程式碼研究室簡介
1. 簡介
總覽
Cloud Run 日前開始支援 GPU。這項功能目前是公開測試計畫的等候名單項目。如果你有興趣試用這項功能,請填寫這份表單,加入候補名單。Cloud Run 是 Google Cloud 上的容器平台,可讓您輕鬆在容器中執行程式碼,無須管理叢集。
目前,我們提供的 GPU 是 Nvidia L4 GPU,具有 24 GB 的 vRAM。每個 Cloud Run 執行個體都有一個 GPU,且 Cloud Run 自動調整資源配置功能仍會套用。這包括最多可擴充至 5 個執行個體 (可提高配額),以及在沒有要求時將執行個體數調降至零。
在本程式碼研究室中,您將建立並部署 TorchServe 應用程式,使用 Stable Diffusion XL 從文字提示產生圖片。系統會將產生的圖片以 base64 編碼字串的形式傳回給呼叫端。
這個範例是根據「在 Torchserve 中使用 Huggingface Diffusers 執行 Stable diffusion 模型」一文所述的做法。本程式碼研究室會說明如何修改這個範例,以便與 Cloud Run 搭配運作。
課程內容
- 如何在 Cloud Run 上使用 GPU 執行 Stable Diffusion XL 模型
2. 啟用 API 並設定環境變數
您必須先啟用幾個 API,才能開始使用本程式碼研究室。本程式碼研究室需要使用下列 API。您可以執行下列指令來啟用這些 API:
gcloud services enable run.googleapis.com \ storage.googleapis.com \ cloudbuild.googleapis.com \
接著,您可以設定在本程式碼研究室中用到的環境變數。
PROJECT_ID=<YOUR_PROJECT_ID> REPOSITORY=repo NETWORK_NAME=default REGION=us-central1 IMAGE=us-central1-docker.pkg.dev/$PROJECT_ID/$REPOSITORY/gpu-torchserve
3. 建立 Torchserve 應用程式
首先,請建立原始碼目錄,然後切換至該目錄。
mkdir stable-diffusion-codelab && cd $_
建立 config.properties
檔案。這是 TorchServe 的設定檔。
inference_address=http://0.0.0.0:8080 enable_envvars_config=true min_workers=1 max_workers=1 default_workers_per_model=1 default_response_timeout=1000 load_models=all max_response_size=655350000 # to enable authorization, see https://github.com/pytorch/serve/blob/master/docs/token_authorization_api.md#how-to-set-and-disable-token-authorization disable_token_authorization=true
請注意,在本例中,監聽位址 http://0.0.0.0 會用於在 Cloud Run 上運作。Cloud Run 的預設通訊埠為 8080。
建立 requirements.txt
檔案。
python-dotenv accelerate transformers diffusers numpy google-cloud-storage nvgpu
建立名為 stable_diffusion_handler.py
的檔案
from abc import ABC import base64 import datetime import io import logging import os from diffusers import StableDiffusionXLImg2ImgPipeline from diffusers import StableDiffusionXLPipeline from google.cloud import storage import numpy as np from PIL import Image import torch from ts.torch_handler.base_handler import BaseHandler logger = logging.getLogger(__name__) def image_to_base64(image: Image.Image) -> str: """Convert a PIL image to a base64 string.""" buffer = io.BytesIO() image.save(buffer, format="JPEG") image_str = base64.b64encode(buffer.getvalue()).decode("utf-8") return image_str class DiffusersHandler(BaseHandler, ABC): """Diffusers handler class for text to image generation.""" def __init__(self): self.initialized = False def initialize(self, ctx): """In this initialize function, the Stable Diffusion model is loaded and initialized here. Args: ctx (context): It is a JSON Object containing information pertaining to the model artifacts parameters. """ logger.info("Initialize DiffusersHandler") self.manifest = ctx.manifest properties = ctx.system_properties model_dir = properties.get("model_dir") model_name = os.environ["MODEL_NAME"] model_refiner = os.environ["MODEL_REFINER"] self.bucket = None logger.info( "GPU device count: %s", torch.cuda.device_count(), ) logger.info( "select the GPU device, cuda is available: %s", torch.cuda.is_available(), ) self.device = torch.device( "cuda:" + str(properties.get("gpu_id")) if torch.cuda.is_available() and properties.get("gpu_id") is not None else "cpu" ) logger.info("Device used: %s", self.device) # open the pipeline to the inferenece model # this is generating the image logger.info("Donwloading model %s", model_name) self.pipeline = StableDiffusionXLPipeline.from_pretrained( model_name, variant="fp16", torch_dtype=torch.float16, use_safetensors=True, ).to(self.device) logger.info("done donwloading model %s", model_name) # open the pipeline to the refiner # refiner is used to remove artifacts from the image logger.info("Donwloading refiner %s", model_refiner) self.refiner = StableDiffusionXLImg2ImgPipeline.from_pretrained( model_refiner, variant="fp16", torch_dtype=torch.float16, use_safetensors=True, ).to(self.device) logger.info("done donwloading refiner %s", model_refiner) self.n_steps = 40 self.high_noise_frac = 0.8 self.initialized = True # Commonly used basic negative prompts. logger.info("using negative_prompt") self.negative_prompt = ("worst quality, normal quality, low quality, low res, blurry") # this handles the user request def preprocess(self, requests): """Basic text preprocessing, of the user's prompt. Args: requests (str): The Input data in the form of text is passed on to the preprocess function. Returns: list : The preprocess function returns a list of prompts. """ logger.info("Process request started") inputs = [] for _, data in enumerate(requests): input_text = data.get("data") if input_text is None: input_text = data.get("body") if