{"payload":{"feedbackUrl":"https://github.com/orgs/community/discussions/53140","repo":{"id":666163354,"defaultBranch":"main","name":"kernel-memory","ownerLogin":"microsoft","currentUserCanPush":false,"isFork":false,"isEmpty":false,"createdAt":"2023-07-13T21:33:28.000Z","ownerAvatar":"https://avatars.githubusercontent.com/u/6154722?v=4","public":true,"private":false,"isOrgOwned":true},"refInfo":{"name":"","listCacheKey":"v0:1718936023.0","currentOid":""},"activityList":{"items":[{"before":"788f2538c68e3b206eab6f53d0a9e5135b46e4ea","after":"3d34260ae513af48030da9a56aa50b8e0162c6f8","ref":"refs/heads/main","pushedAt":"2024-06-21T06:11:17.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"dluc","name":"Devis Lucato","path":"/dluc","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/371009?s=80&v=4"},"commit":{"message":"Add AWS S3 Document Storage (WIP) (#596)\n\n## Motivation and Context (Why the change? What's the scenario?)\r\n\r\nAllow KM to store documents/files in AWS S3 and compatible services.\r\n\r\n## High level description (Approach, Design)\r\n\r\nAdded new extension for AWS S3 storage, configuration settings, and nuget.\r\n\r\n---------\r\n\r\nCo-authored-by: Konstantine Kalbazov \r\nCo-authored-by: Devis Lucato \r\nCo-authored-by: Devis Lucato ","shortMessageHtmlLink":"Add AWS S3 Document Storage (WIP) (#596)"}},{"before":"b11b2ca35e6c20b6f8ffe59fc272c931cd409d6d","after":"788f2538c68e3b206eab6f53d0a9e5135b46e4ea","ref":"refs/heads/main","pushedAt":"2024-06-21T01:58:44.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"dluc","name":"Devis Lucato","path":"/dluc","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/371009?s=80&v=4"},"commit":{"message":"Update Directory.Build.props","shortMessageHtmlLink":"Update Directory.Build.props"}},{"before":"2a75bca87b67d26295a1be41b5145db920d6eadb","after":"b11b2ca35e6c20b6f8ffe59fc272c931cd409d6d","ref":"refs/heads/main","pushedAt":"2024-06-21T01:56:09.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"dluc","name":"Devis Lucato","path":"/dluc","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/371009?s=80&v=4"},"commit":{"message":"Added support for batch embedding generation. (#531)\n\nAdd support for batch embedding generation with Azure AI and OpenAI.\r\nThe default batch size for Azure is `1` to take into account old deployments.\r\nThe value can be changed via configuration with `MaxEmbeddingBatchSize`,\r\nand overridden during a request with RequesContenxt using \r\n`custom_embedding_generation_batch_size_int`\r\n\r\n---------\r\n\r\nCo-authored-by: Devis Lucato \r\nCo-authored-by: Devis Lucato ","shortMessageHtmlLink":"Added support for batch embedding generation. (#531)"}},{"before":"e4d59452fe1d37f78ea2493153c1ab57a6cf7c66","after":"2a75bca87b67d26295a1be41b5145db920d6eadb","ref":"refs/heads/main","pushedAt":"2024-06-20T00:14:53.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"dluc","name":"Devis Lucato","path":"/dluc","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/371009?s=80&v=4"},"commit":{"message":"Cleanup logger DI + other DI work (#674)\n\n* Align the code approach to how logger is injected, using\r\n`ILoggerFactory`, so that the same log configuration can be passed down\r\nwhen using `new` (Postgres, Qdrant, SK, others)\r\n* Change `SearchClient` visibility from internal to public, allowing to\r\ncreate `MemoryServerless` instances without `KernelMemoryBuilder`\r\n* Add example showing how to create a `MemoryServerless` instance\r\nwithout `KernelMemoryBuilder`.