Dataset Viewer
Auto-converted to Parquet Duplicate
name
stringlengths
38
40
instruction
stringlengths
800
931
description
stringlengths
60
173
keywords
listlengths
3
3
schema_version
stringclasses
1 value
config
unknown
metadata
unknown
files
listlengths
6
6
location
stringlengths
26
28
smoldataenvs-eval/0000_886_886039_qa_5
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - battles.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install more if needed...
How many times did Balon/Euron Greyjoy defeat Robb Stark in the dataset?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0000_886_886039_qa_5", "description": "How many times did Balon/Euron Greyjoy defeat Robb Stark in the dataset?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis", "kaggle" ] }, "...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0000/886/886039.ipynb_qa_5", "kaggle_dataset_name": "mylesoneill/game-of-thrones", "gold_answer": "4", "reward_mode_initial": "numeric", "package_tier": 1, "difficulty_level": 2, "difficulty": "medium", "difficulty_tier": "me...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0000_886_886039_qa_5
smoldataenvs-eval/0000_901_901543_qa_2
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - survey.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install more if needed)...
What numerical value was assigned to the 'no_employees' category 'More than 1000' during preprocessing?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0000_901_901543_qa_2", "description": "What numerical value was assigned to the 'no_employees' category 'More than 1000' during preprocessing?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis"...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0000/901/901543.ipynb_qa_2", "kaggle_dataset_name": "osmi/mental-health-in-tech-survey", "gold_answer": "6", "reward_mode_initial": "numeric", "package_tier": 1, "difficulty_level": 2, "difficulty": "medium", "difficulty_tier...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0000_901_901543_qa_2
smoldataenvs-eval/0000_901_901543_qa_3
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - survey.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install more if needed)...
How many columns are present in the dataset after removing the 'Timestamp' column?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0000_901_901543_qa_3", "description": "How many columns are present in the dataset after removing the 'Timestamp' column?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis", "kaggle" ...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0000/901/901543.ipynb_qa_3", "kaggle_dataset_name": "osmi/mental-health-in-tech-survey", "gold_answer": "26", "reward_mode_initial": "numeric", "package_tier": 1, "difficulty_level": 1, "difficulty": "easy", "difficulty_tier"...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0000_901_901543_qa_3
smoldataenvs-eval/0000_903_903504_qa_1
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - (see /home/user/input) Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install mor...
What is the correlation coefficient between Air Temperature and Sea Surface Temperature in the dataset?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0000_903_903504_qa_1", "description": "What is the correlation coefficient between Air Temperature and Sea Surface Temperature in the dataset?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis"...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0000/903/903504.ipynb_qa_1", "kaggle_dataset_name": "uciml/el-nino-dataset", "gold_answer": "0.93", "reward_mode_initial": "numeric", "package_tier": 0, "difficulty_level": 2, "difficulty": "medium", "difficulty_tier": "mediu...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0000_903_903504_qa_1
smoldataenvs-eval/0000_939_939639_qa_2
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - Pokemon.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install more if needed...
What is the total number of rows in the dataset after using the pd.melt() function to transform the data for the swarmplot visualization?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0000_939_939639_qa_2", "description": "What is the total number of rows in the dataset after using the pd.melt() function to transform the data for the swarmplot visualization?", "authors": [], "keywords": [ "sm...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0000/939/939639.ipynb_qa_2", "kaggle_dataset_name": "abcsds/pokemon", "gold_answer": "4800", "reward_mode_initial": "numeric", "package_tier": 1, "difficulty_level": 2, "difficulty": "medium", "difficulty_tier": "medium" }
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0000_939_939639_qa_2
smoldataenvs-eval/0000_939_939639_qa_3
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - Pokemon.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install more if needed...
What is the number of unique statistical categories (e.g., HP, Attack, Defense, etc.) included in the transformed dataset used for the swarmplot visualization?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0000_939_939639_qa_3", "description": "What is the number of unique statistical categories (e.g., HP, Attack, Defense, etc.) included in the transformed dataset used for the swarmplot visualization?", "authors": [], "...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0000/939/939639.ipynb_qa_3", "kaggle_dataset_name": "abcsds/pokemon", "gold_answer": "6", "reward_mode_initial": "numeric", "package_tier": 1, "difficulty_level": 2, "difficulty": "medium", "difficulty_tier": "medium" }
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0000_939_939639_qa_3
smoldataenvs-eval/0000_949_949403_qa_3
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - Video_Games_Sales_as_at_22_Dec_2016.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly...
Which publisher is associated with the most games in the dataset?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0000_949_949403_qa_3", "description": "Which publisher is associated with the most games in the dataset?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis", "kaggle" ] }, "metadat...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0000/949/949403.ipynb_qa_3", "kaggle_dataset_name": "rush4ratio/video-game-sales-with-ratings", "gold_answer": "Electronic Arts", "reward_mode_initial": "exact_short", "package_tier": 1, "difficulty_level": 1, "difficulty": "ea...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0000_949_949403_qa_3
smoldataenvs-eval/0000_975_975906_qa_1
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - database.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install more if neede...
What is the most common weapon used in unsolved homicides in the dataset?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0000_975_975906_qa_1", "description": "What is the most common weapon used in unsolved homicides in the dataset?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis", "kaggle" ] }, ...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0000/975/975906.ipynb_qa_1", "kaggle_dataset_name": "murderaccountability/homicide-reports", "gold_answer": "Handgun", "reward_mode_initial": "exact_short", "package_tier": 1, "difficulty_level": 2, "difficulty": "medium", "d...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0000_975_975906_qa_1
smoldataenvs-eval/0000_986_986420_qa_2
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - creditcard.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install more if nee...
What is the accuracy of a model that predicts all transactions as non-fraudulent in this dataset?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0000_986_986420_qa_2", "description": "What is the accuracy of a model that predicts all transactions as non-fraudulent in this dataset?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis", ...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0000/986/986420.ipynb_qa_2", "kaggle_dataset_name": "mlg-ulb/creditcardfraud", "gold_answer": "99.83", "reward_mode_initial": "numeric", "package_tier": 1, "difficulty_level": 2, "difficulty": "medium", "difficulty_tier": "me...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0000_986_986420_qa_2
