v1.2.0

1.2.0 (2023-01-12)

🔆 Highlights

Data labelling and curation with similarity search

Since 1.2.0 Argilla supports adding vectors to Argilla records which can then be used for finding the most similar records to a given one. This feature uses vector or semantic search combined with more traditional search (keyword and filter based).

View record info

You can now find all record details and fields which can be useful for bookmarking, copy/pasting, and making ES queries

View record timestamp

You can now see the timestamp associated with the record timestamp (event timestamp) which corresponds to the moment when the record was uploaded or a custom timestamp passed when logging the data (e.g., when the prediction was made when using it for monitoring)

Configure the base path of your Argilla UI (useful for proxies)

See: https://docs.argilla.io/en/latest/getting_started/installation/server_configuration.html#using-a-proxy

Features

• Allow to launch the argilla server in a different base_url (#2080) (63d624d), closes #1914 #1899
• Check es connection on startup with retries (#2141) (7a63bea)
• enable partial record update (#2118) (4ed0d95)
• Improve the dataset_labels metric processing (#1978) (1c3235e), closes #1818
• Include record event_timestamp (#2156) (5b75ade), closes #1911
• Include record info view and remove metadata filter (#2079) (901d45a), closes #1927 #1849
• Raw records scan endpoint (#2102) (1b63d95)
• reuse the same httpx async client instance (#1958) (a70cb6c), closes #1886
• Search: Allow passing raw es query in search query (#2098) (0541798)
• set record timestamp by default (#1970) (309fd9f), closes #1892
• Similarity vector search (#1768) (#1998) (32958f4), closes #1757
• UI: remove mixins to hide scroll bar in drop down (#2000) (95ad9b8), closes #1928

Bug Fixes

• #1912 hide empty menu dropdown (#1981) (d90390b)
• Avoid manipulating DOM (#1895) (6939b28), closes #1765
• catch ImportError for telemetry module (#1989) (25513b7)
• Client: check url underscore only for hostnames (#2185) (ec5726a)
• client: prevent python client response json parse error (#2186) (5549ab0)
• Compute predicted properly for token classification [REINDEX_DATASET_REF] (#1975) (a29a198), closes #1955
• Disable shortcuts for pagination when focus is on an input tag (#1995) (af07f3e), closes #1976
• Migration: Set dynamic to false for old indices (#2167) (15a18d7)
• Prevent show "No result" before data is loaded (#2014) (0799425), closes #1936

Documentation

• Add new tutorial about zeroshot sentiment analysis with GPT-3 (#2011) (d3c43ab)
• added additional explanation for datetime ranges (#2120) (c8c3dc9), closes #2119
• Adds Hugging Face Space deployment guide (#2109) (a7a47c4)
• changed DatasetForTextGeneration to DatasetForText2Text (#2090) (8cde28b), closes #2089
• Fix load docstring example (#2050) (7e2af7f), closes #1951
• fixed typo errors for terminology section (#2025) (1056736)
• include new OG image (#2017) (710ab3f)
• Include og image (#2016) (85442e4)
• Maintain menu position during navigation (#1935) (82c6e08), closes #1864
• New setfit tutorial (#2002) (43c66b2)
• Replace OG image (#2018) (894b273)
• Replace video with image (#1990) (359b637)
• reverted to correct apikey reference (#2136) (f32f2b8), closes #2074

As always, thanks to our amazing contributors!

• Add Azure deployment tutorial (#2124) by @burtenshaw
• Create training-textclassification-activelearning-with-GPU.ipynb (#2020) by @MoritzLaurer