- Change to type instead of instantiated type (
e327665) - Misc (
46c6707) - Validate fallback is compatible with output of aggregator (
6fe230b)
- Use Union in pydantic basemodels (
a80a9c2)
- Don't mix types in HasValuesAggregator fallback (
dfdf5dd)
- Expose ValueSpecification (
b670fc8)
- Rename 'pred_time_uuid' to 'prediction_time_uuid' (
fe44fd8)
- Scrambled features with step size (
6bdddbd)
- Bump release (
879e752)
- Ensure entity col name is passed to the valueframe in legacy spec (
6702cb9)
- Add option to set entity_id_col_name_out in legacy predictor group spec (
4094dc1)
- Update pre-commit.yml highlight legacy spec change (
fcff53e)
- Specify outcome colname in boolean (
70cb724)
- Require polars >= 0.19.0 (
c70ca84)
- Docstrings for aggregators (
32b6698)
- Update docs (
196fb50)
- Update major version, deprecate v1 (
1dda823)
- update major version, deprecate v1 (
1dda823)
- Bux in timedelta (
92b90b3)
- #496: Add some docstrings to user facing specs on the expected type of input and general usage (#499) (#500) (
cf30e04) - #496: Add some docstrings to user facing specs on the expected type of input and general usage (#499) (
3cfa8d5) - #496: Add some docstrings to user facing specs on the expected type of input and general usage (
ae0f04a) - #485: Update readme to use v2 (
37dc2ad)
- Bug in
_get_timedelta_frameif timestamp col names were the same (3ade969) - #480: Pure NaNs when flattening with lookbehind-tuple in 01_basic (
6d1122b)
- Text tutorial (
b761131)
- Update tutorials to not use value_col_name (
872cd1f)
- #479: Improve value column conflict error (
4c33465)
- #474: Update 01_basic (
cc74a82)
- #363: Re-add support for Python >=3.9 (
7d9df0c)
- Allow setting format in timedeltaspec (
8383269)
- Remove unused args in flattener (
82e4cc3)
- #440: Type of AggregatedDataframe.df should always be LazyFrame (
ab9be64)
- #441: Return entity_id in aggregated column (
d7f9cfb)
- #439: Support static features (
e8d1977)
- Age / time from event spec (
1856cb1)
- #425: Support from_legacy api interface (
e0f3160)
- Initial rye support (
aaf06ae)
- #399: Handle boolean outcomes without value column automatically (
b4e09bc)
- #449: Propagate timeseries column to aggregation function (#451) (
8b4c7ed) - #449: Propagate timeseries column to aggregation function (
3bea4bb)
- #447: Implement earliest and latest without requiring sorting (#450) (
f88fccb) - #447: Implement earliest and latest without requiring sorting (
24acab1)
- Lookperiod tuple (
ff43abd)
- #396: Add all aggregators from v1 (
a1fa957)
- Validate columns in valueframe (
5019ad5)
- #402: Allow eager polarsframes as input (#433) (
ffe2f0a) - #402: Allow eager polarsframes as input (
f1e4cee)
- #423: Multiprocessing (
71de8c1)
- Expand aggregators (
866a3a1)
- Aggregate with one groupby (
d933d46)
- Further tweak benchmarks (
f29654d)
- #384: Add benchmark (
4d746de)
- Ensure all prediction times are kept when slicing (#397) (
6a5aa16) - Ensure all prediction times are kep when slicing (
9e5cef4)
- #356: Validate shared columns exist in all specs on ingest (
c0aa8cd)
- #379: Check that no value columns have the same name (
28518e0)
- #386: Make debugging easier (
94b3179)
- #376: Add prefix to colname (
d0e01b7)
- #360: Ensure correct ordering before concatenating (
3af55e3)
- #365: Support ingesting from pandas dataframes (
6614836)
- #359: Handle fallbacks (
e72a90b)
- Polars prototype (
36e6837)
- Bump version (
bc0b500)
- List to Sequence type hint in group specs (
133be8b)
- Allow feature generation in time ranges (
924e63a)
- Make generated colnames backwards compatible (
b767c29)
- Add new functionality to tutorials (
bc809aa)
- Upgrade pandas deps (
99568ad)
- Add meaningful error if entity id is not unique in incident outcome (
9abaf75)
- Call function properly (
df13140)
- Remove unneeded tests (
8babaf7) - Remove column_handler (
f3ce8b7) - Remove text utility functions (
dcc3f9b)
- Do not unpack col names (
6c4c236)
- Remove sklearn/sentence transformer from text tutorial (
6bc467a)
- Hopeful ci docs fix (
cd78f98)
- Make docs render properly (
990020d)
- Convert df with multiple values to named dataframes (
ca797e6)
- Add how to use pre-embedded text to text tutorial (
c0f994c)
- Change int detection for new version of pydantic (
752d3bf)
- Remove autodoc pydantic requirements for docs (
34ae391) - Handle non-coerceable values (