smoldataenvs-eval/0000_986_986420_qa_4
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - creditcard.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install more if nee...
What percentage of the transactions in the dataset are classified as non-fraudulent?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0000_986_986420_qa_4", "description": "What percentage of the transactions in the dataset are classified as non-fraudulent?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis", "kaggle" ...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0000/986/986420.ipynb_qa_4", "kaggle_dataset_name": "mlg-ulb/creditcardfraud", "gold_answer": "99.83", "reward_mode_initial": "numeric", "package_tier": 1, "difficulty_level": 1, "difficulty": "easy", "difficulty_tier": "easy...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0000_986_986420_qa_4
smoldataenvs-eval/0000_988_988342_qa_4
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - sources.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install more if needed...
What is the most common cause of tsunamis in the dataset when excluding 'Unknown' and 'Questionable Earthquake' classifications?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0000_988_988342_qa_4", "description": "What is the most common cause of tsunamis in the dataset when excluding 'Unknown' and 'Questionable Earthquake' classifications?", "authors": [], "keywords": [ "smoldataenv...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0000/988/988342.ipynb_qa_4", "kaggle_dataset_name": "noaa/seismic-waves", "gold_answer": "Earthquake", "reward_mode_initial": "exact_short", "package_tier": 0, "difficulty_level": 2, "difficulty": "medium", "difficulty_tier":...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0000_988_988342_qa_4
smoldataenvs-eval/0000_988_988657_qa_4
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - archive.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install more if needed...
How many individuals from India have been selected as Person of the Year in the dataset?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0000_988_988657_qa_4", "description": "How many individuals from India have been selected as Person of the Year in the dataset?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis", "kaggle...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0000/988/988657.ipynb_qa_4", "kaggle_dataset_name": "timemagazine/magazine-covers", "gold_answer": "1", "reward_mode_initial": "numeric", "package_tier": 0, "difficulty_level": 2, "difficulty": "medium", "difficulty_tier": "m...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0000_988_988657_qa_4
smoldataenvs-eval/0000_991_991010_qa_1
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - steam-200k.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install more if nee...
Which game has the highest total cumulative playing time in hours across all users?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0000_991_991010_qa_1", "description": "Which game has the highest total cumulative playing time in hours across all users?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis", "kaggle" ...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0000/991/991010.ipynb_qa_1", "kaggle_dataset_name": "tamber/steam-video-games", "gold_answer": "Dota 2", "reward_mode_initial": "exact_short", "package_tier": 3, "difficulty_level": 2, "difficulty": "medium", "difficulty_tier...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0000_991_991010_qa_1
smoldataenvs-eval/0000_991_991010_qa_2
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - steam-200k.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install more if nee...
What is the average number of games played per user in the dataset?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0000_991_991010_qa_2", "description": "What is the average number of games played per user in the dataset?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis", "kaggle" ] }, "metad...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0000/991/991010.ipynb_qa_2", "kaggle_dataset_name": "tamber/steam-video-games", "gold_answer": "6.21", "reward_mode_initial": "numeric", "package_tier": 3, "difficulty_level": 2, "difficulty": "medium", "difficulty_tier": "me...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0000_991_991010_qa_2
smoldataenvs-eval/0000_991_991010_qa_3
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - steam-200k.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install more if nee...
What percentage of all games played that exceed 1,000 hours are dedicated to Dota 2?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0000_991_991010_qa_3", "description": "What percentage of all games played that exceed 1,000 hours are dedicated to Dota 2?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis", "kaggle" ...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0000/991/991010.ipynb_qa_3", "kaggle_dataset_name": "tamber/steam-video-games", "gold_answer": "49.1987", "reward_mode_initial": "numeric", "package_tier": 3, "difficulty_level": 2, "difficulty": "medium", "difficulty_tier": ...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0000_991_991010_qa_3
smoldataenvs-eval/0000_994_994384_qa_5
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - steam-200k.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install more if nee...
How many users have played at least one game in the dataset?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0000_994_994384_qa_5", "description": "How many users have played at least one game in the dataset?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis", "kaggle" ] }, "metadata": {...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0000/994/994384.ipynb_qa_5", "kaggle_dataset_name": "tamber/steam-video-games", "gold_answer": "11350", "reward_mode_initial": "numeric", "package_tier": 1, "difficulty_level": 2, "difficulty": "medium", "difficulty_tier": "m...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0000_994_994384_qa_5
smoldataenvs-eval/0000_994_994443_qa_4
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - spam.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install more if needed). ...
How many spam messages in the dataset contain phone numbers with four or more digits?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0000_994_994443_qa_4", "description": "How many spam messages in the dataset contain phone numbers with four or more digits?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis", "kaggle" ...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0000/994/994443.ipynb_qa_4", "kaggle_dataset_name": "uciml/sms-spam-collection-dataset", "gold_answer": "623", "reward_mode_initial": "numeric", "package_tier": 1, "difficulty_level": 2, "difficulty": "medium", "difficulty_ti...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0000_994_994443_qa_4
smoldataenvs-eval/0001_004_1004716_qa_2
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - glass.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install more if needed)....
What is the most frequent category in the 'Type' column of the dataset?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0001_004_1004716_qa_2", "description": "What is the most frequent category in the 'Type' column of the dataset?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis", "kaggle" ] }, "...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0001/004/1004716.ipynb_qa_2", "kaggle_dataset_name": "uciml/glass", "gold_answer": "Type 2", "reward_mode_initial": "exact_short", "package_tier": 1, "difficulty_level": 1, "difficulty": "easy", "difficulty_tier": "easy" }
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0001_004_1004716_qa_2
