arabic-audio-reader-worker / scripts /setup_baseer_ocr.ps1
Syncre's picture
Deploy Arabic Audio Reader worker
2e1a095 verified
param(
[string]$Python = "py -3.10"
)
$ErrorActionPreference = "Stop"
$root = Split-Path -Parent (Split-Path -Parent $MyInvocation.MyCommand.Path)
$venv = Join-Path $root ".venv-baseer-ocr"
if (-not (Test-Path $venv)) {
Invoke-Expression "$Python -m venv `"$venv`""
}
$pythonExe = Join-Path $venv "Scripts\python.exe"
& $pythonExe -m pip install --upgrade pip
& $pythonExe -m pip install -r (Join-Path $root "requirements-baseer-ocr.txt")
& $pythonExe -c "import torch; from transformers import AutoProcessor, Qwen2VLForConditionalGeneration; import qwen_vl_utils"
Write-Host "Baseer OCR sidecar is ready at $venv"
Write-Host "Use OCR_ENGINE=baseer-ocr for Arabic document OCR on short samples or a worker."