557371f) - Toml urls (
385706d)
- Create combinations not working for (
8c3c343)
- Add types to aggregation functions (
560539f)
- Specs should be pydantic basemodels (
04781f6)
- Incorrect type coercion in pydantic (
e37eefb) - Types (
3ed4b59) - Remove unnecessary feature_name from group specs (
f29d017)
- Add feature_name arg to group specs and make non-optional for non group specs (
34df7a3)
- Add group spec for text predictor spec (
75a0112)
- Bump version to generate new release (
2316b38)
- Remove undone test (
07ecfca)
- Support multiple python versions for testing (
d929147) - Correct interval str computation (
0cd0437) - Change col str naming (
f853452) - Fix interval days type hints in specs (
e26235c) - Rename test deps (
471a63b) - Type hints (
21726f2)
- Align docstrings (
8d92a25) - Fix tutorial (
f856a56) - Update docstrings (
b771f5b) - Update citation in faq (
8fdfbc4)
- Doi for shamout and imrie (
4823e1e)
- Force new release (
e6522d6)
- Description in .zenodo.json (
1a4e31a)
- Add .zenodo.json (
f6bc274)
- Erroneous "." in .bib (
606c7c6)
- Update supported python versions on pypi (
6f7959b)
- Incorrect python version (
b88d7c9)
- Bump version (
c9c768e) - Allow newest 3.10 version (
b465d2f) - Incorrect imports (
d23da12) - Docstrings after ruff formatting (
1563e57) - Incorrect error (
46eebf1) - Incorrect import (
550ed52)
- Spelling error (
747e1f3)
- Add sklearn embedding function (
6c10d3f) - Add optional loader_kwargs to group specs (
14b872c) - Add loader_kwargs to group specs (
a1e5a14) - Add
embedding_fn_kwargstoTextPredictorSpec(7aa48a5) - Example text embedding functions (
f4ce9a2) - Handle text features and features with multiple columns (
a3cafe9) - TextSpec and option for additional feature name .get_col_str (
2986203) - Added concatenate aggregation_fn (
5dc8c9f)
- Fix semantic release (
7e993e3) - Move
embed_text_valuesto use timeseriesflattener without text dependencies installed (487daec) - Proper guard against neither
dfnoroutput_specspecified inget_value_col_str_name(aeef039) - Correct col str name generation for text specs when column names are ints (e.g. HF embeddings) (
c10565d)
- Add text tutorial to docs (
5bb6a02) - Add textpredictorspec docstring (
d492a40) - Update advanced tutorial (
3c3db93) - Minor changes (
aacd222) - Update basic tutorial (
df8d2d6)
- Don't clear all specs as unprocessed (
b7d4f88) - Iterating while deleting items in list leads to skipped items (
5443408) - Static spec ignored outcome col name override (
5ec6892)
- Fixed titles (
7ebf424) - Added tutorial link to readme (
50cdb85) - Updated relay menu (
44cf0f5) - Updated menu structure (
1e5edda) - Reordered elements form readme to documentation site (
327fbf9)
- Compare all indices for maximum safety (
22b7776) - Sample 5 evenly spaced slices (
a0aa5ef) - Log info on chunksize (
2fa5edc) - Check first and last 1000 indices (
df85c66) - Check that ids are similar (
9d47310) - Better logging when dfs are not of identical index (
2cf78d0)
- Decrease log level of debug info (
9ceece1) - Log time to load values (
162fee9) - Log info about cache hits and misses (
237d5b6) - Improve logs (
a24b155) - Better logging when resolving (
9ecc941)
- Typo (
21ea576) - Don't check if output col name override exists in input df (
4cf755d) - Remove duplicate logging (
447be4c)
- Extended docs for autogenerated docstrings after review (
afadaef) - Simpler example, added output (
313e756) - Simpler readme example (
b2b97f3) - Update docstrings (
36a9aae) - Add docstring to all user facing specs (
1744c79) - Merge tutorials with main (
28dc004) - Add docstring to long_df (
1e5247d) - Update advanced tutorial with df dict func (
919ba4a)
- Incorrect version (
baad0e8)
- Replace psycop-feature-generation with timeseriesflattener (
7e35574) - Updates for prettier html rendering (
1614dab) - Add tutorials to sphinx (
9dc410e) - Rerun tutorial (
b7b6d83)
- Only remove outcome spec if added (
bfa7eed) - Remove unprocessed specs when computing (
6b61195) - Remove unprocessed specs when computing (
38404de)
- Move values_df validation that depends on flattened_ds to flattened_ds (
cad634d)
- Enforce unified timestamp col naming across values_df and pred_time_df (
805e8f2) - Remove hardcoded timestamp columns (
66380d2)
- Unintended hardcoded id col name (
dc788df)
- More informative errors (
3141487)
- More docstrings (