smoldataenvs-eval/0001_024_1024990_qa_2
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - glass.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install more if needed)....
What is the total number of samples across all glass types in the dataset?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0001_024_1024990_qa_2", "description": "What is the total number of samples across all glass types in the dataset?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis", "kaggle" ] }, ...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0001/024/1024990.ipynb_qa_2", "kaggle_dataset_name": "uciml/glass", "gold_answer": "214", "reward_mode_initial": "numeric", "package_tier": 1, "difficulty_level": 1, "difficulty": "easy", "difficulty_tier": "easy" }
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0001_024_1024990_qa_2
smoldataenvs-eval/0001_027_1027372_qa_2
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - pokemonGO.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install more if need...
How many Pokémon in the dataset are classified as dual-type (have a non-null Type 2)?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0001_027_1027372_qa_2", "description": "How many Pokémon in the dataset are classified as dual-type (have a non-null Type 2)?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis", "kaggle" ...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0001/027/1027372.ipynb_qa_2", "kaggle_dataset_name": "abcsds/pokemongo", "gold_answer": "67", "reward_mode_initial": "numeric", "package_tier": 1, "difficulty_level": 1, "difficulty": "easy", "difficulty_tier": "easy" }
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0001_027_1027372_qa_2
smoldataenvs-eval/0001_029_1029150_qa_2
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - database.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install more if neede...
Which state has the highest total number of homicide crimes recorded in the dataset?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0001_029_1029150_qa_2", "description": "Which state has the highest total number of homicide crimes recorded in the dataset?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis", "kaggle" ...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0001/029/1029150.ipynb_qa_2", "kaggle_dataset_name": "murderaccountability/homicide-reports", "gold_answer": "California", "reward_mode_initial": "exact_short", "package_tier": 1, "difficulty_level": 2, "difficulty": "medium", ...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0001_029_1029150_qa_2
smoldataenvs-eval/0001_036_1036744_qa_1
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - My Uber Drives - 2016.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install ...
Which ride purpose has the highest average miles driven per ride in the dataset?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0001_036_1036744_qa_1", "description": "Which ride purpose has the highest average miles driven per ride in the dataset?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis", "kaggle" ]...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0001/036/1036744.ipynb_qa_1", "kaggle_dataset_name": "zusmani/uberdrives", "gold_answer": "Commute", "reward_mode_initial": "exact_short", "package_tier": 1, "difficulty_level": 2, "difficulty": "medium", "difficulty_tier": "...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0001_036_1036744_qa_1
smoldataenvs-eval/0001_042_1042725_qa_2
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - marathon_results_2016.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install ...
Which country has the highest number of top 100 female marathon runners in the dataset?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0001_042_1042725_qa_2", "description": "Which country has the highest number of top 100 female marathon runners in the dataset?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis", "kaggle...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0001/042/1042725.ipynb_qa_2", "kaggle_dataset_name": "rojour/boston-results", "gold_answer": "USA", "reward_mode_initial": "exact_short", "package_tier": 3, "difficulty_level": 2, "difficulty": "medium", "difficulty_tier": "m...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0001_042_1042725_qa_2
smoldataenvs-eval/0001_044_1044175_qa_3
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - My Uber Drives - 2016.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install ...
What is the total number of miles driven by the Uber driver according to the dataset, as indicated in the "Totals" row before it was removed?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0001_044_1044175_qa_3", "description": "What is the total number of miles driven by the Uber driver according to the dataset, as indicated in the \"Totals\" row before it was removed?", "authors": [], "keywords": [ ...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0001/044/1044175.ipynb_qa_3", "kaggle_dataset_name": "zusmani/uberdrives", "gold_answer": "12204.7", "reward_mode_initial": "numeric", "package_tier": 0, "difficulty_level": 1, "difficulty": "easy", "difficulty_tier": "easy" ...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0001_044_1044175_qa_3
smoldataenvs-eval/0001_062_1062946_qa_2
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - LeagueofLegends.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install more i...
Which team has the highest number of games played as the blue team in the dataset?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0001_062_1062946_qa_2", "description": "Which team has the highest number of games played as the blue team in the dataset?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis", "kaggle" ...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0001/062/1062946.ipynb_qa_2", "kaggle_dataset_name": "chuckephron/leagueoflegends", "gold_answer": "SKT", "reward_mode_initial": "exact_short", "package_tier": 1, "difficulty_level": 1, "difficulty": "easy", "difficulty_tier"...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0001_062_1062946_qa_2
smoldataenvs-eval/0001_065_1065379_qa_3
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - pml-training.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install more if n...
What is the most frequently occurring activity class in the training dataset based on the distribution visualization?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0001_065_1065379_qa_3", "description": "What is the most frequently occurring activity class in the training dataset based on the distribution visualization?", "authors": [], "keywords": [ "smoldataenvs", ...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0001/065/1065379.ipynb_qa_3", "kaggle_dataset_name": "athniv/exercisepatternpredict", "gold_answer": "A", "reward_mode_initial": "exact_short", "package_tier": 1, "difficulty_level": 1, "difficulty": "easy", "difficulty_tier"...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0001_065_1065379_qa_3
smoldataenvs-eval/0001_067_1067596_qa_5
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - LowRes_13434_overlapping_pairs.h5 Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip ...
How many channels are present in each image pair in the dataset?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0001_067_1067596_qa_5", "description": "How many channels are present in each image pair in the dataset?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis", "kaggle" ] }, "metadat...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0001/067/1067596.ipynb_qa_5", "kaggle_dataset_name": "jeanpat/overlapping-chromosomes", "gold_answer": "2", "reward_mode_initial": "numeric", "package_tier": 3, "difficulty_level": 2, "difficulty": "medium", "difficulty_tier"...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0001_067_1067596_qa_5
smoldataenvs-eval/0001_068_1068916_qa_1
