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 RawTask executing docs ad verbatim #395

Merged
merged 65 commits into from
Dec 11, 2023
Merged

Add RawTask executing docs ad verbatim #395

merged 65 commits into from
Dec 11, 2023

Conversation

rmitsch
Copy link
Collaborator

@rmitsch rmitsch commented Dec 5, 2023 •
edited
Loading

Description

Add RawTask executing docs ad verbatim, i. e. the surrounding prompt is minimal and doesn't give any instructions except for "read and reply". This is useful for use cases where each doc has its individual prompt or for question answering.

Corresponding documentation PR

explosion/spaCy#13180

Types of change

New feature.

Checklist

  • I confirm that I have the right to submit this contribution under the project's MIT license.
  • I ran all tests in tests and usage_examples/tests, and all new and existing tests passed. This includes
    • all external tests (i. e. pytest ran with --external)
    • all tests requiring a GPU (i. e. pytest ran with --gpu)
  • My changes don't require a change to the documentation, or if they do, I've added all required information.

rmitsch added 30 commits October 17, 2023 12:28
@rmitsch
Add context length info. Refactor BuiltinTask and models to facilitat…
3aba660 
…e this.
@rmitsch
Merge branch 'develop' into feat/inf-doc-len
5699773 
# Conflicts:
#	spacy_llm/tasks/builtin_task.py
@rmitsch
Add token count estimator plumbing.
4213372
@rmitsch
Add plumbing for mapper and reducer.
f440ca4
@rmitsch
Add ShardMapper prototype.
e47f762
@rmitsch
Integrating mapping into prompt generation workflow.
89a5510
@rmitsch
Update response parsing and component to support sharding (WIP).
086dec9
@rmitsch
Fix shard & prompt flow.
23718fc
@rmitsch
Fix shard & prompt flow.
7ce670d
@rmitsch
Remove todo comments.
0d75ea8
@rmitsch
Fix Anthropic, Cohere, NoOp model tests.
9da7098
@rmitsch
Merge branch 'develop' into feat/inf-doc-len
0cb9afd
@rmitsch
Fix test_llm_pipe().
f368412
@rmitsch
Fix type checking test.
b1f111d
@rmitsch
Fix span parsing tests.
44a2787
@rmitsch
Fix internal tests.
6d8cdc7
@rmitsch
Fix _CountTask.
e712f41
@rmitsch
Fix sentiment and summarization tasks and tests.
985fd68
@rmitsch
Fix Azure connection URL. Fix Model test pings.
98842a2
@rmitsch
Fix Lemma parsing.
b54a3d9
@rmitsch
Start work on doc-to-shard property copying.
9bf365d
@rmitsch
Fix REL doc preprocessing.
dddfaab
@rmitsch
Remove comment on doc attribute handling during sharding, as this is …
3af21b5 
…done by spaCy's slicing directly.
@rmitsch
Add reducer implementations.
fee9ca7
@rmitsch
Implement outstanding task reducers.
e508499
@rmitsch
Resolve merge conflicts.
3218541
@rmitsch
Add shardable/non-shardable LLM task typing distinction. Add support …
c104387 
…for handling both types of tasks. Update tests.
@rmitsch
Merge branch 'develop' into feat/inf-doc-len
2c6d899 
# Conflicts:
#	spacy_llm/tests/tasks/test_rel.py
@rmitsch
Fix EL task.
2502c4d
@rmitsch
Fix EL tokenization and highlighting partially.
03055c5
@rmitsch rmitsch added the feat/task Feature: tasks label Dec 5, 2023
@rmitsch rmitsch self-assigned this Dec 5, 2023
@rmitsch rmitsch mentioned this pull request Dec 5, 2023
Merged
3 tasks
@rmitsch rmitsch added the Test external Run external tests label Dec 5, 2023
@rmitsch rmitsch marked this pull request as ready for review December 5, 2023 12:12
@rmitsch
Copy link
Collaborator Author

rmitsch commented Dec 5, 2023

Not sure about the name, open to suggestions.

rmitsch and others added 7 commits December 7, 2023 15:23
@rmitsch @svlandeg
Update spacy_llm/models/langchain/model.py
e94b356 
Co-authored-by: Sofie Van Landeghem <svlandeg@users.noreply.github.com>
@rmitsch @svlandeg
Update spacy_llm/pipeline/llm.py
e68f5d3 
Co-authored-by: Sofie Van Landeghem <svlandeg@users.noreply.github.com>
@rmitsch
Incorporate feedback.
f40bc88
@rmitsch
Move sharding compatibility warning to component constructor.
ac0559d
@rmitsch @svlandeg
Update spacy_llm/tasks/entity_linker/util.py
1763821 
Co-authored-by: Sofie Van Landeghem <svlandeg@users.noreply.github.com>
@rmitsch @svlandeg
Update spacy_llm/models/hf/base.py
ae2e837 
Co-authored-by: Sofie Van Landeghem <svlandeg@users.noreply.github.com>
@rmitsch
Incorporate feedback.
63367fa
svlandeg
svlandeg reviewed Dec 7, 2023
View reviewed changes
Copy link
Member

@svlandeg svlandeg left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Good idea :-)

spacy_llm/tasks/raw/registry.py Outdated Show resolved Hide resolved
spacy_llm/tasks/raw/registry.py Outdated Show resolved Hide resolved
spacy_llm/tasks/raw/registry.py Outdated Show resolved Hide resolved
spacy_llm/tasks/raw/task.py Outdated Show resolved Hide resolved
spacy_llm/tasks/templates/raw.v1.jinja Outdated Show resolved Hide resolved
@svlandeg
Copy link
Member

svlandeg commented Dec 7, 2023

Not sure about the name, open to suggestions.

Naming, my biggest forte!

VerbatimTask? 😁

rmitsch and others added 5 commits December 8, 2023 15:12
@rmitsch @svlandeg
Update spacy_llm/tasks/raw/registry.py
795f675 
Co-authored-by: Sofie Van Landeghem <svlandeg@users.noreply.github.com>
@rmitsch @svlandeg
Apply suggestions from code review
99fe286 
Co-authored-by: Sofie Van Landeghem <svlandeg@users.noreply.github.com>
@rmitsch
Merge branch 'feat/inf-doc-len' into feat/raw-task
e728e2c
@rmitsch
Fix tests.
4070efa
@rmitsch
Remove boilerplate text in raw template.
281998b
@rmitsch
Copy link
Collaborator Author

rmitsch commented Dec 8, 2023

VerbatimTask? 😁

Hm, not that much of a fan (RawTask isn't any better). Something with "direct", "as-is"...I'll think about it a bit more. Feel free to leave more suggestions 😁

Base automatically changed from feat/inf-doc-len to develop December 11, 2023 14:27
svlandeg
svlandeg approved these changes Dec 11, 2023
View reviewed changes
Copy link
Member

@svlandeg svlandeg left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Good to merge after resolving the merge conflicts!

@rmitsch
Merge branch 'develop' into feat/raw-task
191e42b 
# Conflicts:
#	spacy_llm/tasks/__init__.py
#	spacy_llm/tasks/lemma/registry.py
#	spacy_llm/tasks/lemma/util.py
#	spacy_llm/tests/sharding/test_sharding.py
@rmitsch rmitsch merged commit 7c533cd into develop Dec 11, 2023
10 checks passed
@svlandeg svlandeg deleted the feat/raw-task branch December 11, 2023 15:47
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@svlandeg svlandeg svlandeg approved these changes

Assignees

@rmitsch rmitsch

Labels
feat/new New feature feat/task Feature: tasks Test external Run external tests
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
2 participants
@rmitsch @svlandeg