patrickrainer/openai-api-dotnet

A C#/.NET SDK for accessing the OpenAI GPT-3 API

openai-api-dotnet's Issues

TooManyRequests

Hello.
I use similar to your library to work with OpenAI. When executing a request, it throws an exception about exceeding the request limit, this was not only with your library. Do you know what this could be related to? I have not sent a single request, because it throws an exception. (Working through playground on the OpenAI website itself works well)

Make available as Nuget Package

Streaming

I have the following code that is working fine:
string apiKey = mSafeStorage.DecryptString(Config.ApiKey); mOpenApi = new OpenAIAPI(apiKeys: apiKey, engineName: feature.DefaultFeatureModel); // Engine.Ada);// Config.DefaultEngine); CompletionRequest compReq = new CompletionRequest(promptText, max_tokens: maxTokens, temperature: temp, top_p: topp, numOutputs: null, presencePenalty: presPenalty, frequencyPenalty: freqPenalty, logProbs: null, echo: false, stopSequences: null); CompletionResult result = await mOpenApi.Completions.CreateCompletionAsync(compReq); Debug.WriteLine(result.Completions[0].Text);

But the following code for streaming does not produce anything. It just runs without exception and the result is nothing:
CompletionRequest compReq = new CompletionRequest(promptText, max_tokens: maxTokens, temperature: temp, top_p: topp, numOutputs: null, presencePenalty: presPenalty, frequencyPenalty: freqPenalty, logProbs: null, echo: false, stopSequences: null); var allCompletions = new List<Choice>(); await mOpenApi.Completions.StreamCompletionAsync(compReq, (idx, result) => { foreach (var choice in result.Completions) { Console.WriteLine(choice); } });
Any ideas?

Make Endpoints for Finetuning

https://beta.openai.com/docs/guides/fine-tuning

Should be possible to make Fine Tuning, as described in the documentation.