NLU.DevOps
NLU.DevOps copied to clipboard
Tools supporting continuous integration and deployment for NLU services.
Bumps [Newtonsoft.Json](https://github.com/JamesNK/Newtonsoft.Json) from 12.0.1 to 13.0.2. Release notes Sourced from Newtonsoft.Json's releases. 13.0.2 New feature - Add support for DateOnly and TimeOnly New feature - Add UnixDateTimeConverter.AllowPreEpoch property New feature...
dotnet v2.1 is no longer available. And since it is 2022 and .NET v6 is now in LTS, please upgrade the tool to support the latest GA version of .NET....
I'm using the NLUTest task in Azure pipeline with window latest agent version, recently the task start giving this error Tool 'dotnet-nlu' (version '0.8.1') was successfully installed. "C:\hostedtoolcache\windows\dotnet\dotnet.exe" nlu test...
Hello, I am trying to run the batch test behind corporate proxy. I set HTTP_PROXY env variable, but still get ProxyAuthenticationRequired error. Can you advise on any other settings or...
If a "Retry-After" HTTP response header is returned by LUIS, this change ensures that header is respected. Fixes #333
This change allows users to declaratively specify hierarchical entities in their expected utterance results. For example, a user may declare the following: ```json { "text": "Order a pepperoni pizza", "intent":...
A simple approach to this would be to use the same recursive entity flattening we implemented for LUIS v3 results. However, this would be somewhat lossy as we would not...
We should create an `INLUBatchTestClient` interface for NLU providers that accept a batch of utterances as opposed to individual prediction requests.
Currently, the polling duration is hard-coded in the NLU.DevOps tool. If the HTTP services for LUIS, Dialogflow or Lex are returning a 'Retry-After' header, we should respect that.