Deploying a Custom Workflow

First, please make sure you have signed up for a Sieve account as per the quickstart guide.

​

Install the Sieve CLI

pip install https://mango.sievedata.com/v1/client_package/sievedata-0.0.1.1.2-py3-none-any.whl

Then, export your Sieve API key as an environment variable. You can find your API key in the Sieve dashboard.

export SIEVE_API_KEY=YOUR_API_KEY

​

Create a Workflow

Create a new folder and add the files below.

import sieve
from typing import Dict, Tuple
from yolo import Yolo
from splitter import VideoSplitter

@sieve.workflow(name="yolo_object_detection")
def yolo_workflow(video: sieve.Video) -> Dict:
    images = VideoSplitter(video)
    yolo_outputs = Yolo()(images)
    return yolo_outputs

​

Deploy the Workflow

You workflow will be deployed. You can monitor the status via your terminal or on the Sieve dashboard.

sieve deploy

​

Run the Workflow

You can run the workflow on a video file. You can do this via the dashboard, the API, or the CLI. We can use this file to test.

​

CLI

sieve push yolo_object_detection '{"video": {"url": "https://storage.googleapis.com/sieve-test-videos-central/01-lebron-dwade.mp4"}}'

​

API

curl --request POST \
    --url https://mango.sievedata.com/v1/push \
    --header "X-API-Key: $SIEVE_API_KEY" \
    --header "Content-Type: application/json" \
    --data '{
      "workflow_name": "yolo_object_detection",
      "inputs": {
        "video": {
          "url": "https://storage.googleapis.com/sieve-test-videos-central/01-lebron-dwade.mp4"
        }
      }
    }'

​

Dashboard

Navigate to the yolo_object_detection worklow on the dashboard, enter the video URL, and click Send Request.

​

More Examples

You can find more examples in the Sieve Examples Repo.