-
Notifications
You must be signed in to change notification settings - Fork 16
add convert_rank_to_percentile to RankingCalculator #20
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
Merged
Merged
Changes from all commits
Commits
Show all changes
3 commits
Select commit
Hold shift + click to select a range
File filter
Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Some comments aren't visible on the classic Files Changed page.
There are no files selected for viewing
This file contains hidden or 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
This file contains hidden or 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
| Original file line number | Diff line number | Diff line change | ||||
|---|---|---|---|---|---|---|
|
|
@@ -80,7 +80,8 @@ def model_post_init(self, __context: Any) -> None: | |||||
| rank, performance = self.raw_data[i] | ||||||
| next_rank, next_performance = self.raw_data[i + 1] | ||||||
| if rank == next_rank: | ||||||
| assert performance == next_performance, "Something went wrong: `performance` != `next_performance`." | ||||||
| if performance != next_performance: | ||||||
| raise RuntimeError("Something went wrong: `performance` != `next_performance`.") | ||||||
| continue # NOTE: tie | ||||||
| actual_rank = float((rank + next_rank - 1) / 2) # NOTE: Use the average of the rank range for performance | ||||||
| self.data[actual_rank] = performance | ||||||
|
|
@@ -116,7 +117,7 @@ def get_performance(self, rank: int | float) -> int: | |||||
| else: | ||||||
| lose = mid | ||||||
| if win == len(sorted_keys): | ||||||
| assert False, "Something went wrong: `win` should be less than `len(sorted_keys)`." | ||||||
| raise RuntimeError("Something went wrong: `win` should be less than `len(sorted_keys)`.") | ||||||
| if win == 0: | ||||||
| win = 1 # NOTE: to avoid index out of range and use the 1st & 2nd entries | ||||||
| rank_high = sorted_keys[win - 1] | ||||||
|
|
@@ -301,7 +302,8 @@ def model_post_init(self, __context: Any) -> None: | |||||
| for i in range(len(self.score_rank_list) - 1): | ||||||
| _, _, end_rank = self.score_rank_list[i] | ||||||
| _, next_start_rank, _ = self.score_rank_list[i + 1] | ||||||
| assert next_start_rank == end_rank + 1, "Something went wrong: `next_start_rank` != `end_rank + 1`." | ||||||
| if next_start_rank != end_rank + 1: | ||||||
| raise RuntimeError("Something went wrong: `next_start_rank` != `end_rank + 1`.") | ||||||
|
|
||||||
| def get_new_rank(self, private_result: Result) -> tuple[int, float, list[int]]: | ||||||
| """Get the new rank for a given private result. | ||||||
|
|
@@ -364,7 +366,7 @@ def get_new_rank(self, private_result: Result) -> tuple[int, float, list[int]]: | |||||
| return start_rank, float(start_rank), new_case_scores | ||||||
| elif new_score == score: | ||||||
| return start_rank, float((start_rank + end_rank) / 2), new_case_scores # NOTE: Average of the ranks | ||||||
| assert False, "Something went wrong in `get_new_rank` method." | ||||||
| raise RuntimeError("Something went wrong in `get_new_rank` method.") | ||||||
|
|
||||||
|
|
||||||
| class ProblemMetaData(BaseModel): | ||||||
|
|
@@ -761,6 +763,14 @@ def load_contest_schedule() -> dict[str, tuple[dt.datetime, dt.datetime, float]] | |||||
| } | ||||||
|
|
||||||
|
|
||||||
| class RankPercentileMapMethod(str, Enum): | ||||||
| """Method for converting rank to percentile.""" | ||||||
|
|
||||||
| ORIGINAL = "original" | ||||||
| HAZEN = "hazen" | ||||||
| WEIBULL = "weibull" | ||||||
|
|
||||||
|
|
||||||
| class RankingCalculator: | ||||||
| """Ranking calculator for ALE-Bench.""" | ||||||
|
|
||||||
|
|
@@ -781,7 +791,7 @@ def __init__(self, minimum_participation: int = 5) -> None: | |||||
| ranking_file_path = str(local_data_dir / "ranking.csv") | ||||||
| df_ranking = pl.read_csv(ranking_file_path).filter(pl.col("competitions") >= minimum_participation) | ||||||
| # Prepare member variables | ||||||
| num_active_users = len(df_ranking) | ||||||
| self.num_active_users = len(df_ranking) | ||||||
