This repository has been archived by the owner on Aug 26, 2024. It is now read-only.
Implemented sort order matches by common letter count largest to smallest #295
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
maxbachmann
reviewed
Dec 17, 2020
| sorted(sl, key=lambda i: i[1], reverse=True) | ||
| else: | ||
| best_list = sorted(sl, key=lambda i: i[1], reverse=True) | ||
| return sortByCommonLetter(sl, query)[0: min(limit, len(sl))] if limit is not None else \ |
There was a problem hiding this comment.
from a performance standpoint this is a bad idea. When the user is only interested in 5 elements (the default), but has e.g. 1 million choices, this will sort 1 million choices (with a slow algorithm, since it counts all the letters) and then only takes the best 5 elements.
Comment on lines
+242
to
+264
| def sortByCommonLetter(sl, query): | ||
| """This function further sorts the strings with the same scores by common letter count to the query.""" | ||
| current_score, last_index = -1, -1 | ||
| # Iterate over list and look for words with the same scores | ||
| for i in range(0, len(sl)): | ||
| # Identify the indexes of the strings with the same scores | ||
| if sl[i][1] != current_score or i == len(sl) - 1: | ||
| current_score = sl[i][1] | ||
| # First iteration, there are no previous words so we do not have to do anything | ||
| if last_index == -1: | ||
| last_index = i | ||
| continue | ||
| # Found a group of words with the same scores! Now sort them | ||
| if i - last_index > 1: | ||
| count_list = [] | ||
| for j in range(last_index, i): | ||
| count_list.append((sl[j][0], calculateCommonLetter(query, sl[j][0]))) | ||
| count_list = sorted(count_list, key=lambda k: k[1], reverse=True) | ||
| # Copy the sorted portion | ||
| for j in range(0, len(count_list)): | ||
| sl[last_index + j] = (count_list[j][0], current_score) | ||
| last_index = i | ||
| return sl |
There was a problem hiding this comment.
This algorithm is broken for quite a few cases. A couple quick examples:
- When a processor is used the results might not be correct:
query_1 = 'COMPANY'
choices_1 = ['Company 1', 'company', 'company', 'Company 2', 'Awesome Company']
process.extractOne(query_1, choices_1, scorer=fuzz.partial_token_set_ratio, char_sort=True)
-> ('Company 1', 100)
- it matters wether a element is query or choice
query_1 = 'Company'
choices_1 = ['Company 1', 'Company', 'Company 2', 'Awesome Company']
process.extractOne(query_1, choices_1, scorer=fuzz.partial_token_set_ratio, char_sort=True)
-> ('Company 1', 100)
|
Overall I am personally not really convinced, this should be added at all for two reasons
|
Added penalty for character mismatches in sortByCommonLetter Added call to full_process in sortByCommonLetter Fixed incorrect variable names (weird!)
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
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.
This pull request addresses the problem in this issue: #280
All the code changes + unit tests are in process.py and test_fuzzywuzzy.py.
All the old and new test cases in test_fuzzywuzzy.py are passed.