-
Notifications
You must be signed in to change notification settings - Fork 5
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Data visualization based on evaluation CSV files #296
Comments
Closed
7 tasks
ruiAzevedo19
added a commit
that referenced
this issue
Jul 30, 2024
…nto a helper, so to it can be reused Part of #296
ruiAzevedo19
added a commit
that referenced
this issue
Jul 30, 2024
… a CSV file, so it can be used for data visualization Part of #296
ruiAzevedo19
added a commit
that referenced
this issue
Jul 30, 2024
…nto a helper, so to it can be reused Part of #296
ruiAzevedo19
added a commit
that referenced
this issue
Jul 30, 2024
… a CSV file, so it can be used for data visualization Part of #296
ruiAzevedo19
added a commit
that referenced
this issue
Jul 30, 2024
ruiAzevedo19
added a commit
that referenced
this issue
Jul 31, 2024
…e JSON response, to avoid these values to be converted latter on Part of #296
ruiAzevedo19
added a commit
that referenced
this issue
Jul 31, 2024
…der to the model package, since it is model related Part of #296
ruiAzevedo19
added a commit
that referenced
this issue
Jul 31, 2024
…nto a helper, so to it can be reused Part of #296
ruiAzevedo19
added a commit
that referenced
this issue
Jul 31, 2024
…e it can error if the file already exists Part of #296
ruiAzevedo19
added a commit
that referenced
this issue
Jul 31, 2024
…d in data visualization Part of #296
ruiAzevedo19
added a commit
that referenced
this issue
Jul 31, 2024
…d in data visualization Part of #296
ruiAzevedo19
added a commit
that referenced
this issue
Jul 31, 2024
Leaving this here until we have the summing logic in the visualization. # script.sh <evaluation-without-extension> <meta-without-extension>
pip install csvkit
sed -i '1s/-/_/g' $1.csv # SQL does not like hyphens in column names.
sed -i '1s/-/_/g' $2.csv # SQL does not like hyphens in column names.
csvsql --query "SELECT model_id, language, SUM(score) AS score, SUM(coverage) AS coverage, SUM(files_executed) AS files_executed, SUM(files_executed_maximum_reachable) AS files_executed_maximum_reachable, SUM(generate_tests_for_file_character_count) AS generate_tests_for_file_character_count, SUM(processing_time) AS processing_time, SUM(response_character_count) AS response_character_count, SUM(response_no_error) AS response_no_error, SUM(response_no_excess) AS response_no_excess, SUM(response_with_code) AS response_with_code, SUM(tests_passing) AS tests_passing FROM $1 WHERE task NOT LIKE '%-symflower-fix' GROUP BY model_id, language" $1.csv > $1-by-language.csv
csvsql --query "SELECT model_id, SUM(score) AS score, SUM(CASE WHEN language = 'golang' THEN score ELSE 0 END) AS golang_score, SUM(CASE WHEN language = 'java' THEN score ELSE 0 END) AS java_score, SUM(CASE WHEN language = 'ruby' THEN score ELSE 0 END) AS ruby_score FROM $1 WHERE task NOT LIKE '%-symflower-fix' GROUP BY model_id" $1.csv > $1-by-language-score.csv
csvsql --query "SELECT $1.model_id, model_name, (completion + prompt + request) AS cost, SUM(score) AS score, SUM(coverage) AS coverage, SUM(files_executed) AS files_executed, SUM(files_executed_maximum_reachable) AS files_executed_maximum_reachable, SUM(generate_tests_for_file_character_count) AS generate_tests_for_file_character_count, SUM(processing_time) AS processing_time, SUM(response_character_count) AS response_character_count, SUM(response_no_error) AS response_no_error, SUM(response_no_excess) AS response_no_excess, SUM(response_with_code) AS response_with_code, SUM(tests_passing) AS tests_passing FROM $1 LEFT JOIN $2 ON $1.model_id = $2.model_id WHERE task NOT LIKE '%-symflower-fix' GROUP BY $1.model_id" $1.csv $2.csv > $1-total.csv
csvsql --query "SELECT model_id, task, SUM(score) AS score, SUM(coverage) AS coverage, SUM(files_executed) AS files_executed, SUM(files_executed_maximum_reachable) AS files_executed_maximum_reachable, SUM(generate_tests_for_file_character_count) AS generate_tests_for_file_character_count, SUM(processing_time) AS processing_time, SUM(response_character_count) AS response_character_count, SUM(response_no_error) AS response_no_error, SUM(response_no_excess) AS response_no_excess, SUM(response_with_code) AS response_with_code, SUM(tests_passing) AS tests_passing FROM $1 WHERE task NOT LIKE '%-symflower-fix' GROUP BY model_id, task" $1.csv > $1-by-task.csv