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - (see /home/user/input) Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install mor...
What is the average spam score across all entries in the dataset?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0001_068_1068916_qa_1", "description": "What is the average spam score across all entries in the dataset?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis", "kaggle" ] }, "metada...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0001/068/1068916.ipynb_qa_1", "kaggle_dataset_name": "mrisdal/fake-news", "gold_answer": "0.026", "reward_mode_initial": "numeric", "package_tier": 0, "difficulty_level": 1, "difficulty": "easy", "difficulty_tier": "easy" }
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0001_068_1068916_qa_1
smoldataenvs-eval/0001_068_1068916_qa_5
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - fake.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install more if needed). ...
What is the average number of comments per entry in the dataset?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0001_068_1068916_qa_5", "description": "What is the average number of comments per entry in the dataset?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis", "kaggle" ] }, "metadat...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0001/068/1068916.ipynb_qa_5", "kaggle_dataset_name": "mrisdal/fake-news", "gold_answer": "0.038", "reward_mode_initial": "numeric", "package_tier": 1, "difficulty_level": 1, "difficulty": "easy", "difficulty_tier": "easy" }
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0001_068_1068916_qa_5
smoldataenvs-eval/0001_078_1078204_qa_2
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - LeagueofLegends.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install more i...
What is the average duration of games in which SKT T1 participated?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0001_078_1078204_qa_2", "description": "What is the average duration of games in which SKT T1 participated?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis", "kaggle" ] }, "meta...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0001/078/1078204.ipynb_qa_2", "kaggle_dataset_name": "chuckephron/leagueoflegends", "gold_answer": "37.79", "reward_mode_initial": "numeric", "package_tier": 1, "difficulty_level": 2, "difficulty": "medium", "difficulty_tier"...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0001_078_1078204_qa_2
smoldataenvs-eval/0001_079_1079149_qa_2
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - spam.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install more if needed). ...
What is the average word count difference between spam and ham messages?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0001_079_1079149_qa_2", "description": "What is the average word count difference between spam and ham messages?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis", "kaggle" ] }, ...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0001/079/1079149.ipynb_qa_2", "kaggle_dataset_name": "uciml/sms-spam-collection-dataset", "gold_answer": "9.57", "reward_mode_initial": "numeric", "package_tier": 2, "difficulty_level": 2, "difficulty": "medium", "difficulty_...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0001_079_1079149_qa_2
smoldataenvs-eval/0001_082_1082305_qa_1
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - mushrooms.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install more if need...
What is the proportion of edible mushrooms (class 0) in the dataset compared to the total number of mushrooms?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0001_082_1082305_qa_1", "description": "What is the proportion of edible mushrooms (class 0) in the dataset compared to the total number of mushrooms?", "authors": [], "keywords": [ "smoldataenvs", "data-a...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0001/082/1082305.ipynb_qa_1", "kaggle_dataset_name": "uciml/mushroom-classification", "gold_answer": "51.8", "reward_mode_initial": "numeric", "package_tier": 1, "difficulty_level": 2, "difficulty": "medium", "difficulty_tier...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0001_082_1082305_qa_1
smoldataenvs-eval/0001_083_1083006_qa_5
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - Crimes_2012-2016.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install more ...
Which crime type is most frequently associated with the "Invest Cont" status in the dataset?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0001_083_1083006_qa_5", "description": "Which crime type is most frequently associated with the \"Invest Cont\" status in the dataset?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis", ...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0001/083/1083006.ipynb_qa_5", "kaggle_dataset_name": "kingburrito666/los-angeles-crime", "gold_answer": "TRAFFIC DR #", "reward_mode_initial": "exact_short", "package_tier": 0, "difficulty_level": 2, "difficulty": "medium", "...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0001_083_1083006_qa_5
smoldataenvs-eval/0001_091_1091047_qa_4
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - (see /home/user/input) Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install mor...
What is the median trip duration in seconds after removing the top 0.5% of outlier durations?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0001_091_1091047_qa_4", "description": "What is the median trip duration in seconds after removing the top 0.5% of outlier durations?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis", "...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0001/091/1091047.ipynb_qa_4", "kaggle_dataset_name": "benhamner/sf-bay-area-bike-share", "gold_answer": "515", "reward_mode_initial": "numeric", "package_tier": 0, "difficulty_level": 2, "difficulty": "medium", "difficulty_ti...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0001_091_1091047_qa_4
smoldataenvs-eval/0001_091_1091047_qa_5
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - trip.csv - station.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install mor...
What is the total number of unique days represented in the dataset?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0001_091_1091047_qa_5", "description": "What is the total number of unique days represented in the dataset?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis", "kaggle" ] }, "meta...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0001/091/1091047.ipynb_qa_5", "kaggle_dataset_name": "benhamner/sf-bay-area-bike-share", "gold_answer": "733", "reward_mode_initial": "numeric", "package_tier": 1, "difficulty_level": 2, "difficulty": "medium", "difficulty_ti...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0001_091_1091047_qa_5
smoldataenvs-eval/0001_095_1095486_qa_2
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - 911.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install more if needed). ...
Which day of the week has the lowest number of traffic-related 911 calls according to the analysis?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0001_095_1095486_qa_2", "description": "Which day of the week has the lowest number of traffic-related 911 calls according to the analysis?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis", ...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0001/095/1095486.ipynb_qa_2", "kaggle_dataset_name": "mchirico/montcoalert", "gold_answer": "Sunday", "reward_mode_initial": "exact_short", "package_tier": 1, "difficulty_level": 2, "difficulty": "medium", "difficulty_tier": ...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0001_095_1095486_qa_2
smoldataenvs-eval/0001_102_1102505_qa_2
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - up_res.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install more if needed)...
