Intel oneAPI Data Analytics Library (daal)

2026.0.0 verified Mon Apr 27 auth: no python

Intel oneAPI Data Analytics Library (oneDAL) provides optimized algorithms for data analysis and machine learning, including classification, regression, clustering, and dimensionality reduction. The daal package is the Python wrapper for oneDAL. Current version: 2026.0.0. Release cadence: monthly.

pip install daal

Common errors

error ModuleNotFoundError: No module named 'daal' ↓

cause Trying to import 'daal' directly, but the importable module is 'daal4py'.

fix

Install 'daal' via pip and use 'import daal4py' in your Python code.

error AttributeError: module 'daal4py' has no attribute 'kmeans' ↓

cause Incorrect case or misspelling; the function is 'kmeans' (lowercase).

fix

Use 'from daal4py import kmeans' or 'd4p.kmeans(...)'.

Warnings

gotcha The pip package is named 'daal', but the Python import is 'daal4py'. Using 'import daal' will raise ModuleNotFoundError. ↓

fix Install 'pip install daal' and then use 'import daal4py' in your code.

breaking In version 2026.0.0, old DAAL C++ API headers (e.g., daal.h) are removed. The Python package may drop legacy compatibility. ↓

fix Use the daal4py Python API instead of low-level C++ bindings.

deprecated The Matrix, PackedSymmetricMatrix, and PackedTriangularMatrix classes are deprecated and scheduled for removal. ↓

fix Use the new data management API (e.g., NumericTable) or migrate to daal4py.

Imports

daal4py
wrong
```
import daal
```
correct
```
import daal4py
```
'import daal' does not work; use 'import daal4py' for the Python API.

KMeans

wrong

from daal import kmeans

correct

from daal4py import kmeans

KMeans is under daal4py, not daal.

Quickstart

Quick start showing KMeans clustering with daal4py.

import numpy as np
import daal4py as d4p

# Generate sample data
X = np.random.rand(100, 10).astype(np.float32)

# Create KMeans algorithm
kmeans_algo = d4p.kmeans(nClusters=3, maxIterations=100)

# Compute
result = kmeans_algo.compute(X)

print('Cluster centroids:', result.centroids)
print('Assignments:', result.assignments)