{ "cells": [ { "attachments": {}, "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "# AutoML with FLAML Library\n", "\n", "\n", "| | | | |\n", "|-----|--------|--------|--------|\n", "| \"drawing\" \n", "\n", "\n", "\n", "### Goal\n", "In this notebook, we demonstrate how to use AutoML with FLAML to find the best model for our dataset.\n", "\n", "\n", "## 1. Introduction\n", "\n", "FLAML is a Python library (https://github.com/microsoft/FLAML) designed to automatically produce accurate machine learning models \n", "with low computational cost. It is fast and economical. The simple and lightweight design makes it easy to use and extend, such as adding new learners. FLAML can \n", "- serve as an economical AutoML engine,\n", "- be used as a fast hyperparameter tuning tool, or \n", "- be embedded in self-tuning software that requires low latency & resource in repetitive\n", " tuning tasks.\n", "\n", "In this notebook, we use one real data example (binary classification) to showcase how to use FLAML library.\n", "\n", "FLAML requires `Python>=3.7`. To run this notebook example, please install the following packages." ] }, { "cell_type": "code", "execution_count": 39, "metadata": { "jupyter": { "outputs_hidden": true } }, "outputs": [ { "data": { "application/vnd.livy.statement-meta+json": { "execution_finish_time": "2023-04-09T03:11:05.782522Z", "execution_start_time": "2023-04-09T03:11:05.7822033Z", "livy_statement_state": "available", "parent_msg_id": "18b2ee64-09c4-4ceb-8975-e4ed43d7c41a", "queued_time": "2023-04-09T03:10:33.571519Z", "session_id": "7", "session_start_time": null, "spark_jobs": null, "spark_pool": null, "state": "finished", "statement_id": -1 }, "text/plain": [ "StatementMeta(, 7, -1, Finished, Available)" ] }, "metadata": {}, "output_type": "display_data" }, { "data": {}, "execution_count": 39, "metadata": {}, "output_type": "execute_result" }, { "name": "stdout", "output_type": "stream", "text": [ "Collecting flaml[synapse]==1.1.3\n", " Using cached FLAML-1.1.3-py3-none-any.whl (224 kB)\n", "Collecting xgboost==1.6.1\n", " Using cached xgboost-1.6.1-py3-none-manylinux2014_x86_64.whl (192.9 MB)\n", "Collecting pandas==1.5.1\n", " Using cached pandas-1.5.1-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (12.2 MB)\n", "Collecting numpy==1.23.4\n", " Using cached numpy-1.23.4-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (17.1 MB)\n", "Collecting openml\n", " Using cached openml-0.13.1-py3-none-any.whl\n", "Collecting scipy>=1.4.1\n", " Using cached scipy-1.10.1-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (34.5 MB)\n", "Collecting scikit-learn>=0.24\n", " Using cached scikit_learn-1.2.2-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (9.8 MB)\n", "Collecting lightgbm>=2.3.1\n", " Using cached lightgbm-3.3.5-py3-none-manylinux1_x86_64.whl (2.0 MB)\n", "Collecting pyspark>=3.0.0\n", " Using cached pyspark-3.3.2-py2.py3-none-any.whl\n", "Collecting optuna==2.8.0\n", " Using cached optuna-2.8.0-py3-none-any.whl (301 kB)\n", "Collecting joblibspark>=0.5.0\n", " Using cached joblibspark-0.5.1-py3-none-any.whl (15 kB)\n", "Collecting python-dateutil>=2.8.1\n", " Using cached python_dateutil-2.8.2-py2.py3-none-any.whl (247 kB)\n", "Collecting pytz>=2020.1\n", " Using cached pytz-2023.3-py2.py3-none-any.whl (502 kB)\n", "Collecting cliff\n", " Using cached cliff-4.2.0-py3-none-any.whl (81 kB)\n", "Collecting packaging>=20.0\n", " Using cached packaging-23.0-py3-none-any.whl (42 kB)\n", "Collecting cmaes>=0.8.2\n", " Using cached cmaes-0.9.1-py3-none-any.whl (21 kB)\n", "Collecting sqlalchemy>=1.1.0\n", " Using cached SQLAlchemy-2.0.9-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (2.8 MB)\n", "Collecting tqdm\n", " Using cached tqdm-4.65.0-py3-none-any.whl (77 kB)\n", "Collecting alembic\n", " Using cached alembic-1.10.3-py3-none-any.whl (212 kB)\n", "Collecting colorlog\n", " Using cached colorlog-6.7.0-py2.py3-none-any.whl (11 kB)\n", "Collecting xmltodict\n", " Using cached xmltodict-0.13.0-py2.py3-none-any.whl (10.0 kB)\n", "Collecting requests\n", " Using cached requests-2.28.2-py3-none-any.whl (62 kB)\n", "Collecting minio\n", " Using cached minio-7.1.14-py3-none-any.whl (77 kB)\n", "Collecting liac-arff>=2.4.0\n", " Using cached liac_arff-2.5.0-py3-none-any.whl\n", "Collecting pyarrow\n", " Using cached pyarrow-11.0.0-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (35.0 MB)\n", "Collecting joblib>=0.14\n", " Using cached joblib-1.2.0-py3-none-any.whl (297 kB)\n", "Collecting wheel\n", " Using cached wheel-0.40.0-py3-none-any.whl (64 kB)\n", "Collecting py4j==0.10.9.5\n", " Using cached py4j-0.10.9.5-py2.py3-none-any.whl (199 kB)\n", "Collecting six>=1.5\n", " Using cached six-1.16.0-py2.py3-none-any.whl (11 kB)\n", "Collecting threadpoolctl>=2.0.0\n", " Using cached threadpoolctl-3.1.0-py3-none-any.whl (14 kB)\n", "Collecting urllib3\n", " Using cached urllib3-1.26.15-py2.py3-none-any.whl (140 kB)\n", "Collecting certifi\n", " Using cached certifi-2022.12.7-py3-none-any.whl (155 kB)\n", "Collecting idna<4,>=2.5\n", " Using cached idna-3.4-py3-none-any.whl (61 kB)\n", "Collecting charset-normalizer<4,>=2\n", " Using cached charset_normalizer-3.1.0-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (195 kB)\n", "Collecting typing-extensions>=4.2.0\n", " Using cached typing_extensions-4.5.0-py3-none-any.whl (27 kB)\n", "Collecting greenlet!