Export experiment #31
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
nathanmartins
approved these changes
Jan 6, 2020
thalessr
added a commit
that referenced
this pull request
Jan 6, 2020
* Aurum init (#1) * Change script to create src and logs directories * Add tests to guarantee the default directories creation * Fixing runtime * Adding git and a couple of base packages. * Adding `run_add` function and type annotations. * Adding complete add functionality. * Removing ENTRYPOINT. * Adding file extension. * adding docker ignore. * more logging * Adding files to git. * Adding makefile. * minor tweaks to existing test. * adding add test. * better error handling on git failures. * Setup (#3) * Making au available in the terminal * Getting arguments from command line * Fix au data commands * Fix * `rm` removes from metadata as well. (#5) * formatting. * Adding metadata file removal. * Initial MetaData class implementation. * Better docstring. * Adding default values. * Refactoring to use new Meta Data class * docs * Renaming metadata to avoid name conflicts. * Refactoring to create a more generalized MetaData class. * renaming file variable. * Moving to constants * Refactoring to use relative path only. * Separating the metadata files into named directories that represent their meaning. Eg. .au/datasets * Create one directory for the file_path. Eg. .au/datasets/dataset.txt/{mdfs_inside}. * Rename deserialize to serialize. * Generalized metadata deserialization. * Moving a few functions to utils. * Fixing old tests. * Removing unneeded type casting. * from aurum import constants as cons * fixing constant * Relative paths only. * Removing breakpoint * Updating makefile * Fixing `DATASET_METADATA_DIR` constant to be valid and enabling adding nested files. * Adding more tests. * Better docstring * prevent the au command from being run from inside any of the sub-directories of the .au dir. * Moving verification to where command are parsed. * Au script improvement (#9) * minor fmt * Minor bug fix. * - Adding type annotations - fixing `config_parser` output - changing to double quotes * Changing all `.formats` to f-strings * changing single quotes to double quotes * adding au command to makefile * Better au response and fixing tests. * Refactoring to a separate commands file * Fixing problems. * Removing dead code. * Removing unneeded and unused parameter * installed packages hashing algorithm implementation * Better comment * Set parameters (#8) * Add .editorconfig * Adding command line arguments taking precedence * Clean up * Saving parameters to metadata file * Add a basic test * Add test case when argument is not accepted * Change to concatenate the params * Saving parameters * Loading parameters from metadata * Saving parameters metadata for each experiment * Saving parameters to metadata * Change to add parent_file_name to newer versions * Full implementation. * Add metrics (#10) * Basic metrics * Basic metrics * Add gpu info * Clean up * Saving metrics to metadata * Changes required during code review * Remove non static info * Remove duplicated import * Init experiment (#13) * Add InitExperiment class * Renamed class from InitExperiment to VersionTheorem * Renamed class from VersionTheorem to Theorem * Checking any change * Log Tracker (#12) * basic implementation working with `prints` * Adding coverage to take on logging module as well. * Adding logging_tracker to au imports * Merging singleton code. * import fix * General refactoring. * Fixing Dockerfile. * Refactoring Parser and Theorem into separate files. * Removing duplicated package info (__)'s * Passing tests. * Refactoring Parser name. * Adding readme, example and a template. (#14) * Add run time to metrics (#15) * Experiment Metadata (#16) * Major metadata refactoring. * renaming... * Adding experiment metadata * Adding parent behaviour. * Adding example to makefile. * Saner experiment.py imports. * Fixing experiment_parser's imports and missing dependencies. * Moving logger to __init__ and better imports. * Implementing dry run and Dehydrator to keep state. * Relative imports. * Relative imports. * Removing dry run from __init__ * Full implementation. * Slight refactoring. * Fixing return type * Implementing experiment and dataset get_latest_by_date * Adding `use_datasets` to module exports and full DatasetTracker implementation. * Bug fixes and tests * Tracking code change (#17) * Add code tracker * Changes required during code review * Add test when the code changes * New example (#21) * Adding new and more complete experiment for aurum tests. * Minor fix - removing old code * End experiment (#22) * WIP * Change to create a commit & tag * Add code changes hash * Fix duplication * Moved experiment id to Theorem and add warnings * Fixes * No tracking and general bug fixes. (#18) * Fixing broken tests. * Removing unnecessary logging. * Fixing experiment parser bug. * Full implementation. * Removing logging bug * Fixing add test. * Removing breakpoint * Adding is `is_new_requirements` check to experiments. * fixing relative import * Since the whole command line is showing up inside sys.argv[0], break it up by space and then check if the first is au, just because au is a rather common sequence of vowels that can cause confusion if the name of the experiment file or one of the parameters contain those two letters in sequence. * Adding Theorem check * Removing unused import. * Adding Theorem check * Removing code from another branch * Inform dataset change to theorem (#25) * Adding Theorem check * Removing unused import. * Adding Theorem check * Revert "Adding Theorem check" This reverts commit 012b6eb. * Cleaner version. * Fix metadata (#24) * Change parameters metadata * Change metrics metadata * Change code metadata * Change experiment metadata * Change dataset metadata * Changed metadata classes * Implementing theorem check and changing the is_new_code implementation to a hash based check. * Relative imports only. * this test was removed for some reason * Fixing import. * Better process handling. * imports and spacing * Adding initial boilerplate for load CLI command. * Experiment ID f2b998f1-4c55-49f9-9d97-8499c7a5a58c Requiments hash 9d35681a1e132ce04f6e7f27d61b407e96dfcf4b * Experiment ID 34c1ae7b-0489-4b25-8ed8-a9cba1e73137 Requiments hash 9d35681a1e132ce04f6e7f27d61b407e96dfcf4b * Experiment ID 34c1ae7b-0489-4b25-8ed8-a9cba1e73137 Requiments hash 9d35681a1e132ce04f6e7f27d61b407e96dfcf4b * Experiment ID 0964c8e1-b919-4b8e-bc0e-e620d5d00d23 Requiments hash 9d35681a1e132ce04f6e7f27d61b407e96dfcf4b * Experiment ID 0964c8e1-b919-4b8e-bc0e-e620d5d00d23 Requiments hash 9d35681a1e132ce04f6e7f27d61b407e96dfcf4b * Experiment ID 8e92b4ea-a0f5-485f-9faf-3c884ca94549 Requiments hash 9d35681a1e132ce04f6e7f27d61b407e96dfcf4b * Experiment ID 8e92b4ea-a0f5-485f-9faf-3c884ca94549 Requiments hash 9d35681a1e132ce04f6e7f27d61b407e96dfcf4b * Experiment ID 290234f7-4c3f-43ce-bbe3-be8f78949d42 Requiments hash 9d35681a1e132ce04f6e7f27d61b407e96dfcf4b * Experiment ID 7423e244-241a-4b85-9f19-a84278507836 Requiments hash 9d35681a1e132ce04f6e7f27d61b407e96dfcf4b * Updating gitignore * Removing au specific courses * Updating docker ignore. * Adding context to git tests. * Minor dataset tracker bug * base.cwd to base.get_cwd * general imports and bug fixes * implementig checkout * Creating temporary env builder * Add display metrics (#28) * Fix metrics and parameter metadata (#29) * Fix metrics and parameter metadata * Using relative import * pip install working * Adding activation cmd. * Re-adding removed test * minor fix. * fixing test * Adding constant. * Fixing imports. * Fixing super() bug * Export experiment (#31) * Add au export * Improve dataset export * Add a basic test * Add data from web (#33) Co-authored-by: Nathan S. Martins <nathanmartins@users.noreply.github.com> Co-authored-by: Adriano Marques <adrianomarques@users.noreply.github.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.