Which political party secured the highest total number of votes in the UP election based on the dataset analysis?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0001_102_1102505_qa_2", "description": "Which political party secured the highest total number of votes in the UP election based on the dataset analysis?", "authors": [], "keywords": [ "smoldataenvs", "dat...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0001/102/1102505.ipynb_qa_2", "kaggle_dataset_name": "ankit2106/uttar-pradesh-assembly-elections-2017", "gold_answer": "BJP+", "reward_mode_initial": "exact_short", "package_tier": 1, "difficulty_level": 2, "difficulty": "mediu...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0001_102_1102505_qa_2
smoldataenvs-eval/0001_104_1104901_qa_2
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - Tweets.csv - database.sqlite Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip insta...
What is the most common reason cited for negative tweets in the dataset?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0001_104_1104901_qa_2", "description": "What is the most common reason cited for negative tweets in the dataset?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis", "kaggle" ] }, ...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0001/104/1104901.ipynb_qa_2", "kaggle_dataset_name": "crowdflower/twitter-airline-sentiment", "gold_answer": "Customer Service Issue", "reward_mode_initial": "exact_short", "package_tier": 1, "difficulty_level": 2, "difficulty"...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0001_104_1104901_qa_2
smoldataenvs-eval/0001_104_1104901_qa_3
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - Tweets.csv - database.sqlite Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip insta...
Which airline received the highest number of neutral sentiment tweets in the dataset?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0001_104_1104901_qa_3", "description": "Which airline received the highest number of neutral sentiment tweets in the dataset?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis", "kaggle" ...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0001/104/1104901.ipynb_qa_3", "kaggle_dataset_name": "crowdflower/twitter-airline-sentiment", "gold_answer": "Delta", "reward_mode_initial": "exact_short", "package_tier": 1, "difficulty_level": 2, "difficulty": "medium", "di...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0001_104_1104901_qa_3
smoldataenvs-eval/0001_104_1104959_qa_2
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - Tweets.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install more if needed)...
What is the most common sentiment category in the dataset based on tweet distribution?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0001_104_1104959_qa_2", "description": "What is the most common sentiment category in the dataset based on tweet distribution?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis", "kaggle"...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0001/104/1104959.ipynb_qa_2", "kaggle_dataset_name": "crowdflower/twitter-airline-sentiment", "gold_answer": "Negative", "reward_mode_initial": "exact_short", "package_tier": 3, "difficulty_level": 1, "difficulty": "easy", "d...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0001_104_1104959_qa_2
smoldataenvs-eval/0001_105_1105529_qa_5
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - 2015.csv - 2016.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install more i...
What is the strongest negative correlation with the Happiness Score in the 2015 dataset?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0001_105_1105529_qa_5", "description": "What is the strongest negative correlation with the Happiness Score in the 2015 dataset?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis", "kaggl...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0001/105/1105529.ipynb_qa_5", "kaggle_dataset_name": "unsdsn/world-happiness", "gold_answer": "Happiness Rank", "reward_mode_initial": "exact_short", "package_tier": 1, "difficulty_level": 2, "difficulty": "medium", "difficul...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0001_105_1105529_qa_5
smoldataenvs-eval/0001_105_1105892_qa_3
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - hall_of_fame.csv - player.csv - batting.csv - salary.csv - player_award.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, st...
What is the maximum number of times a single player was considered for Hall of Fame induction across all years?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0001_105_1105892_qa_3", "description": "What is the maximum number of times a single player was considered for Hall of Fame induction across all years?", "authors": [], "keywords": [ "smoldataenvs", "data-...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0001/105/1105892.ipynb_qa_3", "kaggle_dataset_name": "seanlahman/the-history-of-baseball", "gold_answer": "20", "reward_mode_initial": "numeric", "package_tier": 1, "difficulty_level": 2, "difficulty": "medium", "difficulty_t...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0001_105_1105892_qa_3
smoldataenvs-eval/0001_105_1105892_qa_5
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - (see /home/user/input) Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install mor...
What is the average number of awards received by players who received any awards in the dataset?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0001_105_1105892_qa_5", "description": "What is the average number of awards received by players who received any awards in the dataset?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis", ...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0001/105/1105892.ipynb_qa_5", "kaggle_dataset_name": "seanlahman/the-history-of-baseball", "gold_answer": "4.59", "reward_mode_initial": "numeric", "package_tier": 0, "difficulty_level": 2, "difficulty": "medium", "difficulty...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0001_105_1105892_qa_5
smoldataenvs-eval/0001_116_1116342_qa_3
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - mushrooms.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install more if need...
What is the most frequent cap surface type in the original dataset before encoding?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0001_116_1116342_qa_3", "description": "What is the most frequent cap surface type in the original dataset before encoding?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis", "kaggle" ...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0001/116/1116342.ipynb_qa_3", "kaggle_dataset_name": "uciml/mushroom-classification", "gold_answer": "y", "reward_mode_initial": "exact_short", "package_tier": 1, "difficulty_level": 1, "difficulty": "easy", "difficulty_tier"...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0001_116_1116342_qa_3
smoldataenvs-eval/0001_116_1116342_qa_5
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - mushrooms.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install more if need...
What is the most common cap color category in the original dataset before encoding?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0001_116_1116342_qa_5", "description": "What is the most common cap color category in the original dataset before encoding?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis", "kaggle" ...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0001/116/1116342.ipynb_qa_5", "kaggle_dataset_name": "uciml/mushroom-classification", "gold_answer": "n", "reward_mode_initial": "exact_short", "package_tier": 1, "difficulty_level": 1, "difficulty": "easy", "difficulty_tier"...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0001_116_1116342_qa_5
smoldataenvs-eval/0001_123_1123075_qa_4
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - up_res.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install more if needed)...
