AutoML Alex
State-of-the art Automated Machine Learning python library for Tabular Data
Works with Tasks:
-
Binary Classification
-
Regression
-
Multiclass Classification (in progress...)
Benchmark Results
The bigger, the better
From AutoML-Benchmark
Scheme
Features
- Automated Data Clean (Auto Clean)
- Automated Feature Engineering (Auto FE)
- Smart Hyperparameter Optimization (HPO)
- Feature Generation
- Feature Selection
- Models Selection
- Cross Validation
- Optimization Timelimit and EarlyStoping
- Save and Load (Predict new data)
Installation
pip install automl-alex
Docs
🚀 Examples
Classifier:
from automl_alex import AutoMLClassifier
model = AutoMLClassifier()
model.fit(X_train, y_train, timeout=600)
predicts = model.predict(X_test)
Regression:
from automl_alex import AutoMLRegressor
model = AutoMLRegressor()
model.fit(X_train, y_train, timeout=600)
predicts = model.predict(X_test)
DataPrepare:
from automl_alex import DataPrepare
de = DataPrepare()
X_train = de.fit_transform(X_train)
X_test = de.transform(X_test)
Simple Models Wrapper:
from automl_alex import LightGBMClassifier
model = LightGBMClassifier()
model.fit(X_train, y_train)
predicts = model.predict_proba(X_test)
model.opt(X_train, y_train,
timeout=600, # optimization time in seconds,
)
predicts = model.predict_proba(X_test)
More examples in the folder ./examples:
- 01_Quick_Start.ipynb
- 02_Data_Cleaning_and_Encoding_(DataPrepare).ipynb
- 03_Models.ipynb
- 04_ModelsReview.ipynb
- 05_BestSingleModel.ipynb
- Production Docker template
What's inside
It integrates many popular frameworks:
- scikit-learn
- XGBoost
- LightGBM
- CatBoost
- Optuna
- ...
Works with Features
-
Categorical Features
-
Numerical Features
-
Binary Features
-
Text
-
Datetime
-
Timeseries
-
Image
Note
- With a large dataset, a lot of memory is required! Library creates many new features. If you have a large dataset with a large number of features (more than 100), you may need a lot of memory.
Realtime Dashboard
Works with optuna-dashboard
Run
$ optuna-dashboard sqlite:///db.sqlite3
Road Map
-
Feature Generation
-
Save/Load and Predict on New Samples
-
Advanced Logging
-
Add opt Pruners
-
Docs Site
-
DL Encoders
-
Add More libs (NNs)
-
Multiclass Classification
-
Build pipelines
Contact
Metadata
Release files for automl-alex 2023.3.11
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 |
|---|---|---|---|---|
| automl-alex-2023.3.11.tar.gz | 32.6 kB | Without provenance | Details for automl-alex-2023.3.11.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| automl_alex-2023.3.11-py3-none-any.whl | Python 3 | none | any | Details for automl_alex-2023.3.11-py3-none-any.whl |
Total release size: 70.8 kB
Release files / automl-alex-2023.3.11.tar.gz
| Download URL | automl-alex-2023.3.11.tar.gz |
|---|---|
| Size | 32.6 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
fbedbb3d32ee74dc472d10b75d583afbe351a1050255e4d978663724068cea78
|
|
BLAKE2b-256 checksum How to use checksums |
4ebe345d3d0a44e7ee8e9432c049a1b7663ecf84923f4babc4465b37d5a0f24c
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
poetry/1.1.14 CPython/3.10.9 Linux/5.4.0-104-generic
|
Release files / automl_alex-2023.3.11-py3-none-any.whl
| Download URL | automl_alex-2023.3.11-py3-none-any.whl |
|---|---|
| Size | 38.2 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
2a303a68b253389e05ee9616554c1bd09ba0d3b6a370ce132732547325b4257d
|
|
BLAKE2b-256 checksum How to use checksums |
789c0041fdcb8379640a765627eaccecf37365ce0bf61517e423ff7337b3d969
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
poetry/1.1.14 CPython/3.10.9 Linux/5.4.0-104-generic
|
Security for automl-alex 2023.3.11
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.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A 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.