=0.4.17\n", " Using cached greenlet-2.0.2-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (618 kB)\n", "Collecting importlib-metadata\n", " Using cached importlib_metadata-6.2.0-py3-none-any.whl (21 kB)\n", "Collecting importlib-resources\n", " Using cached importlib_resources-5.12.0-py3-none-any.whl (36 kB)\n", "Collecting Mako\n", " Using cached Mako-1.2.4-py3-none-any.whl (78 kB)\n", "Collecting autopage>=0.4.0\n", " Using cached autopage-0.5.1-py3-none-any.whl (29 kB)\n", "Collecting cmd2>=1.0.0\n", " Using cached cmd2-2.4.3-py3-none-any.whl (147 kB)\n", "Collecting stevedore>=2.0.1\n", " Using cached stevedore-5.0.0-py3-none-any.whl (49 kB)\n", "Collecting PrettyTable>=0.7.2\n", " Using cached prettytable-3.6.0-py3-none-any.whl (27 kB)\n", "Collecting PyYAML>=3.12\n", " Using cached PyYAML-6.0-cp38-cp38-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl (701 kB)\n", "Collecting attrs>=16.3.0\n", " Using cached attrs-22.2.0-py3-none-any.whl (60 kB)\n", "Collecting pyperclip>=1.6\n", " Using cached pyperclip-1.8.2-py3-none-any.whl\n", "Collecting wcwidth>=0.1.7\n", " Using cached wcwidth-0.2.6-py2.py3-none-any.whl (29 kB)\n", "Collecting zipp>=0.5\n", " Using cached zipp-3.15.0-py3-none-any.whl (6.8 kB)\n", "Collecting pbr!=2.1.0,>=2.0.0\n", " Using cached pbr-5.11.1-py2.py3-none-any.whl (112 kB)\n", "Collecting MarkupSafe>=0.9.2\n", " Using cached MarkupSafe-2.1.2-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (25 kB)\n", "Installing collected packages: wcwidth, pytz, pyperclip, py4j, zipp, xmltodict, wheel, urllib3, typing-extensions, tqdm, threadpoolctl, six, PyYAML, pyspark, PrettyTable, pbr, packaging, numpy, MarkupSafe, liac-arff, joblib, idna, greenlet, colorlog, charset-normalizer, certifi, autopage, attrs, stevedore, sqlalchemy, scipy, requests, python-dateutil, pyarrow, minio, Mako, joblibspark, importlib-resources, importlib-metadata, cmd2, cmaes, xgboost, scikit-learn, pandas, cliff, alembic, optuna, openml, lightgbm, flaml\n", " Attempting uninstall: wcwidth\n", " Found existing installation: wcwidth 0.2.6\n", " Uninstalling wcwidth-0.2.6:\n", " Successfully uninstalled wcwidth-0.2.6\n", " Attempting uninstall: pytz\n", " Found existing installation: pytz 2023.3\n", " Uninstalling pytz-2023.3:\n", " Successfully uninstalled pytz-2023.3\n", " Attempting uninstall: pyperclip\n", " Found existing installation: pyperclip 1.8.2\n", " Uninstalling pyperclip-1.8.2:\n", " Successfully uninstalled pyperclip-1.8.2\n", " Attempting uninstall: py4j\n", " Found existing installation: py4j 0.10.9.5\n", " Uninstalling py4j-0.10.9.5:\n", " Successfully uninstalled py4j-0.10.9.5\n", " Attempting uninstall: zipp\n", " Found existing installation: zipp 3.15.0\n", " Uninstalling zipp-3.15.0:\n", " Successfully uninstalled zipp-3.15.0\n", " Attempting uninstall: xmltodict\n", " Found existing installation: xmltodict 0.13.0\n", " Uninstalling xmltodict-0.13.0:\n", " Successfully uninstalled xmltodict-0.13.0\n", " Attempting uninstall: wheel\n", " Found existing installation: wheel 0.40.0\n", " Uninstalling wheel-0.40.0:\n", " Successfully uninstalled wheel-0.40.0\n", " Attempting uninstall: urllib3\n", " Found existing installation: urllib3 1.26.15\n", " Uninstalling urllib3-1.26.15:\n", " Successfully uninstalled urllib3-1.26.15\n", " Attempting uninstall: typing-extensions\n", " Found existing installation: typing_extensions 4.5.0\n", " Uninstalling typing_extensions-4.5.0:\n", " Successfully uninstalled typing_extensions-4.5.0\n", " Attempting uninstall: tqdm\n", " Found existing installation: tqdm 4.65.0\n", " Uninstalling tqdm-4.65.0:\n", " Successfully uninstalled tqdm-4.65.0\n", " Attempting uninstall: threadpoolctl\n", " Found existing installation: threadpoolctl 3.1.0\n", " Uninstalling threadpoolctl-3.1.0:\n", " Successfully uninstalled threadpoolctl-3.1.0\n", " Attempting uninstall: six\n", " Found existing installation: six 1.16.0\n", " Uninstalling six-1.16.0:\n", " Successfully uninstalled six-1.16.0\n", " Attempting uninstall: PyYAML\n", " Found existing installation: PyYAML 6.0\n", " Uninstalling PyYAML-6.0:\n", " Successfully uninstalled PyYAML-6.0\n", " Attempting uninstall: pyspark\n", " Found existing installation: pyspark 3.3.2\n", " Uninstalling pyspark-3.3.2:\n", " Successfully uninstalled pyspark-3.3.2\n", " Attempting uninstall: PrettyTable\n", " Found existing installation: prettytable 3.6.0\n", " Uninstalling prettytable-3.6.0:\n", " Successfully uninstalled prettytable-3.6.0\n", " Attempting uninstall: pbr\n", " Found existing installation: pbr 5.11.1\n", " Uninstalling pbr-5.11.1:\n", " Successfully uninstalled pbr-5.11.1\n", " Attempting uninstall: packaging\n", " Found existing installation: packaging 23.0\n", " Uninstalling packaging-23.0:\n", " Successfully uninstalled packaging-23.0\n", " Attempting uninstall: numpy\n", " Found existing installation: numpy 1.23.4\n", " Uninstalling numpy-1.23.4:\n", " Successfully uninstalled numpy-1.23.4\n", " Attempting uninstall: MarkupSafe\n", " Found