FaissImputer
Warning: FaissImputer 0.1.x has a known neighbor-mapping bug that can produce incorrect imputations, and it is incompatible with scikit-learn 1.8+. Use version 0.2.0 or newer.
A scikit-learn-compatible missing-value imputer with Faiss-backed neighbor search.
Current release: 0.3.4. See Releases for version history.
Performance at a glance
FaissImputer can accelerate nearest-neighbor imputation, especially when queries reuse the same missing-feature patterns.
Which donor policy should I use?
A donor is a training row used to supply a missing value.
complete(default): Uses only fully observed training rows as donors. Rows containing missing values are excluded from neighbor search. Choose this when you have enough complete training rows.available: Also allows partially observed training rows as donors. Donors are selected separately for each missing feature: they must contain that feature's value and share at least one originally observed feature with the query row. No fully observed training row is required.
The benchmarks below use fully observed training data for complete
and partially missing training data for available. They compare each
policy against KNNImputer, not the two policies against each other.
Selected benchmark results:
| Donor policy | Training rows | Query missingness | KNNImputer | FaissImputer | Speedup |
|---|---|---|---|---|---|
| complete | 1,000 | Random patterns | 22.6 ms | 30.9 ms | 0.73× — slower |
| complete | 20,000 | One shared pattern | 314.3 ms | 34.8 ms | 9.04× |
| complete | 20,000 | Random patterns | 316.8 ms | 133.0 ms | 2.38× |
| available | 500,000 | Random patterns | 9.027 s | 6.179 s | 1.46× |
| available | 1,000,000 | Random patterns | 21.724 s | 19.681 s | 1.10× |
What was measured: fit + transform on synthetic data, with 300 query
rows, 20 features, 5 neighbors, mean aggregation, and one native thread.
Each query has four missing features. Training data is fully observed
for complete, and has 10% MCAR missingness for available.
Complete-policy times are medians across five seeds of three-run medians,
measured at commit 9d179b2b. Available-policy results measure the
128 MiB batching candidate adopted in 0.3.2: three-run medians at
500,000 rows and a single-run pilot at 1,000,000 rows.
These are measured development snapshots, not a fresh benchmark of the
published 0.3.2 package. CPU models differ between experiments;
compare methods within each row. Speedups use unrounded times.
Output agreement: the complete-policy benchmark matched KNNImputer
exactly on the tested inputs. The available-policy experiments had a
maximum absolute difference of approximately 2.38e-7 from KNNImputer.
This measures agreement, not accuracy against ground truth or a guarantee
for other datasets.
Tradeoffs: small datasets with varied missingness can be slower. Memory use is not always lower: in the million-row pilot, whole-worker peak RSS was about 9.1% higher than KNNImputer.
Complete-policy results and reproduction · Available-policy results, memory and thread comparisons
Installation
FaissImputer requires Python 3.10 or newer.
python -m pip install --upgrade "faiss-imputer>=0.3.4"
Usage
Complete donors (default)
Training data must contain enough fully observed rows to supply n_neighbors donors.
import numpy as np
from faiss_imputer import FaissImputer
X_train = np.array(
[
[1.0, 10.0, 100.0],
[2.0, 20.0, 200.0],
[3.0, 30.0, 300.0],
],
dtype=np.float32,
)
X_missing = np.array(
[
[1.5, np.nan, 150.0],
[2.5, 25.0, np.nan],
],
dtype=np.float32,
)
imputer = FaissImputer(n_neighbors=2)
X_imputed = imputer.fit(X_train).transform(X_missing)
print(X_imputed)
Expected output:
[[ 1.5 15. 150. ]
[ 2.5 25. 250. ]]
Partially observed donors
Set donor_policy="available" to choose donors separately for each missing feature. No fully observed training row is required.
import numpy as np
from faiss_imputer import FaissImputer
# No training row is completely observed.
X_train = np.array(
[
[0.0, 10.0, np.nan],
[2.0, 30.0, np.nan],
[1.0, np.nan, 20.0],
[3.0, np.nan, 40.0],
],
dtype=np.float32,
)
X_missing = np.array(
[[0.1, np.nan, np.nan], [np.nan, np.nan, np.nan]],
dtype=np.float32,
)
imputer = FaissImputer(n_neighbors=1, donor_policy="available")
X_imputed = imputer.fit(X_train).transform(X_missing)
print(X_imputed)
Expected output:
[[ 0.1 10. 20. ]
[ 1.5 20. 30. ]]
By default, both policies return a new NumPy float32 array. The training and query inputs are not modified.
Feature names and pandas output
The following example requires FaissImputer 0.3.3 or newer.
It also requires pandas, which is optional:
python -m pip install pandas.
import pandas as pd
from sklearn.pipeline import make_pipeline
from sklearn.preprocessing import StandardScaler
from faiss_imputer import FaissImputer
X = pd.DataFrame({
"age": [20.0, None, 40.0],
"income": [100.0, 190.0, 300.0],
})
pipe = make_pipeline(
FaissImputer(n_neighbors=1),
StandardScaler(),
).set_output(transform="pandas")
result = pipe.fit_transform(X)
print(pipe.get_feature_names_out()) # ['age' 'income']
print(result.columns.tolist()) # ['age', 'income']
Use pipe.set_output(transform="default") to return NumPy arrays instead.
For unnamed array inputs, feature names are generated as x0, x1, and so on.
