public/Request-TuneFileReview.ps1
|
function Request-TuneFileReview { <# .SYNOPSIS Submits a file to Invoke-TuneChat for improvement suggestions. .DESCRIPTION This function sends the contents of specified files to Invoke-TuneChat for analysis and suggestions for improvement. .PARAMETER FilePath Path to the files to be analyzed for improvements. .PARAMETER Model The model you want to use for the analysis. If not specified, the default model is used. .PARAMETER Prompt Custom prompt. The default asks for 5 improvements with corresponding json examples. .PARAMETER MaxTokens Limit the number of tokens in the response from Invoke-TuneChat. Default is 1024. .EXAMPLE Request-TuneFileReview -FilePath C:\path\to\file.jsonl Sends the content of the specified file to Invoke-TuneChat and displays improvement suggestions. .EXAMPLE Get-ChildItem *.jsonl | Request-TuneFileReview Sends the content of all .jsonl files in the current directory to Invoke-TuneChat and displays improvement suggestions for each. .EXAMPLE Request-TuneFileReview -FilePath C:\path\to\file.jsonl -Model "gpt-4" -MaxTokens 500 Uses the GPT-4 model and limits the response to 500 tokens. #> [CmdletBinding()] param ( [Parameter(Mandatory, ValueFromPipeline)] [ValidateScript({ return (Test-Path $_) })] [System.IO.FileInfo[]]$FilePath, [string]$Model, [string]$Prompt = "I have a JSONL file used for training a chatbot. Here's the content. Pls analyze it and suggest 5 improvements. After your review, please include corresponding examples in JSONL format.", [int]$MaxTokens = 1024 ) process { foreach ($file in $FilePath) { Write-Verbose "Processing $file" $file = (Get-ChildItem $file).FullName $basename = (Get-ChildItem $file).BaseName Write-Host "`r`n--- Results for $basename ---`r`n" $fileContent = Get-Content -Path $file -Raw $Prompt = "$Prompt $fileContent" if (-not $Model) { $Model = Get-TuneModelDefault } $params = @{ Message = $Prompt Model = $Model MaxTokens = $MaxTokens } Invoke-TuneChat @params } } } |