Backfill issue -425 — The AI Toolchain

EDITIONS tail grep head diff uniq

The daily firehose — everything the toolchain shipped today, already filtered.

// HOW THIS ISSUE IS MADE

We read every release from the 174 tools on our watchlist at the source — GitHub and GitLab release notes, vendor release pages and changelogs, project blogs and feeds, vendor press releases, and the source code behind the tag. Bug-fix-only releases and non-product newsroom noise are dropped; what's left is summarized down to the new capability, how to try it, and any screenshots or videos the release itself published. Every entry links to the sources it was built from.

VIEW

ISSUE VIEW full issue

Do you prefer this view?

$ tct list   # 3 tools matched

AI & LLM Tooling

◆ AI Model & Data Infrastructure

Microsoft ONNX Runtime

v1.3.1 NOTES

ONNX Runtime v1.3.1 publishes Java (CPU/GPU) and Node.js API packages to Maven and NPM.

└──▷ GET THIS VERSION

$ git clone --branch v1.3.1 https://github.com/microsoft/onnxruntime.git
# already have the repo? check out this version:
$ git checkout v1.3.1

›Publishes Java CPU inference library com.microsoft.onnxruntime/onnxruntime to Maven Central.
›Publishes Java GPU inference library com.microsoft.onnxruntime/onnxruntime_gpu to Maven Central.
›Publishes onnxruntime Node.js API package to NPM.

Was this useful?

NVIDIA Triton Inference Server

v2.0.0 3 RELEASES · 2020-06-01 → 2020-06-30 NOTES STABLE

Triton Inference Server v2.0.0 brings updated KFServing HTTP/REST and gRPC protocols with new Python and C++ client libraries.

└──▷ GET THIS VERSION

$ git clone --branch v2.0.0 https://github.com/triton-inference-server/server.git
# already have the repo? check out this version:
$ git checkout v2.0.0

›Updates KFServing HTTP/REST and gRPC protocols with corresponding V2 Python and C++ client libraries.
›Adds Jetson JetPack 4.4 support via v2.0.0-jetpack4.4ga.tgz, with TensorFlow 1.15.2, TensorRT 7.1, Custom backends, and ensemble support on edge devices.
›Client libraries and examples for Ubuntu 18.04 are packaged in v2.0.0_ubuntu1804.clients.tar.gz and installable via the triton*.whl Python wheel.
›Custom backend SDK for Ubuntu 18.04 is packaged in v2.0.0_ubuntu1804.custombackend.tar.gz.
›Client SDK is also available as an NGC container image.

└──▷ BREAKING ON UPGRADE

!Migration from Triton V1 to Triton V2 requires significant changes — see the 'Backwards Compatibility' and 'Roadmap' sections of the GitHub README.

2 more releases in this issue · 2020-06-01 → 2020-06-30

v1.14.0 NOTES STABLE

Triton v1.14.0 publishes legacy V1 HTTP/REST and GRPC support on a dedicated branch and NGC container.

└──▷ GET THIS VERSION

$ git clone --branch v1.14.0 https://github.com/triton-inference-server/server.git
# already have the repo? check out this version:
$ git checkout v1.14.0

›Releases legacy V1 HTTP/REST and GRPC endpoints, along with corresponding client libraries, on GitHub branch r20.06-v1 and as NGC container 20.06-v1-py3.

v1.13.0 NOTES STABLE

Triton v1.13.0 updates KFServing HTTP/REST and gRPC protocols and bumps gRPC to 1.24.0

└──▷ GET THIS VERSION

$ git clone --branch v1.13.0 https://github.com/triton-inference-server/server.git
# already have the repo? check out this version:
$ git checkout v1.13.0

›Updates KFServing HTTP/REST and gRPC protocols with corresponding Python and C++ client libraries.
›Upgrades gRPC dependency to version 1.24.0.

└──▷ BREAKING ON UPGRADE

!The Caffe2 backend is deprecated; support for Caffe2 models will be removed in a future release.

Was this useful?

