Skip to content

Latest commit

 

History

History
168 lines (117 loc) · 11.6 KB

Finetuning_VSCodeaitoolkit.md

File metadata and controls

168 lines (117 loc) · 11.6 KB

Bienvenido al AI Toolkit para VS Code

AI Toolkit para VS Code reúne varios modelos del Catálogo de Azure AI Studio y otros catálogos como Hugging Face. El toolkit simplifica las tareas comunes de desarrollo para construir aplicaciones de IA con herramientas y modelos de IA generativa a través de:

  • Empezar con el descubrimiento de modelos y el playground.
  • Ajuste fino de modelos e inferencia utilizando recursos informáticos locales.
  • Ajuste fino e inferencia remotos utilizando recursos de Azure.

Instalar AI Toolkit para VSCode

AIToolkit FineTuning

[Vista Previa Privada] Aprovisionamiento con un solo clic para Azure Container Apps para ejecutar el ajuste fino de modelos e inferencia en la nube.

Ahora vamos a sumergirnos en el desarrollo de tu aplicación de IA:

Desarrollo Local

Preparativos

  1. Asegúrate de que el controlador NVIDIA esté instalado en el host.
  2. Ejecuta huggingface-cli login, si estás usando HF para la utilización del conjunto de datos.
  3. Olive explicaciones de configuración clave para cualquier cosa que modifique el uso de memoria.

Activar Conda

Dado que estamos utilizando un entorno WSL y es compartido, necesitas activar manualmente el entorno conda. Después de este paso, puedes ejecutar el ajuste fino o la inferencia.

conda activate [conda-env-name]

Solo ajuste fino del modelo base

Para solo probar el modelo base sin ajuste fino, puedes ejecutar este comando después de activar conda.

cd inference

# Web browser interface allows to adjust a few parameters like max new token length, temperature and so on.
# User has to manually open the link (e.g. http://0.0.0.0:7860) in a browser after gradio initiates the connections.
python gradio_chat.py --baseonly

Ajuste fino del modelo e inferencia

Una vez que el espacio de trabajo esté abierto en un contenedor de desarrollo, abre una terminal (la ruta predeterminada es la raíz del proyecto), luego ejecuta el comando a continuación para ajustar un LLM en el conjunto de datos seleccionado.

python finetuning/invoke_olive.py

Los puntos de control y el modelo final se guardarán en models folder.

Next run inferencing with the fune-tuned model through chats in a console, web browser or prompt flow.

cd inference

# Console interface.
python console_chat.py

# Web browser interface allows to adjust a few parameters like max new token length, temperature and so on.
# User has to manually open the link (e.g. http://127.0.0.1:7860) in a browser after gradio initiates the connections.
python gradio_chat.py

Para usar prompt flow in VS Code, please refer to this Quick Start.

Model Fine-tuning

Next, download the following model depending on the availability of a GPU on your device.

To initiate the local fine-tuning session using QLoRA, select a model you want to fine-tune from our catalog.

Platform(s) GPU available Model name Size (GB)
Windows Yes Phi-3-mini-4k-directml-int4-awq-block-128-onnx 2.13GB
Linux Yes Phi-3-mini-4k-cuda-int4-onnx 2.30GB
Windows
Linux		 No Phi-3-mini-4k-cpu-int4-rtn-block-32-acc-level-4-onnx 2.72GB

Note You do not need an Azure Account to download the models

The Phi3-mini (int4) model is approximately 2GB-3GB in size. Depending on your network speed, it could take a few minutes to download.

Start by selecting a project name and location. Next, select a model from the model catalog. You will be prompted to download the project template. You can then click "Configure Project" to adjust various settings.

Microsoft Olive

We use Olive to run QLoRA fine-tuning on a PyTorch model from our catalog. All of the settings are preset with the default values to optimize to run the fine-tuning process locally with optimized use of memory, but it can be adjusted for your scenario.

Fine Tuning Samples and Resoures

[Private Preview] Remote Development

Prerequisites

  1. To run the model fine-tuning in your remote Azure Container App Environment, make sure your subscription has enough GPU capacity. Submit a support ticket to request the required capacity for your application. Get More Info about GPU capacity
  2. If you are using private dataset on HuggingFace, make sure you have a HuggingFace account and generate an access token
  3. Enable Remote Fine-tuning and Inference feature flag in the AI Toolkit for VS Code
    1. Open the VS Code Settings by selecting File -> Preferences -> Settings.
    2. Navigate to Extensions and select AI Toolkit.
    3. Select the "Enable Remote Fine-tuning And Inference" option.
    4. Reload VS Code to take effect.

