Overview

Status
—
Not connected
Total Predictions
0
server memory
Avg Latency
—
ms / request
Model
—
—
Training Rows
—
feature count: —
🤖 Model Metadata
🔌Connect to a running Deeploi server to see metadata.
🏥 Health
Click Connect to check server health.
Endpoint Check
🚀 Quick Start

Start your Deeploi server, open this dashboard, then use the Playground or docs below to test requests.

deploy()
from deeploi import deploy
deploy(model, port=8000)
load() + serve()
from deeploi import load
pkg = load("artifacts/iris_rf")
pkg.serve(port=8000)
Docker artifact
pkg.save("artifacts/iris_rf", generate_docker=True)

# then build and run
docker build -t iris-model .
⚡ API Playground Send real requests to your running model
Endpoint
Input mode
Response
Waiting for request…
Requests
0
this session
Errors
0
non-200 responses
Avg Confidence
—
max class probability
Avg Latency
—
ms per request
📦 Prediction Distribution
⏱ Latency over Time
🎯 Confidence Buckets
📈 Requests per Minute
📋 Recent Predictions (Server Memory)
📭No predictions yet. Send requests to /predict or /predict_proba.
📖 API Reference

Base URL

All endpoints are relative to your server base URL (set in the top bar). Default: http://127.0.0.1:8000

POST /predict Return class predictions ▾

Returns the predicted class (or value for regressors) for each record.

Request
{ "records": [{ "feature_1": 1.0, "feature_2": 3.5 }] }
Response
{ "predictions": [0] }
POST /predict_proba Return class probabilities ▾

Classifiers only. Returns per-class confidence scores.

{
  "predictions":   [0],
  "probabilities": [{ "0": 0.91, "1": 0.06, "2": 0.03 }]
}
GET /meta Model metadata ▾

Returns model type, framework, task type, features and Deeploi version.

{
  "model_type":     "RandomForestClassifier",
  "framework":      "sklearn",
  "task_type":      "classification",
  "features":       ["sepal length (cm)", "..."],
  "deeploi_version": "0.2.0"
}
GET /health Health check ▾

Always returns 200 OK when the server is running.

{ "status": "ok" }

Errors

Validation failures return 422 Unprocessable Entity with a descriptive detail array. Feature name mismatches, missing fields, and wrong types are all caught automatically.

Python client example

client.py
import requests

BASE = "http://127.0.0.1:8000"
HEADERS = {"Content-Type": "application/json"}

payload = {
    "records": [{
        "sepal length (cm)": 5.1, "sepal width (cm)": 3.5,
        "petal length (cm)": 1.4, "petal width (cm)": 0.2
    }]
}

r = requests.post(f"{BASE}/predict", json=payload, headers=HEADERS)
print(r.json())  # {"predictions": [0]}
🔩 Feature Schema
🔌Connect to a running server to view the feature schema.