Case: Classification

Authentication

Create an access credential from My PhenoNet → API Access. Send it with every request:

Authorization: AccessKey <access_id>:<secret_key>

Keep the secret key on your server. Do not expose it in browser code or a public repository.

Submit a classification task

POST https://api.phenonet.org/v1/openapi/classification/

Send a JSON request body. image_url must be an HTTPS URL from an image host allowed by the platform.

Field	Required	Description
`image_url`	Yes	HTTPS URL of the image to classify.
`task_name`	No	A name for this task. Defaults to `API prediction`.
`model_id`	No	Available single-image model ID. The platform default is used when omitted.

{
  "image_url": "https://files.example.org/images/wheat-001.jpg",
  "task_name": "wheat-001",
  "model_id": "your-model-id"
}

A successful submission returns HTTP 202 and a server-generated task_id:

{
  "code": 202,
  "msg": "accepted",
  "data": {
    "task_id": "server-generated-task-id",
    "status": "queued",
    "status_url": "https://api.phenonet.org/v1/openapi/classification/server-generated-task-id/",
    "created_at": "2026-08-03T00:00:00+00:00",
    "finished_at": null
  }
}

Get task status

GET https://api.phenonet.org/v1/openapi/classification/{task_id}/

The response status is one of queued, running, waiting, succeeded, or failed. When it succeeds, data.result contains the predicted stage and its index.

Python example

import time

import requests

BASE_URL = "https://api.phenonet.org/v1/openapi/classification/"
ACCESS_ID = "your-access-id"
SECRET_KEY = "your-secret-key"

headers = {
    "Authorization": f"AccessKey {ACCESS_ID}:{SECRET_KEY}",
}

payload = {
    "image_url": "https://files.example.org/images/wheat-001.jpg",
    "task_name": "wheat-001",
    # "model_id": "your-model-id",  # Optional.
}

created = requests.post(BASE_URL, headers=headers, json=payload, timeout=30)
created.raise_for_status()
task = created.json()["data"]
task_id = task["task_id"]

while task["status"] not in {"succeeded", "failed"}:
    time.sleep(5)
    response = requests.get(f"{BASE_URL}{task_id}/", headers=headers, timeout=30)
    response.raise_for_status()
    task = response.json()["data"]

if task["status"] == "succeeded":
    print(task["result"])
else:
    print("Task failed:", task)

Error responses

HTTP status	Meaning
`400`	The request body is invalid, contains an unexpected field, or the image URL is not allowed.
`401`	The access credential is missing or invalid.
`404`	The task does not exist or does not belong to this credential.
`409`	Task dispatch is already in progress.
`503`	The compute service did not accept the task.