Setting Up a Remote Development Project

  1. Execute the command palette AI Toolkit: Focus on Resource View.
  2. Navigate to Model Fine-tuning to access the model catalog. Assign a name to your project and select its location on your machine. Then, hit the "Configure Project" button.
  3. Project Configuration
    1. Avoid enabling the "Fine-tune locally" option.
    2. The Olive configuration settings will appear with pre-set default values. Please adjust and fill in these configurations as required.
    3. Move on to Generate Project. This stage leverages WSL and involves setting up a new Conda environment, preparing for future updates that include Dev Containers.
  4. Click on "Relaunch Window In Workspace" to open your remote development project.

Note: The project currently works either locally or remotely within the AI Toolkit for VS Code. If you choose "Fine-tune locally" during project creation, it will operate exclusively in WSL without remote development capabilities. On the other hand, if you forego enabling "Fine-tune locally", the project will be restricted to the remote Azure Container App environment.

Provision Azure Resources

To get started, you need to provision the Azure Resource for remote fine-tuning. Do this by running the AI Toolkit: Provision Azure Container Apps job for fine-tuning from the command palette.

Monitor the progress of the provision through the link displayed in the output channel.

[Optional] Add Huggingface Token to the Azure Container App Secret

If you're using private HuggingFace dataset, set your HuggingFace token as an environment variable to avoid the need for manual login on the Hugging Face Hub. You can do this using the AI Toolkit: Add Azure Container Apps Job secret for fine-tuning command. With this command, you can set the secret name as HF_TOKEN and use your Hugging Face token as the secret value.

Run Fine-tuning

To start the remote fine-tuning job, execute the AI Toolkit: Run fine-tuning command.

To view the system and console logs, you can visit the Azure portal using the link in the output panel (more steps at View and Query Logs on Azure). Or, you can view the console logs directly in the VSCode output panel by running the command AI Toolkit: Show the running fine-tuning job streaming logs.

Note: The job might be queued due to insufficient resources. If the log is not displayed, execute the AI Toolkit: Show the running fine-tuning job streaming logs command, wait for a while and then execute the command again to re-connect to the streaming log.

During this process, QLoRA will be used for fine-tuning, and will create LoRA adapters for the model to use during inference. The results of the fine-tuning will be stored in the Azure Files.

Provision Inference Endpoint

After the adapters are trained in the remote environment, use a simple Gradio application to interact with the model. Similar to the fine-tuning process, you need to set up the Azure Resources for remote inference by executing the AI Toolkit: Provision Azure Container Apps for inference from the command palette.

By default, the subscription and the resource group for inference should match those used for fine-tuning. The inference will use the same Azure Container App Environment and access the model and model adapter stored in Azure Files, which were generated during the fine-tuning step.

Deploy the Inference Endpoint

If you wish to revise the inference code or reload the inference model, please execute the AI Toolkit: Deploy for inference command. This will synchronize your latest code with Azure Container App and restart the replica.

Once deployment is successfully completed, you can access the inference API by clicking on the "Go to Inference Endpoint" button displayed in the VSCode notification. Or, the web API endpoint can be found under ACA_APP_ENDPOINT in ./infra/inference.config.json y en el panel de salida. Ahora estás listo para evaluar el modelo utilizando este endpoint.

Uso Avanzado

Para más información sobre el desarrollo remoto con AI Toolkit, consulta la documentación Ajuste Fino de Modelos Remotamente e Inferencia con el Modelo Ajustado Finamente.

Descargo de responsabilidad: Este documento ha sido traducido utilizando servicios de traducción automática basados en IA. Si bien nos esforzamos por lograr precisión, tenga en cuenta que las traducciones automatizadas pueden contener errores o inexactitudes. El documento original en su idioma nativo debe considerarse la fuente autorizada. Para información crítica, se recomienda la traducción humana profesional. No somos responsables de ningún malentendido o interpretación errónea que surja del uso de esta traducción.