\r\n* Some other cleanup, more for the future, e.g. reducing the use of\r\n`new`\r\n* Add `ToString` override method to `MemoryAnswer`","shortMessageHtmlLink":"Cleanup logger DI + other DI work (#674)"}},{"before":"c652260f64ce12e3fd74d51002ffed70ed83a598","after":"e4d59452fe1d37f78ea2493153c1ab57a6cf7c66","ref":"refs/heads/main","pushedAt":"2024-06-18T17:04:40.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"dluc","name":"Devis Lucato","path":"/dluc","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/371009?s=80&v=4"},"commit":{"message":"Update contributors","shortMessageHtmlLink":"Update contributors"}},{"before":"f1468648620c42aca2724705e8db35fc4bd4f124","after":"c652260f64ce12e3fd74d51002ffed70ed83a598","ref":"refs/heads/main","pushedAt":"2024-06-18T08:47:47.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"dluc","name":"Devis Lucato","path":"/dluc","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/371009?s=80&v=4"},"commit":{"message":"Update Directory.Build.props","shortMessageHtmlLink":"Update Directory.Build.props"}},{"before":"73451d9205b5cc548b4a40583e1e08f9c6cf1b8a","after":"f1468648620c42aca2724705e8db35fc4bd4f124","ref":"refs/heads/main","pushedAt":"2024-06-18T08:47:17.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"dluc","name":"Devis Lucato","path":"/dluc","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/371009?s=80&v=4"},"commit":{"message":"Add Request Context feature: allow to override settings at runtime (#673)\n\n## Motivation and Context (Why the change? What's the scenario?)\r\n\r\nSeveral settings of the solution are hard coded by design, and others\r\ncan be configured but require a service restart to be changed.\r\n\r\nThis PR introduces a Context object that is passed during ingestion and\r\nsearch/ask. The context object is optional and can contain custom\r\nkey-values accessible to handlers and search clients.\r\n\r\nDuring web requests, the context is accessible also via dependency\r\ninjection in those scenarios where a method signature doesn't support\r\nIContext, see `RequestContextProvider`. During ingestion, the context is\r\naccessible through the `DataPipeline` instance.\r\n\r\nThe context allows to override the following settings during a\r\nrequest/during the upload a document, without the need to change the\r\ncode or change the configuration:\r\n\r\n* `custom_partitioning_max_tokens_per_paragraph_int`: the max size of\r\nparagraphs while partitioning a file during the upload\r\n* `custom_partitioning_overlapping_tokens_int`: overlapping tokens while\r\npartitioning a file\r\n* `custom_rag_empty_answer_str`: the answer returned by Ask when no\r\nanswer can be found\r\n* `custom_rag_prompt_str` and `custom_rag_fact_template_str`: prompt\r\nused for RAG, including how facts are injected\r\n* `custom_rag_max_tokens_int`: max number of tokens to generate with the\r\nRAG prompt\r\n* `custom_rag_temperature_float`: temperature used with the RAG prompt\r\n* `custom_rag_nucleus_sampling_float`: nucleus sampling used with the\r\nRAG prompt\r\n* `custom_summary_prompt_str`: prompt used to summarize content\r\n* `custom_summary_target_token_size_int`: size of the summary to\r\ngenerate (best effort)\r\n* `custom_summary_overlapping_tokens_int`: overlapping tokens while\r\ngenerating summaries\r\n\r\nOther keys can be used e.g. when working with custom handlers and custom\r\nclasses.\r\nThe RAG `{{$facts}}` template is now configurable and can include tags\r\nand metadata:\r\n\r\n* `{{$content}}`: text from memory, i.e. chunk of text extracted from\r\nthe source\r\n* `{{$source}}`: name of the source file, or URL of the web page, where\r\nthe content originated.