mutation_ratio to 0.2 #6
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
dymitrlubczyk
added a commit
that referenced
this pull request
Sep 28, 2021
* 🔬 Experiment Class (#3) * Save progress on Experiment Class * Remove binary and text files * Refactor Experiment class * Save Process * Added Experiment to run_algorithm() * Remove pyc file * Turned on self.logs in environment.py because logs were still being displayed * Made changes to experiment, base_algo, and specialist; added line_plot_method, changed variable name in base_alg, and added hyperparams for specialist * Changed generations_number from 5 to 2 * Changed fitnesses var name to avg_generation_fitness for clarity; added DEBUG statements; requesting clarification for population array * Implemented working line_plot function - plots average generation fitness during each experimental run; TODO: boxplots, fix best_solution array - best does not want to be appended to a list of list... so far solutions get merged into a single list * Base_algo: fixed generation iteratable (increased at beginning & end, instead of just end); experiment.py: fixed calculation of total average - successfully plots total average over all generations in a single experiment against experiment number; specialist: changed some hyperparams Co-authored-by: marcgalitski <[email protected]> * Add niche fitness (#4) * 🙆🏻♀️ Crossover methods (#5) * Average crossover * Add hidden layer to network * Niche fitness refactor * ∞ Uniform Mutation (#6) * mutation_ratio to 0.2 * Added uniform_mutation, removed duplicate mutation_ratio (set by specialist as hyperparam) * Added uniform, made changes to attribute names * Changed selection ratio to 0.3 * 🏟 Tournament selection (#7) * 🦧 Add tuning (#8) * 🦫 Fix naming (#9) * Add tuning * Fix * 🐑 Same initial population in tuning (#10) * Add tuning * Fix * Same initial population * 🐏 Same initial population (#11) * Minor improvements * 🦒 Tuning improvements (#12) * Same initial population * Minor improvements * Fix alpha values * Fix * 🐉 Tuning improvements (#13) * Same initial population * Minor improvements * Fix alpha values * Fix * Minor improvement * 🪲 Tuning improvements (#14) * Same initial population * Minor improvements * Fix alpha values * Fix * Minor improvement * Tune params Co-authored-by: marcgalitski <[email protected]> Co-authored-by: marcgalitski <[email protected]>
dymitrlubczyk
added a commit
that referenced
this pull request
Sep 29, 2021
* 🔬 Experiment Class (#3) * Save progress on Experiment Class * Remove binary and text files * Refactor Experiment class * Save Process * Added Experiment to run_algorithm() * Remove pyc file * Turned on self.logs in environment.py because logs were still being displayed * Made changes to experiment, base_algo, and specialist; added line_plot_method, changed variable name in base_alg, and added hyperparams for specialist * Changed generations_number from 5 to 2 * Changed fitnesses var name to avg_generation_fitness for clarity; added DEBUG statements; requesting clarification for population array * Implemented working line_plot function - plots average generation fitness during each experimental run; TODO: boxplots, fix best_solution array - best does not want to be appended to a list of list... so far solutions get merged into a single list * Base_algo: fixed generation iteratable (increased at beginning & end, instead of just end); experiment.py: fixed calculation of total average - successfully plots total average over all generations in a single experiment against experiment number; specialist: changed some hyperparams Co-authored-by: marcgalitski <[email protected]> * Added uniform_mutation, made changes to mutation_selection params & print message * Made adjustments to uniform_mutation * Add niche fitness (#4) * Switched order of generation increment & assigning avg_generation_fitness * Fixed typo * 🙆🏻♀️ Crossover methods (#5) * Average crossover * Add hidden layer to network * Niche fitness refactor * Just making 2D array to start my idea * ∞ Uniform Mutation (#6) * mutation_ratio to 0.2 * Added uniform_mutation, removed duplicate mutation_ratio (set by specialist as hyperparam) * Added uniform, made changes to attribute names * Changed selection ratio to 0.3 * 🏟 Tournament selection (#7) * 🦧 Add tuning (#8) * 🦫 Fix naming (#9) * Add tuning * Fix * 🐑 Same initial population in tuning (#10) * Add tuning * Fix * 🐏 Same initial population (#11) * 🦒 Tuning improvements (#12) * Same initial population * Minor improvements * 🐉 Tuning improvements (#13) * Same initial population * Minor improvements * Fix alpha values * Fix * still wip sorry guys Seperated plotters and experiments Setting up experiment to run as they intend. * Lineplots - Updates for spec Note - I have screwed around in base_evolutionary_algo Primarily work on fixing line-plot to desired. * Added some todo's - lineplots done (nvm a bug appeared) and boxplots WIP * Lineplot works with standard deviation - Boxplot WIP * Saving of the best individual happens. Setting up that the best individuals from each run get to play 5 times and then their data is averaged. * Boxplot and running best individual WIP * 🪲 Tuning improvements (#14) * Same initial population * Minor improvements * Fix alpha values * Fix * Minor improvement * Boxplot + Best individual run = nice * Boxplots + play best individual almost ready * Play best is smarter... * Tune hyperparams (#15) * Same initial population * Minor improvements * Fix alpha values * Fix * Minor improvement * Tune params * Debugging in plots and experiment * Final debugs of plots * Fixed merge with base_evo_alg * split experiment and main * Integrate Experiment with EvolutionaryAlgorithm Co-authored-by: marcgalitski <[email protected]> Co-authored-by: Adrian S.A <[email protected]> Co-authored-by: marcgalitski <[email protected]>
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.