Parameters
n_neighbors(default:3): Positive integer specifying the maximum number of donors used for each missing feature. The complete-donor policy requires at least this many complete training rows.metric(default:"l2"): Supports"l2"and"ip". Raw inner product is not cosine similarity. The available-donor policy requires"l2".strategy(default:"mean"): Supports"mean"and"median"for aggregating donor values and calculating fallback column statistics.index_factory(default:"Flat"): Faiss index description for the complete-donor policy. The available-donor policy accepts only"Flat"and uses the distance backend described below.donor_policy(default:"complete"): Use fully observed training rows with"complete", or allow partially observed training rows with"available".
Important behavior
Shared behavior
- Inputs must be two-dimensional numeric array-like data, with
NaNmarking missing values. Values are converted tofloat32; infinity is not accepted. transform()requires the same number of features asfit().- An entirely missing query row uses column means or medians learned during
fit(). - A failed
fit(), including a failed refit, clears the fitted state.
Complete-donor policy
- Only fully observed training rows are used as donors. At least one complete row is required.
n_neighborscannot exceed the number of complete donors.- Neighbor search uses only the originally observed columns of each query row.
- The default
index_factory="Flat"performs exact neighbor search.
Available-donor policy
- Only
metric="l2"andindex_factory="Flat"are supported. Both"mean"and"median"strategies are supported. - A donor must observe the feature being imputed and share at least one originally observed feature with the query row.
- Donors are ranked by squared L2 distance over shared observed features, scaled by the total feature count divided by the shared feature count.
- Each missing feature uses up to
n_neighborseligible donors. Fewer eligible donors are allowed. - If no eligible donor exists for a feature, its fitted column mean or median is used.
- Entirely missing training rows are ignored; entirely missing training columns are rejected.
The available-donor policy uses batched NaN-aware distances and Faiss neighbor selection, with float64 safeguards for detected numerical risks. This backend is also used when all training donors are complete.
Batching trades memory for throughput. Its internal batch-sizing budget is not a total process-memory limit: fitted donor data, distance calculations, and temporary arrays also consume memory. Larger batches can improve speed but increase peak memory usage. FaissImputer does not set the number of threads; configure thread limits in the application when needed.
Donor-side arrays are prepared during fit() and reused across query batches.
This reduces repeated work but increases retained memory after fitting;
lower peak memory is not guaranteed.
Benchmarks
Performance depends on donor policy, data size, and missingness patterns. Neither faster execution nor lower memory use than KNNImputer is guaranteed. Similar average errors do not imply identical imputed values; ties and numerical precision can affect donor selection.
- Available-donor batching and threads: compares 16/64/128 MiB batch budgets and 1/2/4 threads, including repeated 500,000-row measurements and a single-run million-row pilot. Documents both speed gains and memory tradeoffs.
- Real-data pilot: compares SimpleImputer, KNNImputer, and both Faiss donor policies under MCAR and selected MAR missingness. This small dataset is not a scalability test.
- Complete-donor patterns: measures the impact of training size and query missingness patterns.
- Partial-donor development snapshot: historical results for measured commit
0a3cc077, not the final 0.3.0 implementation. - Historical 0.2.0 benchmark: results for the earlier complete-donor-only implementation.
Reports provide measurement conditions, results, and links to reproduction code and raw data.
Example notebook
See Imputing Missing Values with Faiss Imputer for a complete-donor example.
Contributing
Contributions are welcome! Please open an issue or create a pull request. Further work is tracked in the roadmap.
Author
- GitHub: @ScionKim
License
This project is licensed under the MIT License.
Third-Party Licenses
FaissImputer depends on Meta's Faiss, which is distributed under the MIT License.
FaissImputer is not affiliated with or endorsed by Meta or the Faiss maintainers.
Metadata
Release files for faiss-imputer 0.3.4
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| faiss_imputer-0.3.4.tar.gz | 24.2 kB | With provenance | Details for faiss_imputer-0.3.4.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| faiss_imputer-0.3.4-py3-none-any.whl | Python 3 | none | any | Details for faiss_imputer-0.3.4-py3-none-any.whl |
Total release size: 35.7 kB
Release files / faiss_imputer-0.3.4.tar.gz
| Download URL | faiss_imputer-0.3.4.tar.gz |
|---|---|
| Size | 24.2 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
97916ab01486a1a10ef6abd8f26bc2d578c2fde105e31da3c6866a637d6c94f3
|
|
BLAKE2b-256 checksum How to use checksums |
3ad39ce5c0f84c98194337ccb6ce74fd926dedba468b2a503129a15027cdaa94
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Sep 4, 2026.
Transparency logRelease files / faiss_imputer-0.3.4-py3-none-any.whl
| Download URL | faiss_imputer-0.3.4-py3-none-any.whl |
|---|---|
| Size | 11.4 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
c545454c796f68cd042d69e5a1ae8d0d81ac7d4b5bdafb959fe78fedfdda6398
|
|
BLAKE2b-256 checksum How to use checksums |
cd66bfcdeee69ad013bcd5f6617cdb78740611872e98a037e83f3dd2b02a4caf
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Sep 4, 2026.
Transparency logSecurity for faiss-imputer 0.3.4
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
Full provenance
Both files in this release have been uploaded with provenance metadata.
Attestation for faiss_imputer-0.3.4.tar.gz
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Sep 4, 2026.
Transparency logA note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.