\r\n* `{{$relevance}}`: relevance score of the current chunk of text\r\n* `{{$memoryId}}`: ID of the memory record\r\n* `{{$tags}}`: list of tags, excluding reserved/internal ones\r\n* `{{$tag[X]}}`: tag X value(s), replaced with \"-\" if the value is empty\r\n* `{{$meta[X]}}`: value of memory record payload X field (memory payload\r\nis also known as metadata), replaced with \"-\" if the value is empty\r\n\r\nThe PR includes a few examples.\r\n\r\nSample syntax:\r\n\r\n```csharp\r\nvar context = new RequestContext();\r\n\r\ncontext.SetArg(\"custom_summary_prompt_str\", \"Summarize this: {{$input}}. Summary: \");\r\n\r\ncontext.SetArg(\"custom_summary_overlapping_tokens_int\", 0);\r\n\r\nawait memory.ImportDocumentAsync(\r\n new Document(\"doc1\").AddFile(\"file4-KM-Readme.pdf\"),\r\n steps: Constants.PipelineOnlySummary,\r\n context: context);\r\n```\r\n\r\n```csharp\r\nvar context = new RequestContext();\r\n\r\ncontext.SetArg(\"custom_rag_fact_template_str\", \"=== Last update: {{$meta[last_update]}} ===\\n{{$content}}\\n\");\r\n\r\ncontext.SetArg(\"custom_rag_prompt_str\", \"\"\"\r\n Facts:\r\n {{$facts}}\r\n ======\r\n Given only the timestamped facts above, provide a very short answer, include the relevant dates in brackets.\r\n If you don't have sufficient information, reply with '{{$notFound}}'.\r\n Question: {{$input}}\r\n Answer:\r\n \"\"\");\r\n\r\nvar answer = await s_memory.AskAsync(\"What's Kernel Memory?\", context: context);\r\n```","shortMessageHtmlLink":"Add Request Context feature: allow to override settings at runtime (#673"}},{"before":"c76d5e6ee568bab72eec2e9a7a411af2a3e7a477","after":"73451d9205b5cc548b4a40583e1e08f9c6cf1b8a","ref":"refs/heads/main","pushedAt":"2024-06-18T08:32:06.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"dluc","name":"Devis Lucato","path":"/dluc","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/371009?s=80&v=4"},"commit":{"message":"Azure.Identity update (#667)\n\n## Motivation and Context (Why the change? What's the scenario?)\r\nThe package **Azure.Identity** v1.11.3 has a known moderate severity\r\nvulnerability, https://github.com/advisories/GHSA-m5vv-6r4h-3vj9. This\r\nPR updates the library to v1.12.0","shortMessageHtmlLink":"Azure.Identity update (#667)"}},{"before":"1172d8f85104837714c21bb9a4753ee09c5b54bd","after":"c76d5e6ee568bab72eec2e9a7a411af2a3e7a477","ref":"refs/heads/main","pushedAt":"2024-06-12T01:24:08.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"dluc","name":"Devis Lucato","path":"/dluc","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/371009?s=80&v=4"},"commit":{"message":"Fix reading StopSequences from the config (#660)\n\n## Motivation and Context\r\nCurrently, StopSequences initialized with Array.Empty() are not\r\ndeserialized correctly, resulting in an empty list even when values are\r\nprovided in the configuration.\r\n\r\n## High level description\r\nBy changing the initialization to use new List(), we ensure that the\r\nStopSequences are properly deserialized from the configuration. For more\r\ndetails, I have created a issue:\r\n[dotnet/runtime#103282](https://github.com/dotnet/runtime/issues/103282).","shortMessageHtmlLink":"Fix reading StopSequences from the config (#660)"}},{"before":"837a6aa0e505b21c842b8420e772cc3fed57ca15","after":"1172d8f85104837714c21bb9a4753ee09c5b54bd","ref":"refs/heads/main","pushedAt":"2024-06-11T03:30:04.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"dluc","name":"Devis Lucato","path":"/dluc","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/371009?s=80&v=4"},"commit":{"message":"Add functional tests for the evaluation (#657)\n\n## Motivation and Context (Why the change? What's the scenario?)\r\n\r\nResolve #655 \r\n\r\n## High level description (Approach, Design)\r\n\r\n- [x] Add function tests for the evaluation.