Flatten json #94
Merged
Flatten json #94
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
manycoding
requested review from
raphapassini,
ejulio,
victor-torres and
ivankivanov
manycoding
commented
May 23, 2019
| if self._flat_df is None: | ||
| if self.expand: | ||
| self._flat_df, self._columns_map = pandas.flatten_df(self.df) | ||
| self._flat_df = pd.DataFrame(flatten(i) for i in self.raw) | ||
| self._flat_df["_key"] = self.df.get( |
There was a problem hiding this comment.
We need the same formatted keys as in original df.
manycoding
commented
May 23, 2019
| @@ -40,8 +43,12 @@ def process_df(df: pd.DataFrame) -> pd.DataFrame: | |||
| df["_type"] = df["_type"].astype("category") | |||
| return df | |||
|
|
|||
| def get_origin_column_name(self, column_name: str) -> str: | |||
| return self._columns_map.get(column_name, column_name) | |||
| def origin_column_name(self, new: str) -> str: | |||
There was a problem hiding this comment.
KISS approach. I believe it will work in most cases.
manycoding
commented
May 23, 2019
| {"type": [0, [2, 3]], "str": "s", "type_0": 6}, | ||
| ], | ||
| { | ||
| "str": ["k", "s"], |
There was a problem hiding this comment.
Here we lose type 0,0
Codecov Report
@@ Coverage Diff @@
## master #94 +/- ##
==========================================
- Coverage 71.04% 70.47% -0.57%
==========================================
Files 24 23 -1
Lines 1592 1565 -27
Branches 275 273 -2
==========================================
- Hits 1131 1103 -28
Misses 436 436
- Partials 25 26 +1
Continue to review full report at Codecov.
|
ivankivanov
approved these changes
May 23, 2019
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Outdated suggestions cannot be applied.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.
This one significantly increases performance and decreases memory usage if using
flat_df, which corresponds toexpand=Truedefault parameter.Some things to consider:
See here Data is lost if unflatten column name equals to another original column amirziai/flatten#48
I think it won't happen in 99% of cases, I've never encountered such a case. Alternatively, we can set other separator which potentially bumps it to 99.9%.
Nevertheless, there's is that issue and also a comment in tests.
Time and peak memory, 1k samples from a heavily nested df (see
shared/qa/Experiments/data/items_epicsports_150k_nested.pickle).This pr - 1.63s, 2400 MB
Old - 4min 57s, 4411.51 MiB
#74 - 6min 2s, 2800 MB
And, interestingly, all 150k items with this pr:
9min 40s
peak memory: 3682.77 MiB, increment: 1391.34 MiB