existing installation: MarkupSafe 2.1.2\n", " Uninstalling MarkupSafe-2.1.2:\n", " Successfully uninstalled MarkupSafe-2.1.2\n", " Attempting uninstall: liac-arff\n", " Found existing installation: liac-arff 2.5.0\n", " Uninstalling liac-arff-2.5.0:\n", " Successfully uninstalled liac-arff-2.5.0\n", " Attempting uninstall: joblib\n", " Found existing installation: joblib 1.2.0\n", " Uninstalling joblib-1.2.0:\n", " Successfully uninstalled joblib-1.2.0\n", " Attempting uninstall: idna\n", " Found existing installation: idna 3.4\n", " Uninstalling idna-3.4:\n", " Successfully uninstalled idna-3.4\n", " Attempting uninstall: greenlet\n", " Found existing installation: greenlet 2.0.2\n", " Uninstalling greenlet-2.0.2:\n", " Successfully uninstalled greenlet-2.0.2\n", " Attempting uninstall: colorlog\n", " Found existing installation: colorlog 6.7.0\n", " Uninstalling colorlog-6.7.0:\n", " Successfully uninstalled colorlog-6.7.0\n", " Attempting uninstall: charset-normalizer\n", " Found existing installation: charset-normalizer 3.1.0\n", " Uninstalling charset-normalizer-3.1.0:\n", " Successfully uninstalled charset-normalizer-3.1.0\n", " Attempting uninstall: certifi\n", " Found existing installation: certifi 2022.12.7\n", " Uninstalling certifi-2022.12.7:\n", " Successfully uninstalled certifi-2022.12.7\n", " Attempting uninstall: autopage\n", " Found existing installation: autopage 0.5.1\n", " Uninstalling autopage-0.5.1:\n", " Successfully uninstalled autopage-0.5.1\n", " Attempting uninstall: attrs\n", " Found existing installation: attrs 22.2.0\n", " Uninstalling attrs-22.2.0:\n", " Successfully uninstalled attrs-22.2.0\n", " Attempting uninstall: stevedore\n", " Found existing installation: stevedore 5.0.0\n", " Uninstalling stevedore-5.0.0:\n", " Successfully uninstalled stevedore-5.0.0\n", " Attempting uninstall: sqlalchemy\n", " Found existing installation: SQLAlchemy 2.0.9\n", " Uninstalling SQLAlchemy-2.0.9:\n", " Successfully uninstalled SQLAlchemy-2.0.9\n", " Attempting uninstall: scipy\n", " Found existing installation: scipy 1.10.1\n", " Uninstalling scipy-1.10.1:\n", " Successfully uninstalled scipy-1.10.1\n", " Attempting uninstall: requests\n", " Found existing installation: requests 2.28.2\n", " Uninstalling requests-2.28.2:\n", " Successfully uninstalled requests-2.28.2\n", " Attempting uninstall: python-dateutil\n", " Found existing installation: python-dateutil 2.8.2\n", " Uninstalling python-dateutil-2.8.2:\n", " Successfully uninstalled python-dateutil-2.8.2\n", " Attempting uninstall: pyarrow\n", " Found existing installation: pyarrow 11.0.0\n", " Uninstalling pyarrow-11.0.0:\n", " Successfully uninstalled pyarrow-11.0.0\n", " Attempting uninstall: minio\n", " Found existing installation: minio 7.1.14\n", " Uninstalling minio-7.1.14:\n", " Successfully uninstalled minio-7.1.14\n", " Attempting uninstall: Mako\n", " Found existing installation: Mako 1.2.4\n", " Uninstalling Mako-1.2.4:\n", " Successfully uninstalled Mako-1.2.4\n", " Attempting uninstall: joblibspark\n", " Found existing installation: joblibspark 0.5.1\n", " Uninstalling joblibspark-0.5.1:\n", " Successfully uninstalled joblibspark-0.5.1\n", " Attempting uninstall: importlib-resources\n", " Found existing installation: importlib-resources 5.12.0\n", " Uninstalling importlib-resources-5.12.0:\n", " Successfully uninstalled importlib-resources-5.12.0\n", " Attempting uninstall: importlib-metadata\n", " Found existing installation: importlib-metadata 6.2.0\n", " Uninstalling importlib-metadata-6.2.0:\n", " Successfully uninstalled importlib-metadata-6.2.0\n", " Attempting uninstall: cmd2\n", " Found existing installation: cmd2 2.4.3\n", " Uninstalling cmd2-2.4.3:\n", " Successfully uninstalled cmd2-2.4.3\n", " Attempting uninstall: cmaes\n", " Found existing installation: cmaes 0.9.1\n", " Uninstalling cmaes-0.9.1:\n", " Successfully uninstalled cmaes-0.9.1\n", " Attempting uninstall: xgboost\n", " Found existing installation: xgboost 1.6.1\n", " Uninstalling xgboost-1.6.1:\n", " Successfully uninstalled xgboost-1.6.1\n", " Attempting uninstall: scikit-learn\n", " Found existing installation: scikit-learn 1.2.2\n", " Uninstalling scikit-learn-1.2.2:\n", " Successfully uninstalled scikit-learn-1.2.2\n", " Attempting uninstall: pandas\n", " Found existing installation: pandas 1.5.1\n", " Uninstalling pandas-1.5.1:\n", " Successfully uninstalled pandas-1.5.1\n", " Attempting uninstall: cliff\n", " Found existing installation: cliff 4.2.0\n", " Uninstalling cliff-4.2.0:\n", " Successfully uninstalled cliff-4.2.0\n", " Attempting uninstall: alembic\n", " Found existing installation: alembic 1.10.3\n", " Uninstalling alembic-1.10.3:\n", " Successfully uninstalled alembic-1.10.3\n", " Attempting uninstall: optuna\n", " Found existing installation: optuna 2.8.0\n", " Uninstalling optuna-2.8.0:\n", " Successfully uninstalled optuna-2.8.0\n", " Attempting uninstall: openml\n", " Found existing installation: openml 0.13.1\n", " Uninstalling openml-0.13.1:\n", " Successfully uninstalled openml-0.13.1\n", " Attempting uninstall: lightgbm\n", " Found existing installation: lightgbm 3.3.5\n", " Uninstalling lightgbm-3.3.5:\n", " Successfully uninstalled lightgbm-3.3.5\n", " Attempting uninstall: flaml\n", " Found existing installation: FLAML 1.1.3\n", " Uninstalling FLAML-1.1.3:\n", " Successfully uninstalled FLAML-1.1.3\n", "\u001b[31mERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.