\r\n- [x] Fix DI for both evaluator and test set generator\r\n\r\n---------\r\n\r\nCo-authored-by: Devis Lucato ","shortMessageHtmlLink":"Add functional tests for the evaluation (#657)"}},{"before":"1f7690348193a1b9de1c9c2f0a3eab94ee0c43e7","after":"837a6aa0e505b21c842b8420e772cc3fed57ca15","ref":"refs/heads/main","pushedAt":"2024-06-08T09:52:21.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"dluc","name":"Devis Lucato","path":"/dluc","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/371009?s=80&v=4"},"commit":{"message":"Fix #655 - Fix DI for evaluation builders (#656)\n\n## Motivation and Context (Why the change? What's the scenario?)\r\n\r\nResolve #655 \r\n\r\n## High level description (Approach, Design)\r\n\r\nThe builders didn't have the generators or evaluators in service\r\ncollection blocking the resolution of instances.","shortMessageHtmlLink":"Fix #655 - Fix DI for evaluation builders (#656)"}},{"before":"2ed7555c6a3325c68730ef9ba53a06f0f296909d","after":"1f7690348193a1b9de1c9c2f0a3eab94ee0c43e7","ref":"refs/heads/main","pushedAt":"2024-06-06T21:04:56.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"dluc","name":"Devis Lucato","path":"/dluc","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/371009?s=80&v=4"},"commit":{"message":"Update euclidean-distance.md","shortMessageHtmlLink":"Update euclidean-distance.md"}},{"before":"6e720d4364e7c3961d5212ea8422d6270411b7e7","after":"2ed7555c6a3325c68730ef9ba53a06f0f296909d","ref":"refs/heads/main","pushedAt":"2024-06-06T21:04:30.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"dluc","name":"Devis Lucato","path":"/dluc","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/371009?s=80&v=4"},"commit":{"message":"Update dot-product.md","shortMessageHtmlLink":"Update dot-product.md"}},{"before":"133b412302fd088278f92137ec0948be2f5eeea7","after":"6e720d4364e7c3961d5212ea8422d6270411b7e7","ref":"refs/heads/main","pushedAt":"2024-06-06T20:59:19.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"dluc","name":"Devis Lucato","path":"/dluc","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/371009?s=80&v=4"},"commit":{"message":"Update docs plugins (#654)","shortMessageHtmlLink":"Update docs plugins (#654)"}},{"before":"d562122409027fd6c51672025a88ef033f08fae3","after":"133b412302fd088278f92137ec0948be2f5eeea7","ref":"refs/heads/main","pushedAt":"2024-06-06T20:33:27.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"dluc","name":"Devis Lucato","path":"/dluc","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/371009?s=80&v=4"},"commit":{"message":"Fix mathjax docs (#653)\n\nFix docs rendering of mathjax syntax","shortMessageHtmlLink":"Fix mathjax docs (#653)"}},{"before":"75dd02ae5975bc2a149737647ba0df3de23c603b","after":"d562122409027fd6c51672025a88ef033f08fae3","ref":"refs/heads/main","pushedAt":"2024-06-06T19:32:25.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"dluc","name":"Devis Lucato","path":"/dluc","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/371009?s=80&v=4"},"commit":{"message":"Update dot-product.md","shortMessageHtmlLink":"Update dot-product.md"}},{"before":"239c14af6261827efe819c8ad4251166a149c744","after":"75dd02ae5975bc2a149737647ba0df3de23c603b","ref":"refs/heads/main","pushedAt":"2024-06-06T19:30:53.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"dluc","name":"Devis Lucato","path":"/dluc","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/371009?s=80&v=4"},"commit":{"message":"Update euclidean-distance.md","shortMessageHtmlLink":"Update euclidean-distance.md"}},{"before":"a93804af7069980e7235b962594218ee52dcaa3d","after":"239c14af6261827efe819c8ad4251166a149c744","ref":"refs/heads/main","pushedAt":"2024-06-06T19:28:50.