Skip to main content

Neuralk SDK Quickstart

The Neuralk SDK is a python library that provides a simple interface to the state-of-the-art foundation models & expert modules accessible via the Neuralk API. Get started by generating your access credentials and running your first API call.

Generate credentials

Load credentials​

To connect to the Neuralk API, you need to authenticate with your credentials. Instead of hardcoding them, you can store them securely in environment variables.

The most common approaches are:

1. Environment variables in the operating system​

export NEURALK_USERNAME=your_username
export NEURALK_PASSWORD=your_password

2. Using a .env file (useful during development)​

Many Python projects use environment variables to store sensitive information like API keys or configuration settings. Create a file named .env in the root of your project with:

.env
NEURALK_USERNAME=your_username
NEURALK_PASSWORD=your_password

To access these variables in your scripts, you can use the python-dotenv package. First, install it with pip install python-dotenv.

Then, at the start of your script, load your environment variables using dotenv:

import dotenv
dotenv.load_dotenv()

Don’t forget to add .env to your .gitignore to avoid committing it.

Install the Neuralk SDK and run your first prediction​

The Neuralk SDK is available on PyPI and can be installed using pip. This is the recommended installation method for users who want to integrate the SDK into their Python projects.

Install the SDK with pip
pip install neuralk

Verifying your installation​

You can verify that the Neuralk package was installed correctly by importing it in your terminal:

Test installation in your terminal
python -c "import neuralk; print('Neuralk imported successfully')"

If you see the message above, the installation was successful.

Next, you can create a test file, e.g., toy.py, and copy the example code into it to run a simple classification.

Test a toy classification example
import time

import dotenv
from sklearn.datasets import make_classification
from sklearn.model_selection import train_test_split
from sklearn.metrics import accuracy_score

from neuralk import Classifier

dotenv.load_dotenv()

X, y = make_classification(random_state=0, n_samples=1_000, n_features=10)
X_train, X_test, y_train, y_test = train_test_split(X, y)

start = time.monotonic()
classif = Classifier()
classif.fit(X_train, y_train)
prediction = classif.predict(X_test)
stop = time.monotonic()
print(f"fit & predict took {stop - start:.1f}s")
print(prediction.ravel())
print(f"{X_train.shape=} {X_test.shape=} {y_train.shape=} {y_test.shape=} {prediction.shape=}")
print("accuracy:", accuracy_score(y_test, prediction))

Run the code by executing python toy.py. After a few moments, the results of your API request should appear.

This lets you quickly check that the SDK is working before moving on to your own projects.

Explore more classification examples

On this page