| self.avg_perfs = df_ranking["avg_perf"].sort(descending=True).to_list() | ||||||
| self.ratings = df_ranking["rating"].sort(descending=True).to_list() | ||||||
| self.avg_perf_ranks, self.rating_ranks = [], [] | ||||||
|
|
@@ -798,9 +808,9 @@ def __init__(self, minimum_participation: int = 5) -> None: | |||||
| self.rating_ranks.append(current_rating_rank) | ||||||
| # Append the last entry for the average performance and rating | ||||||
| self.avg_perfs.append(-1000.0) | ||||||
| self.avg_perf_ranks.append(num_active_users + 1) | ||||||
| self.avg_perf_ranks.append(self.num_active_users + 1) | ||||||
| self.ratings.append(0) | ||||||
| self.rating_ranks.append(num_active_users + 1) | ||||||
| self.rating_ranks.append(self.num_active_users + 1) | ||||||
|
|
||||||
| def calculate_avg_perf_rank(self, avg_perf: float) -> int: | ||||||
| """Calculate the rank based on the rating. | ||||||
|
|
@@ -826,7 +836,7 @@ def calculate_avg_perf_rank(self, avg_perf: float) -> int: | |||||
| else: # Exactly matched | ||||||
| return self.avg_perf_ranks[mid] | ||||||
| if ok == len(self.avg_perfs): | ||||||
| assert False, "Something went wrong: `ok` should be less than `len(self.avg_perfs)`." | ||||||
| raise RuntimeError("Something went wrong: `ok` should be less than `len(self.avg_perfs)`.") | ||||||
| return self.avg_perf_ranks[ok] | ||||||
|
|
||||||
| def calculate_rating_rank(self, rating: int) -> int: | ||||||
|
|
@@ -853,5 +863,41 @@ def calculate_rating_rank(self, rating: int) -> int: | |||||
| else: # Exactly matched | ||||||
| return self.rating_ranks[mid] | ||||||
| if ok == len(self.ratings): | ||||||
| assert False, "Something went wrong: `ok` should be less than `len(self.ratings)`." | ||||||
| raise RuntimeError("Something went wrong: `ok` should be less than `len(self.ratings)`.") | ||||||
| return self.rating_ranks[ok] | ||||||
|
|
||||||
| def convert_rank_to_percentile( | ||||||
| self, | ||||||
| rank: int, | ||||||
| method: RankPercentileMapMethod | str = RankPercentileMapMethod.WEIBULL, | ||||||
| ) -> float: | ||||||
| """Convert the rank to percentile. | ||||||
|
|
||||||
| Args: | ||||||
| rank (int): The rank to convert. | ||||||
| method (RankPercentileMapMethod | str): The mode to use for conversion. Defaults to "weibull". | ||||||
| "original": percentile = 100.0 * rank / num_active_users | ||||||
| "hazen": percentile = 100.0 * (rank - 0.5) / (num_active_users + 1) | ||||||
| "weibull": percentile = 100.0 * rank / (num_active_users + 2) | ||||||
|
||||||
|
|
||||||
| Returns: | ||||||
| float: The converted percentile. | ||||||
|
|
||||||
| Raises: | ||||||
| ValueError: If the rank is less than 1 or greater than the number of active users + 1. | ||||||
| ValueError: If the method is invalid. | ||||||
| """ | ||||||
| if rank < 1 or rank > self.num_active_users + 1: | ||||||
| raise ValueError(f"The rank must be between 1 and {self.num_active_users + 1} (the number of users + 1).") | ||||||
| try: | ||||||
| method = RankPercentileMapMethod(method) | ||||||
| except ValueError: | ||||||
| raise ValueError(f"Invalid method: {method}. Supported methods are 'original', 'hazen', and 'weibull'.") | ||||||
| if method == RankPercentileMapMethod.ORIGINAL: | ||||||
| if rank == self.num_active_users + 1: | ||||||
| return 100.0 # NOTE: The lowest rank is always 100.0% (avoid exceeding 100.0%) | ||||||
| return 100.0 * rank / self.num_active_users | ||||||
| elif method == RankPercentileMapMethod.HAZEN: | ||||||
| return 100.0 * (rank - 0.5) / (self.num_active_users + 1) | ||||||
| elif method == RankPercentileMapMethod.WEIBULL: | ||||||
| return 100.0 * rank / (self.num_active_users + 2) | ||||||
|
||||||
| return 100.0 * rank / (self.num_active_users + 2) | |
| return 100.0 * rank / (self.num_active_users + 1) |
This file contains hidden or 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
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.
Uh oh!
There was an error while loading. Please reload this page.