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"dluc","name":"Devis Lucato","path":"/dluc","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/371009?s=80&v=4"},"commit":{"message":"Update cosine-similarity.md","shortMessageHtmlLink":"Update cosine-similarity.md"}},{"before":"e56adc53dfca85a29a890e80d5abdf9389c78ddb","after":"a93804af7069980e7235b962594218ee52dcaa3d","ref":"refs/heads/main","pushedAt":"2024-06-06T19:23:33.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"dluc","name":"Devis Lucato","path":"/dluc","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/371009?s=80&v=4"},"commit":{"message":"Update cosine-similarity.md","shortMessageHtmlLink":"Update cosine-similarity.md"}},{"before":"ca25ca2148dfbb4f7e780a063e278acaf95aa91b","after":"e56adc53dfca85a29a890e80d5abdf9389c78ddb","ref":"refs/heads/main","pushedAt":"2024-06-06T19:22:32.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"dluc","name":"Devis Lucato","path":"/dluc","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/371009?s=80&v=4"},"commit":{"message":"Update euclidean-distance.md","shortMessageHtmlLink":"Update euclidean-distance.md"}},{"before":"f6bfbc170695781f2f8f4cafcc53a5d1cd58b1e4","after":"ca25ca2148dfbb4f7e780a063e278acaf95aa91b","ref":"refs/heads/main","pushedAt":"2024-06-06T19:20:10.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"dluc","name":"Devis Lucato","path":"/dluc","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/371009?s=80&v=4"},"commit":{"message":"Update cosine-similarity.md","shortMessageHtmlLink":"Update cosine-similarity.md"}},{"before":"9599d570f1505f84311c747b88159f283b66c42e","after":"f6bfbc170695781f2f8f4cafcc53a5d1cd58b1e4","ref":"refs/heads/main","pushedAt":"2024-06-05T18:43:21.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"dluc","name":"Devis Lucato","path":"/dluc","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/371009?s=80&v=4"},"commit":{"message":"Fix for Postgres not supporting 0x00 on text fields (#650)\n\n## Motivation and Context (Why the change? What's the scenario?)\r\n\r\nPostgres doesn't support null chars in text fields.\r\nSee https://github.com/microsoft/kernel-memory/issues/649\r\n\r\n## High level description (Approach, Design)\r\n\r\nRemove \"0x00\" from chunks content written to Postgres.\r\nNote: do not remove the char from Payload json field.","shortMessageHtmlLink":"Fix for Postgres not supporting 0x00 on text fields (#650)"}},{"before":"9abbc769d87cd9c792e885e77eede3016ab889d5","after":"9599d570f1505f84311c747b88159f283b66c42e","ref":"refs/heads/main","pushedAt":"2024-06-05T02:51:50.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"dluc","name":"Devis Lucato","path":"/dluc","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/371009?s=80&v=4"},"commit":{"message":"Update config.yml","shortMessageHtmlLink":"Update config.yml"}},{"before":"bd181549c02c56bd005d5a58c174ef3fd1a0bdb7","after":"9abbc769d87cd9c792e885e77eede3016ab889d5","ref":"refs/heads/main","pushedAt":"2024-06-05T02:50:00.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"dluc","name":"Devis Lucato","path":"/dluc","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/371009?s=80&v=4"},"commit":{"message":"Delete .github/ISSUE_TEMPLATE/feature.yaml","shortMessageHtmlLink":"Delete .github/ISSUE_TEMPLATE/feature.yaml"}},{"before":"426e5e1b9fba6cc08eade28f252678478e6d5ed2","after":"bd181549c02c56bd005d5a58c174ef3fd1a0bdb7","ref":"refs/heads/main","pushedAt":"2024-06-05T02:37:03.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"dluc","name":"Devis Lucato","path":"/dluc","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/371009?s=80&v=4"},"commit":{"message":"Delete .github/ISSUE_TEMPLATE/question.yaml","shortMessageHtmlLink":"Delete .github/ISSUE_TEMPLATE/question.yaml"}},{"before":"89fe8ff8d358105635032fafea4928d36ba5e582","after":"426e5e1b9fba6cc08eade28f252678478e6d5ed2","ref":"refs/heads/main","pushedAt":"2024-06-05T00:42:39.