isinstance(input_text, (bytes, bytearray)): input_text = input_text.decode("utf-8") logger.info("Received text: '%s'", input_text) inputs.append(input_text) return inputs def inference(self, inputs): """Generates the image relevant to the received text. Args: input_batch (list): List of Text from the pre-process function is passed here Returns: list : It returns a list of the generate images for the input text """ logger.info("Inference request started") # Handling inference for sequence_classification. image = self.pipeline( prompt=inputs, negative_prompt=self.negative_prompt, num_inference_steps=self.n_steps, denoising_end=self.high_noise_frac, output_type="latent", ).images logger.info("Done model") image = self.refiner( prompt=inputs, negative_prompt=self.negative_prompt, num_inference_steps=self.n_steps, denoising_start=self.high_noise_frac, image=image, ).images logger.info("Done refiner") return image def postprocess(self, inference_output): """Post Process Function converts the generated image into Torchserve readable format. Args: inference_output (list): It contains the generated image of the input text. Returns: (list): Returns a list of the images. """ logger.info("Post process request started") images = [] response_size = 0 for image in inference_output: # Save image to GCS if self.bucket: image.save("temp.jpg") # Create a blob object blob = self.bucket.blob( datetime.datetime.now().strftime("%Y%m%d_%H%M%S") + ".jpg" ) # Upload the file blob.upload_from_filename("temp.jpg") # to see the image, encode to base64 encoded = image_to_base64(image) response_size += len(encoded) images.append(encoded) logger.info("Images %d, response size: %d", len(images), response_size) return images
建立名為 start.sh
的檔案。這個檔案會用於做為容器中的進入點,用於啟動 TorchServe。
#!/bin/bash echo "starting the server" # start the server. By default torchserve runs in backaround, and start.sh will immediately terminate when done # so use --foreground to keep torchserve running in foreground while start.sh is running in a container torchserve --start --ts-config config.properties --models "stable_diffusion=${MAR_FILE_NAME}.mar" --model-store ${MAR_STORE_PATH} --foreground
然後執行下列指令,將其設為可執行檔案。
chmod 755 start.sh
建立 dockerfile
。
# pick a version of torchserve to avoid any future breaking changes # docker pull pytorch/torchserve:0.11.1-cpp-dev-gpu FROM pytorch/torchserve:0.11.1-cpp-dev-gpu AS base USER root WORKDIR /home/model-server COPY requirements.txt ./ RUN pip install --upgrade -r ./requirements.txt # Stage 1 build the serving container. FROM base AS serve-gcs ENV MODEL_NAME='stabilityai/stable-diffusion-xl-base-1.0' ENV MODEL_REFINER='stabilityai/stable-diffusion-xl-refiner-1.0' ENV MAR_STORE_PATH='/home/model-server/model-store' ENV MAR_FILE_NAME='model' RUN mkdir -p $MAR_STORE_PATH COPY config.properties ./ COPY stable_diffusion_handler.py ./ COPY start.sh ./ # creates the mar file used by torchserve RUN torch-model-archiver --force --model-name ${MAR_FILE_NAME} --version 1.0 --handler stable_diffusion_handler.py -r requirements.txt --export-path ${MAR_STORE_PATH} # entrypoint CMD ["./start.sh"]
4. 設定 Cloud NAT
有了 Cloud NAT,您就能享有更高的頻寬,可用於存取網際網路及從 HuggingFace 下載模型,大幅縮短部署時間。
如要使用 Cloud NAT,請執行下列指令來啟用 Cloud NAT 執行個體:
gcloud compute routers create nat-router --network $NETWORK_NAME --region us-central1 gcloud compute routers nats create vm-nat --router=nat-router --region=us-central1 --auto-allocate-nat-external-ips --nat-all-subnet-ip-ranges
5. 建構及部署 Cloud Run 服務
將程式碼提交至 Cloud Build。
gcloud builds submit --tag $IMAGE
接下來,部署至 Cloud Run
gcloud beta run deploy gpu-torchserve \ --image=$IMAGE \ --cpu=8 --memory=32Gi \ --gpu=1 --no-cpu-throttling --gpu-type=nvidia-l4 \ --allow-unauthenticated \ --region us-central1 \ --project $PROJECT_ID \ --execution-environment=gen2 \ --max-instances 1 \ --network $NETWORK_NAME \ --vpc-egress all-traffic
6. 測試服務
您可以執行下列指令來測試服務:
PROMPT_TEXT="a cat sitting in a magnolia tree" SERVICE_URL=$(gcloud run services describe gpu-torchserve --region $REGION --format 'value(status.url)') time curl $SERVICE_URL/predictions/stable_diffusion -d "data=$PROMPT_TEXT" | base64 --decode > image.jpg
您會在目前的目錄中看到 image.jpg
檔案。您可以在 Cloud Shell 編輯器中開啟圖片,查看樹上坐著一隻貓的圖片。
8. 清除所用資源
為避免產生意外費用 (例如,如果這個 Cloud Run 工作在免費方案中按月分配的 Cloud Run 叫用次數),您可以刪除 Cloud Run 工作,或刪除您在步驟 2 中建立的專案。
如要刪除 Cloud Run 工作,請前往 Cloud Run 控制台 (網址:https://console.cloud.google.com/run/) 並刪除 gpu-torchserve
服務。
您也應該刪除 Cloud NAT 設定。
如果您選擇刪除整個專案,可以前往 https://console.cloud.google.com/cloud-resource-manager,選取您在步驟 2 中建立的專案,然後選擇「Delete」(刪除)。如果您刪除專案,就必須在 Cloud SDK 中變更專案。您可以執行 gcloud projects list
來查看所有可用專案的清單。