Frictionless Data and Model Monitoring
Enable AI Observability to detect ML issues faster, deliver continuous improvements, and avoid costly incidents
Any data
Structured or unstructured. Monitor raw data, feature data, predictions and actuals.
Any platform
Batch or streaming. Integrate seamlessly with existing data pipelines and multi-cloud architectures.
Any scale
Go from massive amounts of data to real-time actionable insights in minutes.
AI Observability for everyone.
Free forever, no credit card needed.
Profile 100% of your data. No sampling. No sending data to third parties.
Integrate in minutes. Get data flowing instantly. Get alerted.
Pinpoint data drifts and data quality issues. Get alerts about training-serving skew.
Enable observability for your ML models and data right now
Track model performance continuously, in real time, at any level of granularity
Free edition, onboard in a few minutes
What people are saying about WhyLabs
Easy integration
Integrate in minutes with whylogs, the open-source data logging library
Onboard the WhyLabs SaaS Platform in just three quick steps, on any ML stack
- whylogs Pythonwhylogs Java
Language:
Python
Java
Integration:
Basic
flask
sagemaker
### WhyLabs Platform support for whylogs v1 is coming soon!
### The following WhyLabs Platform integration example requires the latest whylogs v0 version:
### pip install "whylogs<1.0"
import pandas as pd
import os
from whylogs.app import Session
from whylogs.app.writers import WhyLabsWriter
os.environ["WHYLABS_API_KEY"] = "YOUR-API-KEY"
os.environ["WHYLABS_DEFAULT_ORG_ID"] = "YOUR-ORG-ID"
# Adding the WhyLabs Writer to utilize WhyLabs platform
writer = WhyLabsWriter()
session = Session(project="demo-project", pipeline="demo-pipeline", writers=[writer])
# Point to your local CSV if you have your own data
df = pd.read_csv("https://whylabs-public.s3.us-west-2.amazonaws.com/datasets/tour/current.csv")
# Run whylogs on current data and upload to the WhyLabs Platform
# Note: 'datasetId' maps to 'model-id' that is provided when setting-up a model in WhyLabs
with session.logger(tags={"datasetId": "model-1"}) as ylog:
ylog.log_dataframe(df)
Data health / DataOps
Catch missing data, null values, schema changes, and other data quality issues automatically
Prevent training-serving skew by continuously monitoring against a training data baseline
Pinpoint data drifts and data bias before they impact the user experience
Monitor the Feature Store to detect outages and drifts
Model health / ModelOps
Continuously track model outputs and model performance for any model type
Debug model behavior anomalies quickly, with smart correlation and visualization
Get alerted about concept drift and model accuracy degradation
Configure and monitor any custom model metric or KPI
Privacy preserving
WhyLabs profiles model inputs and outputs to capture only statistical profiles of the underlining data
The raw data never leaves the customer VPC/perimeter. All WhyLabs product features operate on statistical profiles
Statistical profiles do not contain proprietary information or PII
All statistical profiles are encrypted during transfer and at rest
Zero maintenance
No schema maintenance. The integration layer automatically infers data schema.
No monitoring configuration. Simply pick your baseline and sensitivity.
No data sampling. WhyLabs profiles 100% of the data to deliver accurate distributions.
No deployment pain. WhyLabs is a SaaS AI Observability layer that suits even the most secure organization.