e1134d4)
- Take multiple features as long format (
7f771e4) - Add resolve from hydrated dict (
2263c60) - Add function for hydrating long df into cat (
84515e9)
- Only test if loader str can be resolved if it's specified (
5730592) - Check if more than one df attribute is passed (
f0f8623) - Add extra description to error (
85684cf) - Change from catalogue to dict (
4506c3a)
- More appropriate logging level (
7122e79) - Infer aggregation_fn str rep for groups (
8058d65) - Infer aggregation_fn str repr from name (
55be07d) - Add lookahead and lookbehind days to feature group specs (
318591b)
- Guard against incident attribute not existing (
3b1329b) - Re-add aggregation str resolution (
18983a7) - Only infer aggregation_str if not specified manually (
f2648f8) - Create dir for diskcache if it doesn't exist (
0e32436) - Create dir if it doesn't exist (
c32f3c7)
- Update output (
a4fa6f7) - Update tutorial based on feedback (
92c3d3b) - Add figures to basic tutorial (
5eb069f) - Misc. updates to advanced notebook (
a4a9380) - Groupspec add output type (
0b3df30) - Update formatting in 02_advanced (
47065e9) - Initial stab at advanced tutorial (
e8128bd)
- Add diff when dropping rows (
0421ed8) - No def arg for drop_if_insufficient_look_direction (
d290153) - Drop pred times with insufficient look distance (
8ec6e2c) - First stab at dropping unused pred times (
66bb7d4) - Check that all specs have required columns in values_df (
9da16f0) - Better logging (
44eb010) - Process all temporal specs in one batch (
01b3957) - When init spec, coerce timestamp if possible (
4a6f817) - Process specs all at once (
4bcdb82) - Collect specs with one interface (
7135c4d)
- Incorrect dim comparison in diff decorator (
c161f9a) - Revert TemporalSpec renaming (
3140cd5) - Missing column should be a keyerror (
452e903) - Undo renaming of PredictorSpec -> TemporalSpec (
150035c) - Don't process as batch if no specs to process (
aba0b67) - Only batch process predictors if any are added (
c21737c) - eq in AnySpec (
4c650d1)
- Improve from review suggestions (
57804df) - Improve from review suggestions (
0316dbe) - Elaborate on drop_pred_time (
254fb3d) - Update tutorial to new interface (
bc0405e) - Raise valueerror if no prediction times remain (
18c7f82) - Improve drop_records docs. (
871ec18) - Minor docstring edit (
1a7267a)
- Allow either interval_days or lookahead/lookbehind days (
a270801)
- Failing imports after merge (
dd17771) - Unify file naming in cache module (
f155217) - Use correct suffix (
6e737b8) - Remove seconds from diskcache to avoid (
f9ac05c) - Key_for_aggregation should be optional (
c569b74)
- Add basic tutorial (
8136a1b)
- Add colored logging (
d230213) - Add logging by default (
d254b69) - Refactor flattened_dataset to use logging instead of msg (
d9fc31d)
- Check that all col names exist in df before creating spec (
7e75001)
- Style (
d56926e)
- Override cache attributes if unset or None (
9f896c8) - Duplication of citizen id columns when reading cache (
d78340c)
- Raise warning if overriding cache attributes (
ffba27e) - Fixed citation (
ef3312a) - Update citation.cff with zonedo (
ef3fc65) - Added pypi badge (
218cbcc) - Improvements (
1726246) - Misc. (
338b2b1)
- Added explanation comment to token (
9ae8716)
timeseriesflattener have been renamed timeseriesflattener
- Add n_hba1c_within_n_lookahead_days (
e84b591) - Add outcome (
cd39dd6) - Add birth year as a predictor (
7b186d2) - Allow exclusion of specific atc codes (
75619a1)
- Date of birth col name should respect output prefix (
6ec6535) - Incorrect column name when adding age as predictor (
cdbf25c) - Errors in sql loaders after refactor (
28c9f63) - Correct type hinting in load_diagnoses (
f2d5c5b)
- Speccify that n_rows = None returns all rows. (
a4720a8)
- Shuffle feature specs to even out compute vs. IO load (
0db9f0f) - Tweak n_workers for more performance (
3eeee4d) - Segment feature loading for more parallelisation (
9ee5c87) - Rotate feature addition for debugging (
76af9c7) - Parallelise temporal predictor loading (
8d53f16) - Only create one subprocess per values loader (
1a3e5de) - Parralelise groupspec combination creation (
9ccba2a)
- At groupspec init, iterate over values_loader and check that they exist in the loader registry (
04dfd7e)
- More explanation in error message (
b784991) - Bettee valueerror message formatting (