csvsql --query "SELECT model_id, task, language, SUM(score) AS score, SUM(coverage) AS coverage, SUM(files_executed) AS files_executed, SUM(files_executed_maximum_reachable) AS files_executed_maximum_reachable, SUM(generate_tests_for_file_character_count) AS generate_tests_for_file_character_count, SUM(processing_time) AS processing_time, SUM(response_character_count) AS response_character_count, SUM(response_no_error) AS response_no_error, SUM(response_no_excess) AS response_no_excess, SUM(response_with_code) AS response_with_code, SUM(tests_passing) AS tests_passing FROM $1 WHERE task NOT LIKE '%-symflower-fix' GROUP BY model_id, task, language" $1.csv > $1-by-task-by-language.csv
csvsql --query "SELECT model_id, SUM(CASE WHEN task NOT LIKE '%-symflower-fix' THEN score ELSE 0 END) AS score, SUM(CASE WHEN task LIKE '%-symflower-fix' THEN score ELSE 0 END) AS score_fix, SUM(CASE WHEN task NOT LIKE '%-symflower-fix' THEN files_executed ELSE 0 END) AS files_executed, SUM(CASE WHEN task LIKE '%-symflower-fix' THEN files_executed ELSE 0 END) AS files_executed_fix FROM $1 WHERE (task LIKE 'transpile%' OR task LIKE 'write-tests%') AND language = 'golang' GROUP BY model_id " $1.csv > $1-by-symflower-fix.csv |
ruiAzevedo19
added a commit
that referenced
this issue
Aug 1, 2024
…e JSON response, to avoid these values to be converted latter on Part of #296
ruiAzevedo19
added a commit
that referenced
this issue
Aug 1, 2024
…der to the model package, since it is model related Part of #296
ruiAzevedo19
added a commit
that referenced
this issue
Aug 1, 2024
…nto a helper, so to it can be reused Part of #296
ruiAzevedo19
added a commit
that referenced
this issue
Aug 1, 2024
…e it can error if the file already exists Part of #296
ruiAzevedo19
added a commit
that referenced
this issue
Aug 1, 2024
…d in data visualization Part of #296
ruiAzevedo19
added a commit
that referenced
this issue
Aug 1, 2024
…d in data visualization Part of #296
ruiAzevedo19
added a commit
that referenced
this issue
Aug 1, 2024
…d in data visualization Part of #296
ruiAzevedo19
added a commit
that referenced
this issue
Aug 2, 2024
… generic name, since it can sort all kind of CSV records Part of #296
ruiAzevedo19
added a commit
that referenced
this issue
Aug 2, 2024
…d in data visualization Part of #296
ruiAzevedo19
added a commit
that referenced
this issue
Aug 2, 2024
… generic name, since it can sort all kind of CSV records Part of #296
ruiAzevedo19
added a commit
that referenced
this issue
Aug 2, 2024
…d in data visualization Part of #296
ruiAzevedo19
added a commit
that referenced
this issue
Aug 6, 2024
…e JSON response, to avoid these values to be converted latter on Part of #296
ruiAzevedo19
added a commit
that referenced
this issue
Aug 6, 2024
…der to the model package, since it is model related Part of #296
ruiAzevedo19
added a commit
that referenced
this issue
Aug 6, 2024
…nto a helper, so to it can be reused Part of #296
ruiAzevedo19
added a commit
that referenced
this issue
Aug 6, 2024
…e it can error if the file already exists Part of #296
ruiAzevedo19
added a commit
that referenced
this issue
Aug 6, 2024
… generic name, since it can sort all kind of CSV records Part of #296
ruiAzevedo19
added a commit
that referenced
this issue
Aug 6, 2024
…d in data visualization Part of #296
Munsio
pushed a commit
that referenced
this issue
Aug 28, 2024
…e JSON response, to avoid these values to be converted latter on Part of #296
Munsio
pushed a commit
that referenced
this issue
Aug 28, 2024
…der to the model package, since it is model related Part of #296
Munsio
pushed a commit
that referenced
this issue
Aug 28, 2024
…nto a helper, so to it can be reused Part of #296
Munsio
pushed a commit
that referenced
this issue
Aug 28, 2024
…e it can error if the file already exists Part of #296
Munsio
pushed a commit
that referenced
this issue
Aug 28, 2024
… generic name, since it can sort all kind of CSV records Part of #296
Munsio
pushed a commit
that referenced
this issue
Aug 28, 2024
…d in data visualization Part of #296
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Goal: create a HTML report with graphs for data visualization.
Tool: D3.js library for data visualization graphs
TODO
Create a table for the evaluation CSV file
Scatter plot
MetaInformation
function that returns a model's meta informationerror bars over multiple runs, to show variance
The text was updated successfully, but these errors were encountered: