Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/sujith52/fraud/llms.txt

Use this file to discover all available pages before exploring further.

​
Endpoint

POST /predict
Makes fraud predictions on new transaction data using the trained model. This endpoint accepts data file paths and returns predictions in a CSV file.

​
Request Body

The endpoint accepts requests in two formats: JSON or form data.

​
JSON Format

​
filepath
string
required
The file system path to the folder containing prediction data files. This folder should contain properly formatted CSV files with transaction data to classify.
{
  "filepath": "/path/to/prediction/data"
}

​
Form Data Format

​
filepath
string
required
The file system path to the folder containing prediction data files, sent as form data.

​
Prediction Workflow

When the /predict endpoint is called, the following process is executed:
  1. Validation - The pred_validation object validates the prediction data structure
  2. Data Processing - Prediction data is validated and prepared for inference
  3. Prediction - The prediction object loads the trained model and generates predictions
  4. Output - Predictions are saved to a CSV file
pred_val = pred_validation(path)
pred_val.prediction_validation()

pred = prediction(path)
path = pred.predictionFromModel()

​
Response

​
success
string
Returns the file path where the prediction results have been saved.

​
Success Response

Prediction File created at /path/to/predictions/Predictions.csv!!!
The response includes the full path to the generated predictions file.

​
Error Responses

​
error
string
Returns an error message if prediction fails.
Possible error responses: 
Error Occurred! <class 'ValueError'>
Error Occurred! <class 'KeyError'>
Error Occurred! {exception_message}

​
Example Requests

​
cURL (JSON)

curl -X POST http://127.0.0.1:5001/predict \
  -H "Content-Type: application/json" \
  -d '{
    "filepath": "/home/user/fraud_detection/prediction_data"
  }'

​
cURL (Form Data)

curl -X POST http://127.0.0.1:5001/predict \
  -F "filepath=/home/user/fraud_detection/prediction_data"

​
Python (requests - JSON)

import requests

url = "http://127.0.0.1:5001/predict"
payload = {
    "filepath": "/home/user/fraud_detection/prediction_data"
}

response = requests.post(url, json=payload)
print(response.text)
# Output: Prediction File created at /path/to/predictions/Predictions.csv!!!

​
Python (requests - Form Data)

import requests

url = "http://127.0.0.1:5001/predict"
data = {
    "filepath": "/home/user/fraud_detection/prediction_data"
}

response = requests.post(url, data=data)
print(response.text)

​
JavaScript (fetch - JSON)

fetch('http://127.0.0.1:5001/predict', {
  method: 'POST',
  headers: {
    'Content-Type': 'application/json',
  },
  body: JSON.stringify({
    filepath: '/home/user/fraud_detection/prediction_data'
  })
})
.then(response => response.text())
.then(data => console.log(data));

​
Error Handling

The endpoint handles the following error cases:
Raised when the provided file path or data format is invalid.
Error Occurred! <class 'ValueError'>
Raised when the required filepath parameter is missing from the request.
Error Occurred! <class 'KeyError'>
Catches any other exceptions during the prediction process and returns the specific error message.
Error Occurred! {specific_error_message}

​
Implementation Details

The endpoint implementation from main.py:25-61: 
@app.route("/predict", methods=['POST'])
@cross_origin()
def predictRouteClient():
    try:
        if request.json is not None:
            path = request.json['filepath']
            
            pred_val = pred_validation(path)
            pred_val.prediction_validation()
            
            pred = prediction(path)
            path = pred.predictionFromModel()
            return Response("Prediction File created at %s!!!" % path)

        elif request.form is not None:
            path = request.form['filepath']
            
            pred_val = pred_validation(path)
            pred_val.prediction_validation()
            
            pred = prediction(path)
            path = pred.predictionFromModel()
            return Response("Prediction File created at %s!!!" % path)

    except ValueError:
        return Response("Error Occurred! %s" %ValueError)
    except KeyError:
        return Response("Error Occurred! %s" %KeyError)
    except Exception as e:
        return Response("Error Occurred! %s" %e)

​
Notes

The endpoint accepts both JSON and form data formats for flexibility. Both formats trigger the same prediction workflow and return identical results.
Prediction results are automatically saved to a CSV file. The exact location is determined by the prediction module configuration.

Build docs developers (and LLMs) love

Get started for freeTalk to us