\n", "virtualenv 20.14.0 requires platformdirs<3,>=2, but you have platformdirs 3.2.0 which is incompatible.\n", "tensorflow 2.4.1 requires six~=1.15.0, but you have six 1.16.0 which is incompatible.\n", "tensorflow 2.4.1 requires typing-extensions~=3.7.4, but you have typing-extensions 4.5.0 which is incompatible.\n", "pmdarima 1.8.2 requires numpy~=1.19.0, but you have numpy 1.23.4 which is incompatible.\n", "koalas 1.8.0 requires numpy<1.20.0,>=1.14, but you have numpy 1.23.4 which is incompatible.\n", "gevent 21.1.2 requires greenlet<2.0,>=0.4.17; platform_python_implementation == \"CPython\", but you have greenlet 2.0.2 which is incompatible.\n", "azureml-dataset-runtime 1.34.0 requires pyarrow<4.0.0,>=0.17.0, but you have pyarrow 11.0.0 which is incompatible.\n", "azureml-core 1.34.0 requires urllib3<=1.26.6,>=1.23, but you have urllib3 1.26.15 which is incompatible.\u001b[0m\u001b[31m\n", "\u001b[0mSuccessfully installed Mako-1.2.4 MarkupSafe-2.1.2 PrettyTable-3.6.0 PyYAML-6.0 alembic-1.10.3 attrs-22.2.0 autopage-0.5.1 certifi-2022.12.7 charset-normalizer-3.1.0 cliff-4.2.0 cmaes-0.9.1 cmd2-2.4.3 colorlog-6.7.0 flaml-1.1.3 greenlet-2.0.2 idna-3.4 importlib-metadata-6.2.0 importlib-resources-5.12.0 joblib-1.2.0 joblibspark-0.5.1 liac-arff-2.5.0 lightgbm-3.3.5 minio-7.1.14 numpy-1.23.4 openml-0.13.1 optuna-2.8.0 packaging-23.0 pandas-1.5.1 pbr-5.11.1 py4j-0.10.9.5 pyarrow-11.0.0 pyperclip-1.8.2 pyspark-3.3.2 python-dateutil-2.8.2 pytz-2023.3 requests-2.28.2 scikit-learn-1.2.2 scipy-1.10.1 six-1.16.0 sqlalchemy-2.0.9 stevedore-5.0.0 threadpoolctl-3.1.0 tqdm-4.65.0 typing-extensions-4.5.0 urllib3-1.26.15 wcwidth-0.2.6 wheel-0.40.0 xgboost-1.6.1 xmltodict-0.13.0 zipp-3.15.0\n", "\u001b[33mWARNING: You are using pip version 22.0.4; however, version 23.0.1 is available.\n", "You should consider upgrading via the '/nfs4/pyenv-bfada21f-d1ed-44b9-a41d-4ff480d237e7/bin/python -m pip install --upgrade pip' command.\u001b[0m\u001b[33m\n", "\u001b[0mNote: you may need to restart the kernel to use updated packages.\n" ] }, { "data": {}, "execution_count": 39, "metadata": {}, "output_type": "execute_result" }, { "name": "stdout", "output_type": "stream", "text": [ "Warning: PySpark kernel has been restarted to use updated packages.\n", "\n" ] } ], "source": [ "%pip install flaml[synapse]==1.1.3 xgboost==1.6.1 pandas==1.5.1 numpy==1.23.4 openml --force-reinstall" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "## 2. Classification Example\n", "### Load data and preprocess\n", "\n", "Download [Airlines dataset](https://www.openml.org/d/1169) from OpenML. The task is to predict whether a given flight will be delayed, given the information of the scheduled departure." ] }, { "cell_type": "code", "execution_count": 41, "metadata": { "jupyter": { "outputs_hidden": true }, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "outputs": [ { "data": { "application/vnd.livy.statement-meta+json": { "execution_finish_time": "2023-04-09T03:11:11.6973622Z", "execution_start_time": "2023-04-09T03:11:09.4074274Z", "livy_statement_state": "available", "parent_msg_id": "25ba0152-0936-464b-83eb-afa5f2f517fb", "queued_time": "2023-04-09T03:10:33.8002088Z", "session_id": "7", "session_start_time": null, "spark_jobs": null, "spark_pool": "automl", "state": "finished", "statement_id": 67 }, "text/plain": [ "StatementMeta(automl, 7, 67, Finished, Available)" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stderr", "output_type": "stream", "text": [ "/home/trusted-service-user/cluster-env/env/lib/python3.8/site-packages/dask/dataframe/backends.py:187: FutureWarning: pandas.Int64Index is deprecated and will be removed from pandas in a future version. Use pandas.Index with the appropriate dtype instead.\n", " _numeric_index_types = (pd.Int64Index, pd.Float64Index, pd.UInt64Index)\n", "/home/trusted-service-user/cluster-env/env/lib/python3.8/site-packages/dask/dataframe/backends.py:187: FutureWarning: pandas.Float64Index is deprecated and will be removed from pandas in a future version. Use pandas.Index with the appropriate dtype instead.\n", " _numeric_index_types = (pd.Int64Index, pd.Float64Index, pd.UInt64Index)\n", "/home/trusted-service-user/cluster-env/env/lib/python3.8/site-packages/dask/dataframe/backends.py:187: FutureWarning: pandas.UInt64Index is deprecated and will be removed from pandas in a future version. Use pandas.Index with the appropriate dtype instead.\n", " _numeric_index_types = (pd.Int64Index, pd.Float64Index, pd.UInt64Index)\n" ] } ], "source": [ "from flaml.data import load_openml_dataset\n", "X_train, X_test, y_train, y_test = load_openml_dataset(dataset_id=1169, data_dir='./')" ] }, { "cell_type": "code", "execution_count": 42, "metadata": {}, "outputs": [ { "data": { "application/vnd.livy.statement-meta+json": { "execution_finish_time": "2023-04-09T03:11:12.2518637Z", "execution_start_time": "2023-04-09T03:11:11.9466307Z", "livy_statement_state": "available", "parent_msg_id": "c6f3064c-401e-447b-bd1d-65cd00f48fe1", "queued_time": "2023-04-09T03:10:33.901764Z", "session_id": "7", "session_start_time": null, "spark_jobs": null, "spark_pool": "automl", "state": "finished", "statement_id": 68 }, "text/plain": [ "StatementMeta(automl, 7, 68, Finished, Available)" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