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"dluc","name":"Devis Lucato","path":"/dluc","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/371009?s=80&v=4"},"commit":{"message":"Update list of contributors","shortMessageHtmlLink":"Update list of contributors"}},{"before":"e10da239126c66e7fa5c096380d7b933784eb687","after":"89fe8ff8d358105635032fafea4928d36ba5e582","ref":"refs/heads/main","pushedAt":"2024-06-04T23:16:48.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"dluc","name":"Devis Lucato","path":"/dluc","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/371009?s=80&v=4"},"commit":{"message":"New version: 0.62.0","shortMessageHtmlLink":"New version: 0.62.0"}},{"before":"9a476099529b2f54b215db7bf22e5c29b19e8357","after":"e10da239126c66e7fa5c096380d7b933784eb687","ref":"refs/heads/main","pushedAt":"2024-06-04T23:14:47.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"dluc","name":"Devis Lucato","path":"/dluc","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/371009?s=80&v=4"},"commit":{"message":"Postgres: improve docs and code handling “db not found” scenario (#602)\n\n## Motivation and Context (Why the change? What's the scenario?)\r\n\r\nHelp configuring Postgres and debugging scenarios where DB name != username,\r\nrequiring the DB name to be specified.\r\nSee https://github.com/microsoft/kernel-memory/issues/598\r\n\r\n## High level description (Approach, Design)\r\n\r\nImprove docs and default settings, adding `Database` param.\r\nHandle specific error with additional error logs to help resolution.","shortMessageHtmlLink":"Postgres: improve docs and code handling “db not found” scenario (#602)"}},{"before":"8c0ad8c144c5b9b2b8f1fa0de72ba54e2aa723a0","after":"9a476099529b2f54b215db7bf22e5c29b19e8357","ref":"refs/heads/main","pushedAt":"2024-06-03T19:00:28.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"dluc","name":"Devis Lucato","path":"/dluc","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/371009?s=80&v=4"},"commit":{"message":"Added bulk embedding support for azure. (#526)\n\nEmbedding in Azure OpenAI (also standard openai) supports bulk embedding\r\nwhere we can reduce the number of calls requesting embedding for more\r\nthan one single piece of text.\r\n\r\nThis is a really needed feature when we index lots of data (extraction\r\nfrom big database).\r\n\r\n---------\r\n\r\nCo-authored-by: Devis Lucato ","shortMessageHtmlLink":"Added bulk embedding support for azure. (#526)"}},{"before":"b64f7f8fa6c23d30553aee673e93a190232b5489","after":"8c0ad8c144c5b9b2b8f1fa0de72ba54e2aa723a0","ref":"refs/heads/main","pushedAt":"2024-06-01T19:54:18.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"dluc","name":"Devis Lucato","path":"/dluc","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/371009?s=80&v=4"},"commit":{"message":"Fix #527 - Use NVarchar(max) db type for tags in SQL Server (#530)\n\n## Motivation and Context (Why the change? What's the scenario?)\r\n\r\nResolve the issue #527 by removing size limit for tags field.\r\n\r\n## High level description (Approach, Design)\r\n\r\nthe merge command used nvarchar(256) field for tags management but it\r\nrequires to remove this limit.","shortMessageHtmlLink":"Fix #527 - Use NVarchar(max) db type for tags in SQL Server (#530)"}}],"hasNextPage":true,"hasPreviousPage":false,"activityType":"all","actor":null,"timePeriod":"all","sort":"DESC","perPage":30,"cursor":"djE6ks8AAAAEa0K6EQA","startCursor":null,"endCursor":null}},"title":"Activity · microsoft/kernel-memory"}