A visual representation of data generated by operators and checks

Chart

uptrain

Changelog

Blog

Contributing

Community

API Reference

UpTrain

Introduction

Get started with UpTrain in a few simple steps

Quickstart

Why you must use UpTrain in your LLM pipeline

Core Idea

Quickest way to perform evaluations on your data

Evals

Operator

An easy way to chain your operators and create charts

Check

CheckSet

Use Uptrain Client to remotely access the Uptrain toolkit

Overview

Simple and easy pricing. Pay as you go. No hidden fees.

Pricing

Customize your evaluations with UpTrain's framework

Customized Evaluations

How to use UpTrain to experiment with multiple prompts

Prompt Experiments

How to use UpTrain to validate LLM responses

Validation

Performance evaluation - LLM-assisted Q&A

OpenAI Evals

Learn how to create your own custom operator

Custom Operator

How to run your evaluations remotely using the Uptrain API Client

Uptrain API Client

Start monitoring your ML frameworks with UpTrain in minutes

Learn about the key features UpTrain has to offer

Key Features

Concept Drift

Data Drift

Define checks for data integrity in UpTrain

Data Integrity

Edge Case Detection

Measuring inherent bias in Machine Learning models

Model Bias

Monitor metrics specific to your ML model

Custom Monitors

Use predefined and custom measurables to track your model's performance.

Predefined measurables available in UpTrain and how to use them

Predefined

Creating and using custom measurables in UpTrain

Custom

Plot Line charts, Bar charts, Histograms and more

Plot Visualization

Understand the Relative Importance of Input Features on Predictions

SHAP Explainability

UMAP Dimensionality Reduction for Data Visualization

UMAP Visualization

t-SNE Dimensionality Reduction

Measure distances between data points to analyze the model

Distances

Evaluate the performance of embedding algorithms

Convergence Analysis

Monitoring performance of a text summarization task in NLP

Text Summarization

Monitoring performance of a fraud detection task

Fraud Detection

Measuring data drift in production data for automated retraining of ML model

Human Orientation Classifier

Quickly explore the features of UpTrain in more detail

Deep Dive Examples

Monitoring Biases in a Recommender System

Shopping Cart Recommendation

Ride Time Estimation

Masked Language Modeling

Collecting a finetuning dataset to improve a model that summarizes conversations

Getting Started 🚀

Key Components 🗝️

Managed Services ☁️

Tutorials 📑

Chart