AirlineFlightAirportFromAirportToDayOfWeekTimeLength
249392EV5309.0MDTATL3794.0131.0
166918CO1079.0IAHSAT5900.060.0
89110US1636.0CLECLT1530.0103.0
70258WN928.0CMHLAS7480.0280.0
492985WN729.0GEGLAS3630.0140.0
\n", "
" ], "text/plain": [ " Airline Flight AirportFrom AirportTo DayOfWeek Time Length\n", "249392 EV 5309.0 MDT ATL 3 794.0 131.0\n", "166918 CO 1079.0 IAH SAT 5 900.0 60.0\n", "89110 US 1636.0 CLE CLT 1 530.0 103.0\n", "70258 WN 928.0 CMH LAS 7 480.0 280.0\n", "492985 WN 729.0 GEG LAS 3 630.0 140.0" ] }, "execution_count": 9, "metadata": {}, "output_type": "execute_result" } ], "source": [ "X_train.head()" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "### Run FLAML\n", "In the FLAML automl run configuration, users can specify the task type, time budget, error metric, learner list, whether to subsample, resampling strategy type, and so on. All these arguments have default values which will be used if users do not provide them. For example, the default classifiers are `['lgbm', 'xgboost', 'xgb_limitdepth', 'catboost', 'rf', 'extra_tree', 'lrl1']`. " ] }, { "cell_type": "code", "execution_count": 43, "metadata": { "slideshow": { "slide_type": "slide" } }, "outputs": [ { "data": { "application/vnd.livy.statement-meta+json": { "execution_finish_time": "2023-04-09T03:11:12.8001867Z", "execution_start_time": "2023-04-09T03:11:12.5256701Z", "livy_statement_state": "available", "parent_msg_id": "f2fba5ab-4e87-41e8-8a76-b7b7367e6fc6", "queued_time": "2023-04-09T03:10:34.0855462Z", "session_id": "7", "session_start_time": null, "spark_jobs": null, "spark_pool": "automl", "state": "finished", "statement_id": 69 }, "text/plain": [ "StatementMeta(automl, 7, 69, Finished, Available)" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "''' import AutoML class from flaml package '''\n", "from flaml import AutoML\n", "automl = AutoML()" ] }, { "cell_type": "code", "execution_count": 44, "metadata": { "slideshow": { "slide_type": "slide" } }, "outputs": [ { "data": { "application/vnd.livy.statement-meta+json": { "execution_finish_time": "2023-04-09T03:11:13.391257Z", "execution_start_time": "2023-04-09T03:11:13.1109201Z", "livy_statement_state": "available", "parent_msg_id": "d5e4a7ed-3192-4e43-a7a8-44cf1469e685", "queued_time": "2023-04-09T03:10:34.3172166Z", "session_id": "7", "session_start_time": null, "spark_jobs": null, "spark_pool": "automl", "state": "finished", "statement_id": 70 }, "text/plain": [ "StatementMeta(automl, 7, 70, Finished, Available)" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "settings = {\n", " \"time_budget\": 120, # total running time in seconds\n", " \"metric\": 'accuracy', \n", " # check the documentation for options of metrics (https://microsoft.github.io/FLAML/docs/Use-Cases/Task-Oriented-AutoML#optimization-metric)\n", " \"task\": 'classification', # task type\n", " \"log_file_name\": 'airlines_experiment.log', # flaml log file\n", " \"seed\": 7654321, # random seed\n", "}\n" ] }, { "cell_type": "code", "execution_count": 45, "metadata": { "slideshow": { "slide_type": "slide" }, "tags": [ "outputPrepend" ] }, "outputs": [ { "data": { "application/vnd.livy.statement-meta+json": { "execution_finish_time": "2023-04-09T03:13:20.8381216Z", "execution_start_time": "2023-04-09T03:11:13.647266Z", "livy_statement_state": "available", "parent_msg_id": "29dd0ba0-8f0d-428b-acb9-1d8e62f1b157", "queued_time": "2023-04-09T03:10:34.4667686Z", "session_id": "7", "session_start_time": null, "spark_jobs": null, "spark_pool": "automl", "state": "finished", "statement_id": 71 }, "text/plain": [ "StatementMeta(automl, 7, 71, Finished, Available)" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "[flaml.automl.automl: 04-09 03:11:13] {2726} INFO - task = classification\n", "[flaml.automl.automl: 04-09 03:11:13] {2728} INFO - Data split method: stratified\n", "[flaml.automl.automl: 04-09 03:11:13] {2731} INFO - Evaluation method: holdout\n", "[flaml.automl.automl: 04-09 03:11:14] {2858} INFO - Minimizing error metric: 1-accuracy\n", "[flaml.automl.automl: 04-09 03:11:14] {3004} INFO - List of ML learners in AutoML Run: ['lgbm', 'rf', 'xgboost', 'extra_tree', 'xgb_limitdepth', 'lrl1']\n", "[flaml.automl.automl: 04-09 03:11:14] {3334} INFO - iteration 0, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:11:14] {3472} INFO - Estimated sufficient time budget=17413s. Estimated necessary time budget=401s.