Which constituency (ac) had the highest total votes secured by BJP+ candidates?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0001_123_1123075_qa_4", "description": "Which constituency (ac) had the highest total votes secured by BJP+ candidates?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis", "kaggle" ] ...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0001/123/1123075.ipynb_qa_4", "kaggle_dataset_name": "ankit2106/uttar-pradesh-assembly-elections-2017", "gold_answer": "Sahibabad", "reward_mode_initial": "exact_short", "package_tier": 1, "difficulty_level": 2, "difficulty": "...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0001_123_1123075_qa_4
smoldataenvs-eval/0001_124_1124736_qa_1
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - creditcard.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install more if nee...
What percentage of transactions in the dataset are classified as fraudulent?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0001_124_1124736_qa_1", "description": "What percentage of transactions in the dataset are classified as fraudulent?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis", "kaggle" ] }...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0001/124/1124736.ipynb_qa_1", "kaggle_dataset_name": "mlg-ulb/creditcardfraud", "gold_answer": "0.1727", "reward_mode_initial": "numeric", "package_tier": 1, "difficulty_level": 1, "difficulty": "easy", "difficulty_tier": "ea...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0001_124_1124736_qa_1
smoldataenvs-eval/0001_124_1124736_qa_2
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - creditcard.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install more if nee...
What is the maximum transaction amount recorded in the dataset?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0001_124_1124736_qa_2", "description": "What is the maximum transaction amount recorded in the dataset?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis", "kaggle" ] }, "metadata...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0001/124/1124736.ipynb_qa_2", "kaggle_dataset_name": "mlg-ulb/creditcardfraud", "gold_answer": "25691.16", "reward_mode_initial": "numeric", "package_tier": 1, "difficulty_level": 1, "difficulty": "easy", "difficulty_tier": "...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0001_124_1124736_qa_2
smoldataenvs-eval/0001_133_1133224_qa_4
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - WA_Fn-UseC_-HR-Employee-Attrition.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (...
What is the total number of features removed during the classification process?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0001_133_1133224_qa_4", "description": "What is the total number of features removed during the classification process?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis", "kaggle" ] ...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0001/133/1133224.ipynb_qa_4", "kaggle_dataset_name": "pavansubhasht/ibm-hr-analytics-attrition-dataset", "gold_answer": "4", "reward_mode_initial": "numeric", "package_tier": 0, "difficulty_level": 2, "difficulty": "medium", ...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0001_133_1133224_qa_4
smoldataenvs-eval/0001_133_1133625_qa_3
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - (see /home/user/input) Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install mor...
What is the difference between the highest and lowest number of votes received by any BJP+ candidate?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0001_133_1133625_qa_3", "description": "What is the difference between the highest and lowest number of votes received by any BJP+ candidate?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis",...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0001/133/1133625.ipynb_qa_3", "kaggle_dataset_name": "ankit2106/uttar-pradesh-assembly-elections-2017", "gold_answer": "256219", "reward_mode_initial": "numeric", "package_tier": 0, "difficulty_level": 2, "difficulty": "medium"...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0001_133_1133625_qa_3
smoldataenvs-eval/0001_134_1134780_qa_3
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - up_res.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install more if needed)...
How many candidates were affiliated with the "others" category?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0001_134_1134780_qa_3", "description": "How many candidates were affiliated with the \"others\" category?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis", "kaggle" ] }, "metada...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0001/134/1134780.ipynb_qa_3", "kaggle_dataset_name": "ankit2106/uttar-pradesh-assembly-elections-2017", "gold_answer": "1883", "reward_mode_initial": "numeric", "package_tier": 0, "difficulty_level": 2, "difficulty": "medium", ...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0001_134_1134780_qa_3
smoldataenvs-eval/0001_134_1134780_qa_5
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - up_res.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install more if needed)...
What is the total number of candidates from the "INC" party?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0001_134_1134780_qa_5", "description": "What is the total number of candidates from the \"INC\" party?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis", "kaggle" ] }, "metadata"...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0001/134/1134780.ipynb_qa_5", "kaggle_dataset_name": "ankit2106/uttar-pradesh-assembly-elections-2017", "gold_answer": "114", "reward_mode_initial": "numeric", "package_tier": 0, "difficulty_level": 2, "difficulty": "medium", ...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0001_134_1134780_qa_5
smoldataenvs-eval/0001_548_1548552_qa_1
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - CompleteDataset.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install more i...
Which football club has the highest average Overall rating among its players in the dataset?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0001_548_1548552_qa_1", "description": "Which football club has the highest average Overall rating among its players in the dataset?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis", "k...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0001/548/1548552.ipynb_qa_1", "kaggle_dataset_name": "thec03u5/fifa-18-demo-player-dataset", "gold_answer": "FC Barcelona", "reward_mode_initial": "exact_short", "package_tier": 1, "difficulty_level": 3, "difficulty": "medium",...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0001_548_1548552_qa_1
smoldataenvs-eval/0001_551_1551019_qa_3
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - data_set_ALL_AML_independent.csv - data_set_ALL_AML_train.csv - actual.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, sta...
What is the total number of patients in the training dataset after merging with patient cancer labels?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0001_551_1551019_qa_3", "description": "What is the total number of patients in the training dataset after merging with patient cancer labels?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis"...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0001/551/1551019.ipynb_qa_3", "kaggle_dataset_name": "crawford/gene-expression", "gold_answer": "38", "reward_mode_initial": "numeric", "package_tier": 1, "difficulty_level": 3, "difficulty": "medium", "difficulty_tier": "med...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0001_551_1551019_qa_3
smoldataenvs-eval/0001_573_1573460_qa_3
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - utils.py Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install more if needed). ...
What is the total duration of the training data (train_cat and train_dog) in seconds?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0001_573_1573460_qa_3", "description": "What is the total duration of the training data (train_cat and train_dog) in seconds?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis", "kaggle" ...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0001/573/1573460.ipynb_qa_3", "kaggle_dataset_name": "mmoreaux/audio-cats-and-dogs", "gold_answer": "1259.28625", "reward_mode_initial": "numeric", "package_tier": 2, "difficulty_level": 3, "difficulty": "medium", "difficulty...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0001_573_1573460_qa_3
