v1.1.0

What's New

• Rename EmbeddingGenerator to EmbeddingEncoder
• Fixed serverOptions weren't passed through properly in test cases
• Upgrade to @huggingface/transformers v3.2.4
• Upgrade onnxruntime-node v1.20.1
• Avoid including unused models in docker images (smaller image size)
• Increase probe timeout seconds
• Use worker pool
• Process sentence list with separate model runs
• set default workerTaskTimeout to 60 seconds
• use quantized version (q8) default model
• set default limits.memory to 850M
• set default replicas number to 2
• Add max_length config to model config (configurable via helm config)
• set max_length of default model to 1024 due to excessive memory usage when working on text longer than 2048 (the default model supports up to 8192)
• only use padding for multiple inputs received when encoding the input

Full Changelog: v1.0.0...v1.1.0

v1.1.0-alpha.3

What's New

• Rename EmbeddingGenerator to EmbeddingEncoder
• Fixed serverOptions weren't passed through properly in test cases
• Upgrade to @huggingface/transformers v3.2.4
• Upgrade onnxruntime-node v1.20.1
• Avoid including unused models in docker images (smaller image size)
• Increase probe timeout seconds
• Use worker pool
• Process sentence list with separate model runs
• set default workerTaskTimeout to 60 seconds
• use quantized version (q8) default model
• set default limits.memory to 850M
• set default replicas number to 2
• Add max_length config to model config (configurable via helm config)
• set max_length of default model to 1024 due to excessive memory usage when working on text longer than 2048 (the default model supports up to 8192)
• only use padding for multiple inputs received when encoding the input

Full Changelog: v1.0.0...v1.1.0-alpha.2

v1.1.0-alpha.2

What's New

• Rename EmbeddingGenerator to EmbeddingEncoder
• Fixed serverOptions weren't passed through properly in test cases
• Upgrade to @huggingface/transformers v3.2.4
• Upgrade onnxruntime-node v1.20.1
• Avoid including unused models in docker images (smaller image size)
• Increase probe timeout seconds
• Use worker pool
• Process sentence list with separate model runs
• set default workerTaskTimeout to 60 seconds
• set default limits.memory to 2000M
• set default replicas number to 2

Full Changelog: v1.0.0...v1.1.0-alpha.2

v1.1.0-alpha.1

What's New

• Rename EmbeddingGenerator to EmbeddingEncoder
• Fixed serverOptions weren't passed through properly in test cases
• Upgrade to @huggingface/transformers v3.2.4
• Upgrade onnxruntime-node v1.20.1
• Avoid including unused models in docker images (smaller image size)
• Increase probe timeout seconds
• Use worker pool
• Process sentence list with separate model runs
• set default workerTaskTimeout to 60 seconds
• set default limits.memory to 1100M
• set default replicas number to 2

Full Changelog: v1.0.0...v1.1.0-alpha.1

v1.1.0-alpha.0

What's New

• Rename EmbeddingGenerator to EmbeddingEncoder
• Use non-quantized default model by default for better embedding performance but require more memory
• Fixed serverOptions weren't passed through properly in test cases
• Upgrade to @huggingface/transformers v3.2.4
• Upgrade onnxruntime-node v1.20.1
• Avoid including unused models in docker images
• Increase probe timeout seconds
• Use worker pool
• Process sentence list with separate model runs

Full Changelog: v1.0.0...v1.1.0-alpha.0

v1.0.1

What's New

• Fixed: incorrect the svc target port value

Full Changelog: v1.0.0...v1.0.1

v1.0.1-alpha.0

What's NEW

• Fixed: incorrect the svc target port value

Full Changelog: v1.0.0...v1.0.1-alpha.0

v1.0.0

What's New

• Initial production release

Full Changelog: https://github.com/magda-io/magda-embedding-api/commits/v1.0.0

v1.0.0-alpha.2

What's New

• Bug fixes
• Add support of specifying a different model

Full Changelog: v1.0.0-alpha.1...v1.0.0-alpha.2

v1.0.0-alpha.1

What's New

• Add more helm chart config fields
• Re-install prebuilt binary for onnxruntime-node & sharp during docker build

Full Changelog: v1.0.0-alpha.0...v1.0.0-alpha.1