\n", "[flaml.automl.automl: 04-09 03:11:14] {3519} INFO - at 0.5s,\testimator lgbm's best error=0.3777,\tbest estimator lgbm's best error=0.3777\n", "[flaml.automl.automl: 04-09 03:11:14] {3334} INFO - iteration 1, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:11:14] {3519} INFO - at 0.5s,\testimator lgbm's best error=0.3777,\tbest estimator lgbm's best error=0.3777\n", "[flaml.automl.automl: 04-09 03:11:14] {3334} INFO - iteration 2, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:11:14] {3519} INFO - at 0.5s,\testimator lgbm's best error=0.3614,\tbest estimator lgbm's best error=0.3614\n", "[flaml.automl.automl: 04-09 03:11:14] {3334} INFO - iteration 3, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:11:14] {3519} INFO - at 0.6s,\testimator lgbm's best error=0.3614,\tbest estimator lgbm's best error=0.3614\n", "[flaml.automl.automl: 04-09 03:11:14] {3334} INFO - iteration 4, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:11:14] {3519} INFO - at 0.6s,\testimator lgbm's best error=0.3614,\tbest estimator lgbm's best error=0.3614\n", "[flaml.automl.automl: 04-09 03:11:14] {3334} INFO - iteration 5, current learner xgboost\n", "[flaml.automl.automl: 04-09 03:11:14] {3519} INFO - at 1.0s,\testimator xgboost's best error=0.3787,\tbest estimator lgbm's best error=0.3614\n", "[flaml.automl.automl: 04-09 03:11:14] {3334} INFO - iteration 6, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:11:14] {3519} INFO - at 1.0s,\testimator lgbm's best error=0.3614,\tbest estimator lgbm's best error=0.3614\n", "[flaml.automl.automl: 04-09 03:11:14] {3334} INFO - iteration 7, current learner extra_tree\n", "[flaml.automl.automl: 04-09 03:11:14] {3519} INFO - at 1.2s,\testimator extra_tree's best error=0.3787,\tbest estimator lgbm's best error=0.3614\n", "[flaml.automl.automl: 04-09 03:11:14] {3334} INFO - iteration 8, current learner rf\n", "[flaml.automl.automl: 04-09 03:11:14] {3519} INFO - at 1.2s,\testimator rf's best error=0.3816,\tbest estimator lgbm's best error=0.3614\n", "[flaml.automl.automl: 04-09 03:11:14] {3334} INFO - iteration 9, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:11:14] {3519} INFO - at 1.3s,\testimator lgbm's best error=0.3614,\tbest estimator lgbm's best error=0.3614\n", "[flaml.automl.automl: 04-09 03:11:14] {3334} INFO - iteration 10, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:11:14] {3519} INFO - at 1.3s,\testimator lgbm's best error=0.3614,\tbest estimator lgbm's best error=0.3614\n", "[flaml.automl.automl: 04-09 03:11:14] {3334} INFO - iteration 11, current learner rf\n", "[flaml.automl.automl: 04-09 03:11:15] {3519} INFO - at 1.5s,\testimator rf's best error=0.3791,\tbest estimator lgbm's best error=0.3614\n", "[flaml.automl.automl: 04-09 03:11:15] {3334} INFO - iteration 12, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:11:15] {3519} INFO - at 1.6s,\testimator lgbm's best error=0.3550,\tbest estimator lgbm's best error=0.3550\n", "[flaml.automl.automl: 04-09 03:11:15] {3334} INFO - iteration 13, current learner extra_tree\n", "[flaml.automl.automl: 04-09 03:11:15] {3519} INFO - at 1.7s,\testimator extra_tree's best error=0.3787,\tbest estimator lgbm's best error=0.3550\n", "[flaml.automl.automl: 04-09 03:11:15] {3334} INFO - iteration 14, current learner xgboost\n", "[flaml.automl.automl: 04-09 03:11:15] {3519} INFO - at 1.8s,\testimator xgboost's best error=0.3746,\tbest estimator lgbm's best error=0.3550\n", "[flaml.automl.automl: 04-09 03:11:15] {3334} INFO - iteration 15, current learner extra_tree\n", "[flaml.automl.automl: 04-09 03:11:15] {3519} INFO - at 1.9s,\testimator extra_tree's best error=0.3787,\tbest estimator lgbm's best error=0.3550\n", "[flaml.automl.automl: 04-09 03:11:15] {3334} INFO - iteration 16, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:11:15] {3519} INFO - at 1.9s,\testimator lgbm's best error=0.3550,\tbest estimator lgbm's best error=0.3550\n", "[flaml.automl.automl: 04-09 03:11:15] {3334} INFO - iteration 17, current learner xgboost\n", "[flaml.automl.automl: 04-09 03:11:15] {3519} INFO - at 2.2s,\testimator xgboost's best error=0.3699,\tbest estimator lgbm's best error=0.3550\n", "[flaml.automl.automl: 04-09 03:11:15] {3334} INFO - iteration 18, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:11:15] {3519} INFO - at 2.4s,\testimator lgbm's best error=0.3545,\tbest estimator lgbm's best error=0.3545\n", "[flaml.automl.automl: 04-09 03:11:15] {3334} INFO - iteration 19, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:11:16] {3519} INFO - at 2.5s,\testimator lgbm's best error=0.3545,\tbest estimator lgbm's best error=0.3545\n", "[flaml.automl.automl: 04-09 03:11:16] {3334} INFO - iteration 20, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:11:16] {3519} INFO - at 2.9s,\testimator lgbm's best error=0.3545,\tbest estimator lgbm's best error=0.3545\n", "[flaml.automl.automl: 04-09 03:11:16] {3334} INFO - iteration 21, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:11:16] {3519} INFO - at 3.0s,\testimator lgbm's best error=0.3536,\tbest estimator lgbm's best error=0.3536\n", "[flaml.automl.automl: 04-09 03:11:16] {3334} INFO - iteration 22, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:11:16] {3519} INFO - at 3.1s,\testimator lgbm's best error=0.3536,\tbest estimator lgbm's best error=0.3536\n", "[flaml.automl.automl: 04-09 03:11:16] {3334} INFO - iteration 23, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:11:17] {3519} INFO - at 3.4s,\testimator lgbm's best error=0.3536,\tbest estimator lgbm's best error=0.3536\n", "[flaml.automl.automl: 04-09 03:11:17] {3334} INFO - iteration 24, current learner rf\n", "[flaml.automl.automl: 