v1.16.0
☀️ Highlights
This release comes with an auto save feature for the UI, an enhanced Argilla CLI app, new keyboard shortcuts for the annotation process in the Feedback Dataset and new integrations for the ArgillaTrainer
.
💾 Auto save
Have you been writing a long corrected text in a TextField
for a completion given by an LLM and you have refreshed the page before submitting it? Well, since this release you are covered! The Argilla UI will save every few seconds the responses given in the annotation form of a FeedbackDataset
. Annotators can partially annotate one record and then come back to finish the annotation process without losing the previous work.
👨🏻💻 More operations directly from the Argilla CLI
The Argilla CLI has been updated to include an extensive list of new commands, from users and datasets management to training models all from the terminal!
⌨️ New keyboard shorcuts for the Feedback Dataset
Now, you can seamlessly navigate within the feedback form using just your keyboard. We've extended the functionality of these shortcuts to cover all types of available questions: Label, Multi-label, Ranking, Rating and Text
QnA, Chat Completion with OpenAI and Sentence Transformers model training now in the ArgillaTrainer
The ArgillaTrainer
doesn't stop getting new features and improvements!
- A new
TrainingTask
has been added for Question and Answering (QnA) - Use a
FeedbackDataset
for fine-tuning an OpenAI model for Chat Completion - New integration with Sentence Transformers for fine-tuning a model for embedding generation
1.16.0
Added
- Added
ArgillaTrainer
integration with sentence-transformers, allowing fine tuning for sentence similarity (#3739) - Added
ArgillaTrainer
integration withTrainingTask.for_question_answering
(#3740) - Added
Auto save record
to save automatically the current record that you are working on (#3541) - Added
ArgillaTrainer
integration with OpenAI, allowing fine tuning for chat completion (#3615) - Added
workspaces list
command to list Argilla workspaces (#3594). - Added
datasets list
command to list Argilla datasets (#3658). - Added
users create
command to create users (#3667). - Added
whoami
command to get current user (#3673). - Added
users delete
command to delete users (#3671). - Added
users list
command to list users (#3688). - Added
workspaces delete-user
command to remove a user from a workspace (#3699). - Added
datasets list
command to list Argilla datasets (#3658). - Added
users create
command to create users (#3667). - Added
users delete
command to delete users (#3671). - Added
workspaces create
command to create an Argilla workspace (#3676). - Added
datasets push-to-hub
command to push aFeedbackDataset
from Argilla into the HuggingFace Hub (#3685). - Added
info
command to get info about the used Argilla client and server (#3707). - Added
datasets delete
command to delete aFeedbackDataset
from Argilla (#3703). - Added
created_at
andupdated_at
properties toRemoteFeedbackDataset
andFilteredRemoteFeedbackDataset
(#3709). - Added handling
PermissionError
when executing a command with a logged in user with not enough permissions (#3717). - Added
workspaces add-user
command to add a user to workspace (#3712). - Added
workspace_id
param toGET /api/v1/me/datasets
endpoint (#3727). - Added
workspace_id
arg tolist_datasets
in the Python SDK (#3727). - Added
argilla
script that allows to execute Argilla CLI using theargilla
command (#3730). - Added
server_info
function to check the Argilla server information (also accessible viarg.server_info
) (#3772).
Changed
- Move
database
commands underserver
group of commands (#3710) server
commands only included in the CLI app whenserver
extra requirements are installed (#3710).- Updated
PUT /api/v1/responses/{response_id}
to replacevalues
stored with receivedvalues
in request (#3711). - Display a
UserWarning
when theuser_id
inWorkspace.add_user
andWorkspace.delete_user
is the ID of an user with the owner role as they don't require explicit permissions (#3716). - Rename
tasks
sub-package tocli
(#3723). - Changed
argilla database
command in the CLI to now be accessed viaargilla server database
, to be deprecated in the upcoming release (#3754). - Changed
visible_options
(of label and multi label selection questions) validation in the backend to check that the provided value is greater or equal than/to 3 and less or equal than/to the number of provided options (#3773).
Fixed
- Fixed
remove user modification in text component on clear answers
(#3775) - Fixed
Highlight raw text field in dataset feedback task
(#3731) - Fixed
Field title too long
(#3734) - Fixed error messages when deleting a
DatasetForTextClassification
(#3652) - Fixed
Pending queue
pagination problems when during data annotation (#3677) - Fixed
visible_labels
default value to be 20 just whenvisible_labels
not provided andlen(labels) > 20
, otherwise it will either be the providedvisible_labels
value orNone
, forLabelQuestion
andMultiLabelQuestion
(#3702). - Fixed
DatasetCard
generation whenRemoteFeedbackDataset
contains suggestions (#3718). - Add missing
draft
status inResponseSchema
as now there can be responses withdraft
status when annotating via the UI (#3749). - Searches when queried words are distributed along the record fields (#3759).
- Fixed Python 3.11 compatibility issue with
/api/datasets
endpoints due to theTaskType
enum replacement in the endpoint URL (#3769).
As always, thanks to our amazing contributors
Full Changelog: v1.15.1...v1.16.0