Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Add experimental coref docs #11291

Merged
merged 17 commits into from
Sep 27, 2022
Merged

Add experimental coref docs #11291

Show file tree
Hide file tree
Changes from 10 commits
Commits
Show all changes
17 commits
Select commit Hold shift + click to select a range
39fe3f5
Add experimental coref docs
polm Aug 10, 2022
c7c269b
Docs cleanup
polm Aug 10, 2022
fbceea1
Apply suggestions from code review
polm Aug 15, 2022
8b2b11f
Apply changes from code review
polm Aug 15, 2022
e74c009
Fix prettier formatting
polm Aug 15, 2022
8d0faa2
Update docs on examples for initialize
polm Aug 16, 2022
6560a01
Add docs for coref scorers
polm Aug 16, 2022
f032fa3
Remove 3.4 notes from coref
polm Aug 18, 2022
77b7d64
Add docs for span cleaner
polm Sep 14, 2022
212791c
Fix docs
polm Sep 14, 2022
cd1d8ec
Fix docs to match spacy-experimental
polm Sep 15, 2022
4165184
More doc fixes
polm Sep 15, 2022
a4cdb76
Formatting
polm Sep 15, 2022
63b3577
Merge branch 'upstream_master' into docs/experimental-coref
svlandeg Sep 15, 2022
b6106eb
Update architectures
polm Sep 16, 2022
54e6f32
Fix links
polm Sep 27, 2022
18bfe73
Fix another link
polm Sep 27, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
92 changes: 86 additions & 6 deletions website/docs/api/architectures.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -11,6 +11,7 @@ menu:
- ['Text Classification', 'textcat']
- ['Span Classification', 'spancat']
- ['Entity Linking', 'entitylinker']
- ['Coreference', 'coref-architectures']
---

A **model architecture** is a function that wires up a
Expand Down Expand Up @@ -587,8 +588,8 @@ consists of either two or three subnetworks:
run once for each batch.
- **lower**: Construct a feature-specific vector for each `(token, feature)`
pair. This is also run once for each batch. Constructing the state
representation is then a matter of summing the component features and
applying the non-linearity.
representation is then a matter of summing the component features and applying
the non-linearity.
- **upper** (optional): A feed-forward network that predicts scores from the
state representation. If not present, the output from the lower model is used
as action scores directly.
Expand Down Expand Up @@ -628,8 +629,8 @@ same signature, but the `use_upper` argument was `True` by default.
> ```

Build a tagger model, using a provided token-to-vector component. The tagger
model adds a linear layer with softmax activation to predict scores given
the token vectors.
model adds a linear layer with softmax activation to predict scores given the
token vectors.

| Name | Description |
| ----------- | ------------------------------------------------------------------------------------------ |
Expand Down Expand Up @@ -920,5 +921,84 @@ A function that reads an existing `KnowledgeBase` from file.
A function that takes as input a [`KnowledgeBase`](/api/kb) and a
[`Span`](/api/span) object denoting a named entity, and returns a list of
plausible [`Candidate`](/api/kb/#candidate) objects. The default
`CandidateGenerator` uses the text of a mention to find its potential
aliases in the `KnowledgeBase`. Note that this function is case-dependent.
`CandidateGenerator` uses the text of a mention to find its potential aliases in
the `KnowledgeBase`. Note that this function is case-dependent.

## Coreference {#coref-architectures tag="experimental"}

A [`CoreferenceResolver`](/api/coref) component identifies tokens that refer to
the same entity. A [`SpanResolver`](/api/span-resolver) component infers spans
from single tokens. Together these components can be used to reproduce
traditional coreference models. You can also omit the `SpanResolver` if working
with only token-level clusters is acceptable.

### spacy.Coref.v1 {#Coref tag="experimental"}
polm marked this conversation as resolved.
Show resolved Hide resolved

> #### Example Config
>
> ```ini
>
> [model]
> @architectures = "spacy.Coref.v1"
polm marked this conversation as resolved.
Show resolved Hide resolved
> distance_embedding_size = 20
> dropout = 0.3
> hidden_size = 1024
> depth = 2
> antecedent_limit = 50
> antecedent_batch_size = 512
>
> [model.tok2vec]
> @architectures = "spacy-transformers.TransformerListener.v1"
> grad_factor = 1.0
> upstream = "transformer"
> pooling = {"@layers":"reduce_mean.v1"}
> ```

The `Coref` model architecture is a Thinc `Model`.

| Name | Description |
| ------------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| `tok2vec` | The [`tok2vec`](#tok2vec) layer of the model. ~~Model~~ |
| `distance_embedding_size` | A representation of the distance between candidates. ~~int~~ |
| `dropout` | The dropout to use internally. Unlike some Thinc models, this has separate dropout for the internal PyTorch layers. ~~float~~ |
| `hidden_size` | Size of the main internal layers. ~~int~~ |
| `depth` | Depth of the internal network. ~~int~~ |
| `antecedent_limit` | How many candidate antecedents to keep after rough scoring. This has a significant effect on memory usage. Typical values would be 50 to 200, or higher for very long documents. ~~int~~ |
| `antecedent_batch_size` | Internal batch size. ~~int~~ |
| **CREATES** | The model using the architecture. ~~Model[List[Doc], Floats2d]~~ |

### spacy.SpanResolver.v1 {#SpanResolver tag="experimental"}
polm marked this conversation as resolved.
Show resolved Hide resolved

> #### Example Config
>
> ```ini
>
> [model]
> @architectures = "spacy.SpanResolver.v1"
> hidden_size = 1024
> distance_embedding_size = 64
> conv_channels = 4
> window_size = 1
> max_distance = 128
> prefix = "coref_head_clusters"
>
> [model.tok2vec]
> @architectures = "spacy-transformers.TransformerListener.v1"
> grad_factor = 1.0
> upstream = "transformer"
> pooling = {"@layers":"reduce_mean.v1"}
> ```

The `SpanResolver` model architecture is a Thinc `Model`.

| Name | Description |
| ------------------------- | ----------------------------------------------------------------------------------------------------------------------------- |
| `tok2vec` | The [`tok2vec`](#tok2vec) layer of the model. ~~Model~~ |
| `distance_embedding_size` | A representation of the distance between two candidates. ~~int~~ |
| `dropout` | The dropout to use internally. Unlike some Thinc models, this has separate dropout for the internal PyTorch layers. ~~float~~ |
polm marked this conversation as resolved.
Show resolved Hide resolved
| `hidden_size` | Size of the main internal layers. ~~int~~ |
| `conv_channels` | The number of channels in the internal CNN. ~~int~~ |
| `window_size` | The number of neighboring tokens to consider in the internal CNN. `1` means consider one token on each side. ~~int~~ |
| `max_distance` | The longest possible length of a predicted span. ~~int~~ |
| `prefix` | The prefix that indicates spans to use for input data. ~~string~~ |
| **CREATES** | The model using the architecture. ~~Model[List[Doc], TupleFloats2d]~~ |
Loading