Other / Uncategorized

◆ VECTOR DB RAG

Weaviate

0.22.12 3 RELEASES · 2020-06-17 → 2020-06-26 NOTES STABLE

Weaviate 0.22.12 adds _featureProjection to reduce vector dimensionality for 2D/3D visualization via REST and GraphQL.

└──▷ GET THIS VERSION

$ git clone --branch 0.22.12 https://github.com/weaviate/weaviate.git
# already have the repo? check out this version:
$ git checkout 0.22.12

└──▷ USE IT

Retrieve objects with 3D t-SNE projections in GraphQL to feed a scatter-plot visualization of a large corpus.

graphql

{
  Get {
    Article(limit: 100) {
      title
      _featureProjection(dimensions: 3, algorithm: "tsne", perplexity: 5, learningRate: 25, iterations: 100) {
        vector
      }
    }
  }
}

Quickly fetch objects with default 2D projections via REST without writing a GraphQL query.

$ curl -X GET 'http://localhost:8080/v1/things?include=_featureProjection&limit=100'

›Adds _featureProjection underscore prop to REST GET /v1/{kinds}/?include=_featureProjection and GraphQL Get {} queries, reducing object vectors to lower-dimensional representations (default 2D) for visualization.
›Supports GraphQL parameters for _featureProjection: dimensions (int, default 2), algorithm (string, default tsne), perplexity (int, default min(5, len(results)-1)), learningRate (int, default 25), and iterations (int, default 100).
›Ships t-SNE as the first supported dimensionality-reduction algorithm under _featureProjection, with the underlying algorithm designed to be exchangeable in future releases.

2 more releases in this issue · 2020-06-17 → 2020-06-26

0.22.11 NOTES STABLE

Weaviate 0.22.11 adds _nearestNeighbors underscore prop to REST and GraphQL APIs for neighbor concept discovery.

└──▷ GET THIS VERSION

$ git clone --branch 0.22.11 https://github.com/weaviate/weaviate.git
# already have the repo? check out this version:
$ git checkout 0.22.11

└──▷ TRY IT

Retrieve an object's nearest neighboring concepts inline with a single REST call — useful when investigating semantic similarity without a separate query.

$ curl 'https://<weaviate-host>/v1/things/<id>?include=_nearestNeighbors'

›Adds _nearestNeighbors underscore prop to the single-object REST endpoint (GET /v1/{kind}/{id}) and list endpoint (GET /v1/{kinds}) via ?include=_nearestNeighbors query parameter, surfacing neighboring concept data alongside standard responses.
›Adds _nearestNeighbors{} prop support to GraphQL Get {} queries, allowing nearest-neighbor data to be requested alongside schema-defined props.

0.22.8 NOTES STABLE

Weaviate 0.22.8 adds _classification and _interpretation underscore props to REST and GraphQL (use 0.22.9 instead — this release has a known regression).

└──▷ GET THIS VERSION

$ git clone --branch 0.22.8 https://github.com/weaviate/weaviate.git
# already have the repo? check out this version:
$ git checkout 0.22.8

›Adds _classification underscore prop to REST (?include=_classification) and GraphQL, exposing classification metadata for objects that were subject to a classification — previously only available via the now-deprecated ?meta=true REST parameter, and not available in GraphQL at all.
›Adds _interpretation underscore prop to REST (?include=_interpretation) and GraphQL, exposing vectorization metadata including which words were usable, their weights, and per-concept occurrence frequency from the contextionary; requires contextionary version ...-v0.4.12 or later.
›Deprecates meta?=true/false REST query parameter in favor of explicit ?include=_classification and ?include=_vector underscore props; estimated removal in 0.23.0.

Was this useful?

// END OF TRANSMISSION

The AI Toolchain · curated daily · June 30, 2020
Releases & features from the open-source and commercial AI tools that practitioners actually run.

The AI Toolchain — issue -425, June 30, 2020

Microsoft ONNX Runtime

NVIDIA Triton Inference Server

Weaviate