From a9c63f245aea907ece6b83dc0a63bd073f97624d Mon Sep 17 00:00:00 2001 From: Farshid Zavareh Date: Tue, 4 Jul 2023 15:21:03 +1000 Subject: [PATCH] Add release notes for 0.1.0 (#528) --- RELEASE.md | 37 +++++++++++++++++++++++++++++++++++++ 1 file changed, 37 insertions(+) diff --git a/RELEASE.md b/RELEASE.md index da91546dc..1cd767594 100644 --- a/RELEASE.md +++ b/RELEASE.md @@ -1,3 +1,40 @@ +# Release 0.1.0 + +## New features +- Telemetry. Marqo now includes various timing metrics for the `search`, `bulk_search` and `add_documents` endpoints +when the query parameter `telemetry=True` is specified (https://github.com/marqo-ai/marqo/pull/506). The metrics will be +returned in the response body and provide a breakdown of latencies for various stages of the API call. +- Consolidate default device to CUDA when available (https://github.com/marqo-ai/marqo/pull/508). By default, +Marqo now uses CUDA devices for search and indexing if available. +See [here](https://docs.marqo.ai/0.1.0/API-Reference/search/#query-parameters) for more information. This helps ensure +you get the best indexing and search experience without having to explicitly add the device parameter to search and +add_documents calls. +- Model download integrity verification (https://github.com/marqo-ai/marqo/pull/502). Model files are validated and +removed if corrupted during download. This helps ensure that models are not loaded if they are corrupted. + +## Breaking changes +- Remove deprecated `add_or_update_documents` endpoint (https://github.com/marqo-ai/marqo/pull/517). +- Disable automatic index creation. Marqo will no longer automatically create an index if it does not exist +(https://github.com/marqo-ai/marqo/pull/516). +Attempting to add documents to a non-existent index will now result in an error. This helps provide more certainty about +the properties of the index you are adding documents to, and also helps prevent accidental indexing to the wrong index. +- Remove parallel indexing (https://github.com/marqo-ai/marqo/pull/523). Marqo no longer supports server-side parallel +indexing. This helps deliver a more stable and efficient indexing experience. Parallelisation can still be implemented +by the user. + +## Bug fixes and minor changes +- Improve error messages (https://github.com/marqo-ai/marqo/pull/494, https://github.com/marqo-ai/marqo/pull/499). +- Improve API request validation (https://github.com/marqo-ai/marqo/pull/495). +- Add new multimodal search example (https://github.com/marqo-ai/marqo/pull/503). +- Remove autocast for CPU to speed up vectorisation on ARM64 machines (https://github.com/marqo-ai/marqo/pull/491). +- Enhance test stability (https://github.com/marqo-ai/marqo/pull/514). +- Ignore `.kibana` index (https://github.com/marqo-ai/marqo/pull/512). +- Improve handling of whitespace when indexing documents (https://github.com/marqo-ai/marqo/pull/521). +- Update CUDA version to 11.4.3 (https://github.com/marqo-ai/marqo/pull/525). + +## Contributor shout-outs +- Thank you to our 3.1k stargazers! + # Release 0.0.21 ## New features