Skip to main content

Using the OnPremiseClassifier

The OnPremiseClassifier provides a scikit-learn compatible interface for using Neuralk’s In-Context Learning model with an on-premise or self-hosted NICL server. It’s ideal for users who have deployed NICL on their own infrastructure or need to work with on-premise deployments.

NOTE​

For this example to run, you need access to a running NICL server. The server URL should be specified via the host parameter when initializing the classifier.

Simple example on toy data​

We start by using the OnPremiseClassifier on simple data that needs no preprocessing.

Generate simple data:

import numpy as np
from sklearn.datasets import make_classification
from sklearn.model_selection import train_test_split

X, y = make_classification(n_samples=200, n_features=10, n_classes=2, random_state=42)
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

# Ensure data is in the correct format (float32 for features, int64 for labels)
X_train = X_train.astype(np.float32)
X_test = X_test.astype(np.float32)
y_train = y_train.astype(np.int64)

print(f"{X_train.shape=} {y_train.shape=} {X_test.shape=} {y_test.shape=}")
X_train.shape=(160, 10) y_train.shape=(160,) X_test.shape=(40, 10) y_test.shape=(40,)

Now we apply Neuralk’s OnPremiseClassifier.

NOTE​

Replace “http://localhost:8000” with the actual URL of your NICL server. If your server requires authentication, you can pass it via the default_headers parameter.

from sklearn.metrics import accuracy_score

from neuralk import OnPremiseClassifier

# Initialize the classifier with your NICL server URL
# Replace with your actual server URL
classifier = OnPremiseClassifier(
    host="http://localhost:8000",  # Replace with your NICL server URL
    model="nicl-small",
    timeout_s=300,
)

# Note: nothing actually happens during fit() -- in-context learning models are
# pretrained but require no fitting on our specific dataset. The fit method
# only stores the training data.
classifier = classifier.fit(X_train, y_train)

# Make predictions
predictions = classifier.predict(X_test)
probabilities = classifier.predict_proba(X_test)

accuracy = accuracy_score(y_test, predictions)
print(f"Accuracy: {accuracy}")
print(f"Predictions shape: {predictions.shape}")
print(f"Probabilities shape: {probabilities.shape}")
Accuracy: 0.875
Predictions shape: (40,)
Probabilities shape: (40, 2)

Working with authentication​

If your NICL server requires authentication, you can pass authentication headers via the default_headers parameter:

Example with authentication headers

classifier_with_auth = OnPremiseClassifier(
    host="http://localhost:8000",
    model="nicl-small",
    default_headers={"Authorization": "Bearer your-token"},
)

Advanced configuration​

The OnPremiseClassifier supports various configuration options for fine-tuning the connection and request behavior:

Example with advanced configuration

classifier_advanced = OnPremiseClassifier(
    host="http://localhost:8000",
    dataset_name="my-dataset",
    model="nicl-large",  # Use a different model
    timeout_s=600,  # Longer timeout
    metadata={"source": "example", "version": "1.0"},
    user="user123",
    api_version="v1",
)

Integration with scikit-learn pipelines​

Like the cloud-based Classifier, OnPremiseClassifier can be integrated into scikit-learn pipelines:

from sklearn.decomposition import PCA
from sklearn.pipeline import make_pipeline
from sklearn.preprocessing import StandardScaler

# Create a pipeline with preprocessing and OnPremiseClassifier
pipeline = make_pipeline(
    StandardScaler(),
    PCA(n_components=5),
    OnPremiseClassifier(
        host="http://localhost:8000",
        model="nicl-small",
    ),
)

# Fit and predict
pipeline.fit(X_train, y_train)
pipeline_predictions = pipeline.predict(X_test)
pipeline_accuracy = accuracy_score(y_test, pipeline_predictions)
print(f"Pipeline accuracy: {pipeline_accuracy}")
Pipeline accuracy: 0.85

Total running time of the script: (0 minutes 0.610 seconds)

On this page