04-09 03:11:17] {3519} INFO - at 3.6s,\testimator rf's best error=0.3791,\tbest estimator lgbm's best error=0.3536\n", "[flaml.automl.automl: 04-09 03:11:17] {3334} INFO - iteration 25, current learner xgboost\n", "[flaml.automl.automl: 04-09 03:11:17] {3519} INFO - at 3.9s,\testimator xgboost's best error=0.3596,\tbest estimator lgbm's best error=0.3536\n", "[flaml.automl.automl: 04-09 03:11:17] {3334} INFO - iteration 26, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:11:17] {3519} INFO - at 4.3s,\testimator lgbm's best error=0.3528,\tbest estimator lgbm's best error=0.3528\n", "[flaml.automl.automl: 04-09 03:11:17] {3334} INFO - iteration 27, current learner xgboost\n", "[flaml.automl.automl: 04-09 03:11:18] {3519} INFO - at 4.6s,\testimator xgboost's best error=0.3596,\tbest estimator lgbm's best error=0.3528\n", "[flaml.automl.automl: 04-09 03:11:18] {3334} INFO - iteration 28, current learner xgboost\n", "[flaml.automl.automl: 04-09 03:11:18] {3519} INFO - at 4.7s,\testimator xgboost's best error=0.3596,\tbest estimator lgbm's best error=0.3528\n", "[flaml.automl.automl: 04-09 03:11:18] {3334} INFO - iteration 29, current learner xgboost\n", "[flaml.automl.automl: 04-09 03:11:18] {3519} INFO - at 5.3s,\testimator xgboost's best error=0.3586,\tbest estimator lgbm's best error=0.3528\n", "[flaml.automl.automl: 04-09 03:11:18] {3334} INFO - iteration 30, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:11:20] {3519} INFO - at 6.5s,\testimator lgbm's best error=0.3405,\tbest estimator lgbm's best error=0.3405\n", "[flaml.automl.automl: 04-09 03:11:20] {3334} INFO - iteration 31, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:11:20] {3519} INFO - at 6.9s,\testimator lgbm's best error=0.3405,\tbest estimator lgbm's best error=0.3405\n", "[flaml.automl.automl: 04-09 03:11:20] {3334} INFO - iteration 32, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:11:21] {3519} INFO - at 8.1s,\testimator lgbm's best error=0.3370,\tbest estimator lgbm's best error=0.3370\n", "[flaml.automl.automl: 04-09 03:11:21] {3334} INFO - iteration 33, current learner rf\n", "[flaml.automl.automl: 04-09 03:11:21] {3519} INFO - at 8.2s,\testimator rf's best error=0.3791,\tbest estimator lgbm's best error=0.3370\n", "[flaml.automl.automl: 04-09 03:11:21] {3334} INFO - iteration 34, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:11:23] {3519} INFO - at 9.5s,\testimator lgbm's best error=0.3370,\tbest estimator lgbm's best error=0.3370\n", "[flaml.automl.automl: 04-09 03:11:23] {3334} INFO - iteration 35, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:11:24] {3519} INFO - at 10.5s,\testimator lgbm's best error=0.3370,\tbest estimator lgbm's best error=0.3370\n", "[flaml.automl.automl: 04-09 03:11:24] {3334} INFO - iteration 36, current learner xgboost\n", "[flaml.automl.automl: 04-09 03:11:24] {3519} INFO - at 11.0s,\testimator xgboost's best error=0.3577,\tbest estimator lgbm's best error=0.3370\n", "[flaml.automl.automl: 04-09 03:11:24] {3334} INFO - iteration 37, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:11:25] {3519} INFO - at 12.4s,\testimator lgbm's best error=0.3318,\tbest estimator lgbm's best error=0.3318\n", "[flaml.automl.automl: 04-09 03:11:25] {3334} INFO - iteration 38, current learner xgb_limitdepth\n", "[flaml.automl.automl: 04-09 03:11:26] {3519} INFO - at 12.6s,\testimator xgb_limitdepth's best error=0.3630,\tbest estimator lgbm's best error=0.3318\n", "[flaml.automl.automl: 04-09 03:11:26] {3334} INFO - iteration 39, current learner xgb_limitdepth\n", "[flaml.automl.automl: 04-09 03:11:26] {3519} INFO - at 12.7s,\testimator xgb_limitdepth's best error=0.3630,\tbest estimator lgbm's best error=0.3318\n", "[flaml.automl.automl: 04-09 03:11:26] {3334} INFO - iteration 40, current learner xgb_limitdepth\n", "[flaml.automl.automl: 04-09 03:11:26] {3519} INFO - at 13.1s,\testimator xgb_limitdepth's best error=0.3630,\tbest estimator lgbm's best error=0.3318\n", "[flaml.automl.automl: 04-09 03:11:26] {3334} INFO - iteration 41, current learner xgb_limitdepth\n", "[flaml.automl.automl: 04-09 03:11:26] {3519} INFO - at 13.3s,\testimator xgb_limitdepth's best error=0.3630,\tbest estimator lgbm's best error=0.3318\n", "[flaml.automl.automl: 04-09 03:11:26] {3334} INFO - iteration 42, current learner xgb_limitdepth\n", "[flaml.automl.automl: 04-09 03:11:26] {3519} INFO - at 13.4s,\testimator xgb_limitdepth's best error=0.3630,\tbest estimator lgbm's best error=0.3318\n", "[flaml.automl.automl: 04-09 03:11:26] {3334} INFO - iteration 43, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:11:28] {3519} INFO - at 14.8s,\testimator lgbm's best error=0.3282,\tbest estimator lgbm's best error=0.3282\n", "[flaml.automl.automl: 04-09 03:11:28] {3334} INFO - iteration 44, current learner xgb_limitdepth\n", "[flaml.automl.automl: 04-09 03:11:28] {3519} INFO - at 15.1s,\testimator xgb_limitdepth's best error=0.3630,\tbest estimator lgbm's best error=0.3282\n", "[flaml.automl.automl: 04-09 03:11:28] {3334} INFO - iteration 45, current learner xgb_limitdepth\n", "[flaml.automl.automl: 04-09 03:11:28] {3519} INFO - at 15.2s,\testimator xgb_limitdepth's best error=0.3623,\tbest estimator lgbm's