7b3b994) - Better valueerror message (
d92f798) - Find invalid loaders (
ba2d4c5)
- Allow load_medications to concat a list of medications (
d78f465)
- Remove original functions (
da59110)
- Improve docs (
9aad0af)
- Full run (
142212f) - Rename aggregation registry keys to their previous one (
3fd3f35) - Reimplement (
c99585f) - Use lru cache decorator for values_df loading (
4006818) - Add support for loader kwargs (
127f821) - Move values_df resolution to anyspec object (
714e83f) - Make date of birth output prefix a param (
0ed1198) - Ensure that dfs are sorted and of same length before concat (
84a4d65) - Use pandas with set_index for concat (
b93290a) - Use pandas with set_index for concat (
995da41) - Speed up dask join by using index (
3402281) - Require feature name for all features, ensures proper specification (
6af454a) - First stab at adapting generate_main (
7243130) - Add exclusion timestamp (
b02de1a) - Improve dd.concat (
429da34) - Handle strs for generate_feature_spec (
7d54488) - Convert to dd before concat (
06101d8) - Add n hba1c (
3780d84) - Add n hba1c (
614245e)
- Coerce by default (
60adb99) - Output_col_name_override applied at loading, not flattening (
95a96ce) - Typo (
01240ed) - Incorrect attribute addressing (
a6e82b5) - Correctly resolve values_df (
def67cd) - MinGroupSpec should take a sequence of name to permute over (
f0c8140) - Typo (
61c7241) - Remove aggregation_fn_name (
617d386) - Old concat resulted in wrong ordering of rrows. (
3759f71) - Set hba1c as eval (
89fe6d2) - Typos (
6eac440) - Correct col name inference for static predictors (
dfe5dc7) - Misc. fixes (
45f8348) - Generate the correct amount of combinations when creating specs (
c472b3c) - Typo resulted in cache breaking (
fdd47d7) - Correct col naming (
bc74ae3) - Do not infer feature name from values_df (
150569f) - Misc. errors found from tests (
3a1b5db) - Revert falttened dataset to use specs (
e4fada7) - Misc. errors after introducing feature specs (
0308eca) - Correctly merge dataframes (
a907885) - Cache error because of loss off UUID (
89d7f6f) - New bugs in aggregation (
5714a39) - Rename outcomespec appropriately (
41fa220) - Lookbehind_days must be iterable (
cc879e9)
- Move pd->dd into subprocesses (
dc5f38d)
- Remove shak_code + operator check (
f97aee8)
- Ignore cat_features (
2052505) - Failing test (
f8190b4) - Incorrect 'latest' and handling of NaN in cache (
dc33f7e)
- Check for value column prediction_times_df (
5356464) - Change variable name (
990a848) - More flex loaders (
bcad700)
- Use wandb to monitor script errors (
67ae9b9)
- Duplicate loading when pre_loading dfs (
7f864dc)
- Add variance to resolve multiple functions (
8c471df)
- Add vairance resolve multiple (
7a64c5b)
- Deleted_irritating_blank_space (
a4cdfc5)
- Auto inferred cat features (
ea0d946) - Auto inferred cat features error (
f244715) - Resolves errors caused from auto cat features (
667a905)
- Incorrect function argument (
33e0a3e) - Expanded test to include outcome, now passes locally (
640e7ec) - Passing local tests (
6ed4b2e) - First stab at bug fix (
339d793)
- Add parents to wandb dir init (
5eefe3a)
- Add BMI loader (
b6681ea)
- Refactor feature spec generation (
17e9f16) - Align arguments with colnames in SQL (
09ae5f7) - Refactor feature specification (
373b0f0)
- Hardcoded file suffix (
0101acc)
- Mismatched version in .tomll (
292979b)
- Pass value_col only when necessary (
dc1019f) - Pass value_col (
4674e4a) - Don't remove NaNs, might be informative. (
1ad5d81) - Remove parquet default argument except in top level functions (
ec3a98b) - Align .toml and release version (
80adbde) - Failing tests (
b5e4321) - Incorrect feature sets path, linting (
605ccb7) - Handle dicts for duplicate checking (
34524c0) - Check for duplicates in feature combinations (
63ad162) - Remove duplicate alat key which prevented file saving (
f0c3e00) - Incorrect argument (
b97d54b) - Linting (
7406288) - Use suffix instead of string parsing (
cfa96f0) - Refactor dataset loading into a separate function (
bca8cbf) - More migration to parquet (
f1bc2b7) - Mark hf embedding test as slow, only run if passing --runslow to pytest (
0e03395)
- Wandb not logging on overtaci. (
3baab57)
- Use dask for concatenation, increases perf (
4235f5c)
- Use pypi release of psycopmlutils (
5283b05)
- First release to pypi (
c29aa3c)
- Add test for chunking logic (
199ee6b)