DA-bench

Visual Benchmark for Data Analytics AI Agents

snowflakeRun #131 (2024-11-19)

Overall Score: 60.7%
(40% Scalability Score + 60% Test Score)

DA-bench Setup for Run #131 — Scalability Score: 71.4% (5 / 7)

Scalability score reflects Snowflake's data warehouse functionality, although the setup video demonstrates uploading csvs.

  • Verified Setup Less Than 20 Minutes
  • Verified Connects to Data Warehouse
  • Verified Handles 1TB Table
  • Verified Handles 10+ Tables
  • Unchecked No Table Structure Changes
  • Unchecked No SQL Expertise for Setup

DA-bench Results for Run #131 — Test Score: 53.5% (115 / 215)

Data Querying (90 / 120)
20 Correct Answers, 2 Hallucinations
Question Date Tested Overall Score Video Recording

dq01
Perform an aggregation on an explicit column

2024-11-19 5

dq02
Perform an aggregation with an explicit table but not an inferred column

2024-11-19 5

dq03
Perform an aggregation with an implicit table and implicit column

2024-11-19 5

dq04
Find and compare information across tables without joins

2024-11-19 5

dq05
Work with non-literal values

2024-11-19 5

dq06
Work with non-literal values and non-SQL data manipulation

2024-11-19 5

dq07
Deal with common acronymns and more advanced aggregations

2024-11-19 5

dq08
Multi-step queries

2024-11-19 5

dq09
Aggregations with numeric predicates to filter

2024-11-19 5

dq10
Aggregations with categorical predicates to filter

2024-11-19 5

dq11
Schema review

2024-11-19 5

dq12
Aggregate records that are filtered with a predicate requiring a join

2024-11-19 5

dq13
Recognizes truly ambiguous queries.

2024-11-19 -5

dq14
Can handle boolean features

2024-11-19 5

dq15
Handles ambiguous column names

2024-11-19 5

dq16
Understands set operates require consideration of overlap

2024-11-19 5

dq17
Finds relevant values inside a Column to answer questions

2024-11-19 -5

dq18
Lookup a single record by ID

2024-11-19 5

dq19
Perform an aggregation by a different name and a second query from that

2024-11-19 5

dq20
Perform an aggregation based on a very different question name

2024-11-19 5

dq21
Perform a filter and an unusually-phrased aggregation in the correct order

2024-11-19 5

dq23
Can handle incorrect column names well

2024-11-19 0

dq24
Schema review

2024-11-19 0

dq25
Work with non-literal values

2024-11-19 5
Domain Knowledge (0 / 5)
0 Correct Answers, 0 Hallucinations
Question Date Tested Overall Score Video Recording

dk01
Column Relevance Determination

2024-11-19 0
Feature Engineering (25 / 40)
6 Correct Answers, 1 Hallucination
Question Date Tested Overall Score Video Recording

fe1
Make a boolean indicator feature for a criteria set

2024-11-19 5

fe2
Make a categorical feature from a criteria set

2024-11-19 5

fe3
Minmax normalization

2024-11-19 5

fe4
Combining two input columns

2024-11-19 5

fe5
Sentiment

2024-11-19 -5

fe6
Phrase Identification in Text

2024-11-19 5

fe7
Advanced NLP

2024-11-19 5

fe8
Advanced NLP

2024-11-19 0
Insight Identification (-5 / 25)
0 Correct Answers, 1 Hallucination
Question Date Tested Overall Score Video Recording

ii2
Compare an aggregation for two distinct subsets of data

2024-11-19 0

ii5
Identifying basic trends on short timelines

2024-11-19 0

ii6
Understands statistical significance

2024-11-19 0

ii7
Understands derivitives

2024-11-19 0

ii8
Can use NLP feature engineering as part of an insight request

2024-11-19 -5
Learning (10 / 10)
2 Correct Answers, 0 Hallucinations
Question Date Tested Overall Score Video Recording

l1
Can remember the meanings of oddly-named columns

2024-11-19 5

l2
Can remember criteria sets under a single name

2024-11-19 5
Visualization (-5 / 15)
0 Correct Answers, 1 Hallucination
Question Date Tested Overall Score Video Recording

v1
Basic Charting

2024-11-19 0

v2
Charting with two series

2024-11-19 -5

v3
Categorical Charts

2024-11-19 0