Time Series Data

Model Overview

This API uses a neuro-symbolic approach to analyse time series data, identifying patterns to predict future market trends. Its modular base architecture enables efficient fine-tuning across different time horizons, significantly reducing model development time.

Supported Markets

Bonds

Commodities

Cryptocurrencies

Derivatives

Equities

Foreign Exchange

Indices

REITs

Tokenised Assets

Base URL

 https://www.sumtyme.com/shared

Request Endpoint

POST /auth/ti 

OHLC Input Data Lengths

Required Length

200 Periods

Recommended Length

500 Periods

Extended Length

501-3000 Periods

Request Header

Content-Type

string

required

Must be set to application/json

Api-Key

string

required

The API key used to authenticate requests to the API.

Request Body

Datetime

list

required

Open

list

required

High

list

required

Low

list

required

Close

list

required

Response Fields

datetime

string

Trade opportunity timestamp

trend_identified

string

Show Options

1

int

Bullish trend

-1

int

Bearish trend

Request Example

When forecasting future price trends, a future date must be specified (e.g. Jan 2nd at 10:00:00) and unknown values be filled with zeros (0):

{
    'Datetime': ["2024-01-01 10:00:00","2024-01-01 11:00:00", ..., "2025-01-02 10:00:00"],
    'Open': [194.33, 198.75, ..., 0],         # Unknown future value
    'High': [197.02, 197.84, ..., 0],         # Unknown future value
    'Low': [192.87, 196.33, ..., 0],          # Unknown future value
    'Close': [195.45, 196.01, ..., 0]         # Unknown future value
}

Example - API Request

# Import necessary libraries
import pandas as pd
import requests
import json

# Define function to create json for API 
def time_series_dict(df):
    # Convert DataFrame to dictionary format
    return {
        "Datetime": df['Datetime'].tolist(),
        "Open": df['Open'].tolist(),
        "High": df['High'].tolist(),
        "Low": df['Low'].tolist(),
        "Close": df['Close'].tolist()
    }

# Import Data
df = pd.read_csv("")[['Datetime','Open','High','Low','Close']]

# Create JSON payload for API request
data = time_series_dict(df)

# Specify Base URL & API KEY
BASE_URL = ""
API_KEY = ""

# Set API endpoint URL
api_url = f"{BASE_URL}/auth/ti"

# Set header for API request
headers = {"Api-Key": f"{API_KEY}", 'Content-Type': 'application/json'}

# Send POST request to API
response = requests.post(api_url, json=data, headers=headers)

data = json.loads(response.text)

Example - API Response

{

"2024-01-01 10:00:00": {
      "trend_identified": null
    },
    
"2024-01-01 11:00:00": {
      "trend_identified": null
    },

...

"2025-01-02 10:00:00": {
      "trend_identified": 1
    }
    
}