Product Information

NVIDIA RAPIDS NEWS DETAIL

Current Position:Home > News and Insights
News and technical insightsNews and Insights ITZKXY enterprise networking and AI infrastructure support2025-06-09

This product is a networking product designed for enterprise networks.

This product is a networking product designed for enterprise and data center networking.

DoubleML is a networking product designed for enterprise networks.

RAPIDS GPU and AI cuML is a networking product designed for enterprise and data center networking.

RAPIDS is a networking product designed for enterprise and data center networking.

This product is a networking product designed for enterprise and data center networking.

scikit-learn is a networking product designed for enterprise and data center networking.

A/B is a networking product designed for enterprise and data center networking.

This product is a networking product designed for enterprise networks.

XGBoost is a networking product designed for enterprise and data center networking.

for

CPU is a networking product designed for enterprise and data center networking.

DoubleML is a networking product designed for enterprise and data center networking.

import doubleml as dml
from doubleml.datasets import make_plr_CCDDHNR2018
from sklearn.ensemble import RandomForestRegressor
from sklearn.base import clone
import cuml
 
NROWS = [10000, 100000, 1000000, 10000000]
USE_GPU = True
 
for N in NROWS:
    data = make_plr_CCDDHNR2018(alpha=0.5, n_obs=N, dim_x=100, return_type="DataFrame").astype("float32")
    obj_dml_data = dml.DoubleMLData(data, "y", "d")
 
    if USE_GPU == "cuml":
  learner = cuml.ensemble.RandomForestRegressor(n_estimators=200, max_features=100, max_depth=10, min_samples_leaf=2)
    else: # standard scikit-learn
  learner = RandomForestRegressor(n_estimators=200, max_features=100, max_depth=10, min_samples_leaf=2, n_jobs=-1)
 
    ml_l = clone(learner)
    ml_m = clone(learner)
    dml_plr_obj = dml.DoubleMLPLR(obj_dml_data, ml_l, ml_m).fit()

CPU is a networking product designed for enterprise and data center networking.


A bar chart showing the time required to train the DoubleML PLR using Random Forest Regression from Scikit-learn (CPU) vs. RAPIDS cuML (GPU) as the underlying learner. As datasets grow, GPU-accelerated machine learning can provide significant performance improvements (up to 12x).
Scikit-learn is a networking product designed for enterprise and data center networking.


scikit-learn is a networking product designed for enterprise and data center networking.

Product Information 

This product is a networking product designed for enterprise networks.

CPU is a networking product designed for enterprise networks.

RAPIDS is a networking product designed for enterprise and data center networking.