smoldataenvs-eval/0001_574_1574525_qa_2
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - all_annotated.tsv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install more if ...
What percentage of tweets were identified as English by langid?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0001_574_1574525_qa_2", "description": "What percentage of tweets were identified as English by langid?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis", "kaggle" ] }, "metadata...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0001/574/1574525.ipynb_qa_2", "kaggle_dataset_name": "rtatman/the-umass-global-english-on-twitter-dataset", "gold_answer": "40.96", "reward_mode_initial": "numeric", "package_tier": 3, "difficulty_level": 3, "difficulty": "medi...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0001_574_1574525_qa_2
smoldataenvs-eval/0001_583_1583897_qa_5
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - xAPI-Edu-Data.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install more if ...
Which topic shows the largest absolute difference in student count between the highest (H) and lowest (L) performance categories from the crosstab analysis?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0001_583_1583897_qa_5", "description": "Which topic shows the largest absolute difference in student count between the highest (H) and lowest (L) performance categories from the crosstab analysis?", "authors": [], "ke...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0001/583/1583897.ipynb_qa_5", "kaggle_dataset_name": "aljarah/xAPI-Edu-Data", "gold_answer": "IT", "reward_mode_initial": "exact_short", "package_tier": 1, "difficulty_level": 3, "difficulty": "medium", "difficulty_tier": "me...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0001_583_1583897_qa_5
smoldataenvs-eval/0001_586_1586397_qa_4
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - (see /home/user/input) Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install mor...
Which year had the highest combined total of injured and fatality victims from mass shootings according to the grouped analysis?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0001_586_1586397_qa_4", "description": "Which year had the highest combined total of injured and fatality victims from mass shootings according to the grouped analysis?", "authors": [], "keywords": [ "smoldataen...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0001/586/1586397.ipynb_qa_4", "kaggle_dataset_name": "zusmani/us-mass-shootings-last-50-years", "gold_answer": "2017", "reward_mode_initial": "exact_short", "package_tier": 0, "difficulty_level": 3, "difficulty": "medium", "d...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0001_586_1586397_qa_4
smoldataenvs-eval/0001_594_1594994_qa_1
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - Mass Shootings Dataset.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install...
What was the highest number of fatalities recorded in a single year according to the mass shooting dataset?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0001_594_1594994_qa_1", "description": "What was the highest number of fatalities recorded in a single year according to the mass shooting dataset?", "authors": [], "keywords": [ "smoldataenvs", "data-anal...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0001/594/1594994.ipynb_qa_1", "kaggle_dataset_name": "zusmani/us-mass-shootings-last-50-years", "gold_answer": "255", "reward_mode_initial": "numeric", "package_tier": 1, "difficulty_level": 3, "difficulty": "medium", "diffic...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0001_594_1594994_qa_1
smoldataenvs-eval/0001_598_1598064_qa_2
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - Family Income and Expenditure.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip ...
Which expenditure category shows the highest correlation with Total Household Income based on the correlation matrix analysis?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0001_598_1598064_qa_2", "description": "Which expenditure category shows the highest correlation with Total Household Income based on the correlation matrix analysis?", "authors": [], "keywords": [ "smoldataenvs...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0001/598/1598064.ipynb_qa_2", "kaggle_dataset_name": "grosvenpaul/family-income-and-expenditure", "gold_answer": "Communication Expenditure", "reward_mode_initial": "exact_short", "package_tier": 1, "difficulty_level": 3, "diff...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0001_598_1598064_qa_2
smoldataenvs-eval/0001_613_1613977_qa_3
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - tmdb_5000_movies.csv - tmdb_5000_credits.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, p...
Which movie genre generates the highest profit (calculated as mean revenue minus mean budget) in the dataset?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0001_613_1613977_qa_3", "description": "Which movie genre generates the highest profit (calculated as mean revenue minus mean budget) in the dataset?", "authors": [], "keywords": [ "smoldataenvs", "data-an...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0001/613/1613977.ipynb_qa_3", "kaggle_dataset_name": "tmdb/tmdb-movie-metadata", "gold_answer": "Animation", "reward_mode_initial": "exact_short", "package_tier": 1, "difficulty_level": 3, "difficulty": "medium", "difficulty_...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0001_613_1613977_qa_3
smoldataenvs-eval/0001_649_1649151_qa_1
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - deliveries.csv - matches.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip insta...
Which batsman has the highest average runs per game among players who have participated in more than 10 matches?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0001_649_1649151_qa_1", "description": "Which batsman has the highest average runs per game among players who have participated in more than 10 matches?", "authors": [], "keywords": [ "smoldataenvs", "data...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0001/649/1649151.ipynb_qa_1", "kaggle_dataset_name": "manasgarg/ipl", "gold_answer": "LMP Simmons", "reward_mode_initial": "exact_short", "package_tier": 1, "difficulty_level": 3, "difficulty": "medium", "difficulty_tier": "m...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0001_649_1649151_qa_1
smoldataenvs-eval/0001_666_1666101_qa_2
You are a data-analysis agent working in a sandbox. Use your code-execution tool to inspect the files and compute the answer. Files (in /home/user/input, no subfolders): - cereal.csv Installed: pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, statsmodels, tabulate, sqlite3, plotly (pip install more if needed)...
What is the highest average cereal rating among all manufacturers?
[ "smoldataenvs", "data-analysis", "kaggle" ]
1.2
{ "schema_version": "1.2", "artifacts": [], "task": { "name": "smoldataenvs-eval/0001_666_1666101_qa_2", "description": "What is the highest average cereal rating among all manufacturers?", "authors": [], "keywords": [ "smoldataenvs", "data-analysis", "kaggle" ] }, "metad...
{ "source_dataset": "jupyter-agent/jupyter-agent-dataset", "source_row_id": "0001/666/1666101.ipynb_qa_2", "kaggle_dataset_name": "crawford/80-cereals", "gold_answer": "67.97", "reward_mode_initial": "numeric", "package_tier": 0, "difficulty_level": 3, "difficulty": "medium", "difficulty_tier": "mediu...
[ "environment/Dockerfile", "environment/pull_bucket.py", "instruction.md", "task.toml", "tests/grader.py", "tests/test.sh" ]
tasks/0001_666_1666101_qa_2
End of preview. Expand in Data Studio
SmolDataEnvs

