in other components, see `Tok2VecTransformer`. | Name | Description |