📊 SmolDataEnvs: Harbor (eval)

Collection Harbor Visualiser

5.5K+ RL tasks for hill-climbing small models in code and data science.

Reward and held-out pass@k climbing over 1,119 GRPO steps

A 2B model on these tasks. Left: what it optimises. Right: 144 held-out tasks it never trains on.
Two runs over the same 5,000 tasks: shuffled against a curriculum ordered easiest to hardest.

The validation suite: 144 tasks, small enough to run every few hundred training steps without the eval becoming the expensive part of the loop.

Packaged in Harbor format, so each task is a ready-made agentic environment: its own container, its own data, its own verifier.

What's inside

  • 144 verified tasks: quick validation during a run
  • 61 reviewed tasks in eval_verified.txt
  • Difficulty: easy 16, medium 74, hard 54 (difficulty_tier, plus difficulty_level 1–5)

Verified subset

eval_verified.txt lists 61 of the 144 tasks that passed a review (gold answer reproduced by an independent solution, one reading of the question, no unstated model setup, no file copied from a train task). registry_eval_verified.json is the same list as a Harbor registry. The task folders are the same as in the full suite. The criteria are on the SmolDataEnvs card.

To fetch only those tasks:

from huggingface_hub import hf_hub_download, snapshot_download

repo = "FineEnvs/SmolDataEnvs-harbor-eval"
ids = open(hf_hub_download(repo, "eval_verified.txt", repo_type="dataset")).read().split()
snapshot_download(repo, repo_type="dataset", local_dir="smoldataenvs-eval-verified",
                  allow_patterns=[f"tasks/{i}/*" for i in ids])

and pass the local folder as --dataset.

Versions

Revision What changed
v1.0 The original release.
v1.1 Removes HF_TOKEN from every task.toml, pins docker_image to the image the train suite uses (so HF Sandboxes can start the tasks), fixes 5 wrong gold answers (listed in CHANGELOG.md) and adds the eval_verified list.

Don't run tasks from revisions before v1.1 with a token in your environment. Their task.toml passes HF_TOKEN into the sandbox, where the agent can read it and print it into traces and logs. The bucket with the data files is public, so no token is needed.

Known issues (planned for v2)

The splits were drawn task by task, not by source dataset, so most eval and test tasks use a Kaggle dataset that some train task also uses (122 of 144 eval, 185 of 250 test). Some questions, mostly about a model's accuracy or feature importance, depend on an unstated split, seed or model. Details and the reviewed subsets are on the SmolDataEnvs card.

How a task is laid out

tasks/<task_id>/
  task.toml         # metadata, the question, the gold answer, grading tolerances
  instruction.md    # the prompt the agent sees
  environment/      # Dockerfile (shared base image) + the data-pull hook
  tests/            # grader.py (deterministic) + test.sh
registry.json       # the suite manifest
manifest.parquet    # one row per task, for filtering without walking the tree

Install

pip install "openenv[harbor]"   # tested with openenv 0.6.0, needs Python 3.12+

The [harbor] extra is what brings in the sandboxes. A plain pip install openenv gives you the CLI but nothing to run a task in. --sandbox e2b also needs E2B_API_KEY set.

Serve it

openenv harbor serve \
  --dataset FineEnvs/SmolDataEnvs-harbor-eval \
  --llm-url http://127.0.0.1:8000/v1 --model <your-model> \
  --port 8000 --capture-port 8100

Pass several with --dataset a,b and each arrives as its own split, which is how you train against -train and validate against -eval from one server.

One rollout, no trainer

openenv harbor rollout \
  --dataset FineEnvs/SmolDataEnvs-harbor-eval \
  --llm-url http://127.0.0.1:8000/v1 --model <your-model> \
  --harness opencode --sandbox e2b --task-index 0

Where it comes from

Built from the jupyter-agent dataset, real data-science notebooks over 471 Kaggle datasets. Every question–answer pair was extracted and then verified: strong agent models had to solve the task in a live sandbox and reproduce the gold answer under deterministic grading. Anything ambiguous or un-checkable was dropped. So every task here is known-solvable and unambiguously gradable.

Verified by a checker, not judged by a model. Grading is an exact comparison against a known answer, through a ladder of checks: exact match → numeric with tolerances → list and percent normalisation → symbolic equivalence. No LLM sits in the reward path, so the signal does not drift when you change the grader's model, because there isn't one.

The family

Repo What it is
SmolDataEnvs the tasks as plain rows, load it and prompt any model
SmolDataEnvs-sft 4,677 verified agent trajectories, TRL-ready
SmolDataEnvs-harbor-train 5,000 tasks as Harbor environments
SmolDataEnvs-harbor-test 250 held-out, deliberately harder
SmolDataEnvs-harbor-eval 144 for quick validation during a run

Citation

@misc{fineenvs,
  author = {Kolavi, Adithya S},
  title  = {FineEnvs: Open Source RL Environments for LLM Agents},
  year   = {2026},
  url    = {https://github.com/adithya-s-k/FineEnvs}
}
Downloads last month
7,334

Collection including FineEnvs/SmolDataEnvs-harbor-eval