best error=0.3282\n", "[flaml.automl.automl: 04-09 03:11:28] {3334} INFO - iteration 46, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:11:30] {3519} INFO - at 16.6s,\testimator lgbm's best error=0.3282,\tbest estimator lgbm's best error=0.3282\n", "[flaml.automl.automl: 04-09 03:11:30] {3334} INFO - iteration 47, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:11:31] {3519} INFO - at 18.0s,\testimator lgbm's best error=0.3282,\tbest estimator lgbm's best error=0.3282\n", "[flaml.automl.automl: 04-09 03:11:31] {3334} INFO - iteration 48, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:11:33] {3519} INFO - at 20.3s,\testimator lgbm's best error=0.3282,\tbest estimator lgbm's best error=0.3282\n", "[flaml.automl.automl: 04-09 03:11:33] {3334} INFO - iteration 49, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:11:35] {3519} INFO - at 22.2s,\testimator lgbm's best error=0.3282,\tbest estimator lgbm's best error=0.3282\n", "[flaml.automl.automl: 04-09 03:11:35] {3334} INFO - iteration 50, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:11:37] {3519} INFO - at 23.6s,\testimator lgbm's best error=0.3282,\tbest estimator lgbm's best error=0.3282\n", "[flaml.automl.automl: 04-09 03:11:37] {3334} INFO - iteration 51, current learner lrl1\n", "No low-cost partial config given to the search algorithm. For cost-frugal search, consider providing low-cost values for cost-related hps via 'low_cost_partial_config'. More info can be found at https://microsoft.github.io/FLAML/docs/FAQ#about-low_cost_partial_config-in-tune\n", "[flaml.automl.automl: 04-09 03:11:37] {3519} INFO - at 23.8s,\testimator lrl1's best error=0.4339,\tbest estimator lgbm's best error=0.3282\n", "[flaml.automl.automl: 04-09 03:11:37] {3334} INFO - iteration 52, current learner lrl1\n", "[flaml.automl.automl: 04-09 03:11:37] {3519} INFO - at 24.0s,\testimator lrl1's best error=0.4339,\tbest estimator lgbm's best error=0.3282\n", "[flaml.automl.automl: 04-09 03:11:37] {3334} INFO - iteration 53, current learner lrl1\n", "[flaml.automl.automl: 04-09 03:11:37] {3519} INFO - at 24.2s,\testimator lrl1's best error=0.4339,\tbest estimator lgbm's best error=0.3282\n", "[flaml.automl.automl: 04-09 03:11:37] {3334} INFO - iteration 54, current learner lrl1\n", "[flaml.automl.automl: 04-09 03:11:38] {3519} INFO - at 25.0s,\testimator lrl1's best error=0.4334,\tbest estimator lgbm's best error=0.3282\n", "[flaml.automl.automl: 04-09 03:11:38] {3334} INFO - iteration 55, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:11:39] {3519} INFO - at 26.3s,\testimator lgbm's best error=0.3282,\tbest estimator lgbm's best error=0.3282\n", "[flaml.automl.automl: 04-09 03:11:39] {3334} INFO - iteration 56, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:11:42] {3519} INFO - at 28.7s,\testimator lgbm's best error=0.3282,\tbest estimator lgbm's best error=0.3282\n", "[flaml.automl.automl: 04-09 03:11:42] {3334} INFO - iteration 57, current learner rf\n", "[flaml.automl.automl: 04-09 03:11:42] {3519} INFO - at 28.9s,\testimator rf's best error=0.3789,\tbest estimator lgbm's best error=0.3282\n", "[flaml.automl.automl: 04-09 03:11:42] {3334} INFO - iteration 58, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:11:48] {3519} INFO - at 35.0s,\testimator lgbm's best error=0.3282,\tbest estimator lgbm's best error=0.3282\n", "[flaml.automl.automl: 04-09 03:11:48] {3334} INFO - iteration 59, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:11:49] {3519} INFO - at 35.6s,\testimator lgbm's best error=0.3282,\tbest estimator lgbm's best error=0.3282\n", "[flaml.automl.automl: 04-09 03:11:49] {3334} INFO - iteration 60, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:12:01] {3519} INFO - at 47.9s,\testimator lgbm's best error=0.3282,\tbest estimator lgbm's best error=0.3282\n", "[flaml.automl.automl: 04-09 03:12:01] {3334} INFO - iteration 61, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:12:01] {3519} INFO - at 48.3s,\testimator lgbm's best error=0.3282,\tbest estimator lgbm's best error=0.3282\n", "[flaml.automl.automl: 04-09 03:12:01] {3334} INFO - iteration 62, current learner xgboost\n", "[flaml.automl.automl: 04-09 03:12:02] {3519} INFO - at 49.1s,\testimator xgboost's best error=0.3561,\tbest estimator lgbm's best error=0.3282\n", "[flaml.automl.automl: 04-09 03:12:02] {3334} INFO - iteration 63, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:12:04] {3519} INFO - at 51.3s,\testimator lgbm's best error=0.3282,\tbest estimator lgbm's best error=0.3282\n", "[flaml.automl.automl: 04-09 03:12:04] {3334} INFO - iteration 64, current learner xgboost\n", "[flaml.automl.automl: 04-09 03:12:05] {3519} INFO - at 52.0s,\testimator xgboost's best error=0.3561,\tbest estimator lgbm's best error=0.3282\n", "[flaml.automl.automl: 04-09 03:12:05] {3334} INFO - iteration 65, current learner xgboost\n", "[flaml.automl.automl: 04-09 03:12:06] {3519} INFO - at 53.0s,\testimator xgboost's best error=0.3561,\tbest estimator lgbm's best error=0.3282\n", "[flaml.automl.automl: 04-09 03:12:06] {3334} INFO - iteration 66, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:12:07] {3519} INFO - at 54.2s,\testimator lgbm's best error=0.3282,\tbest estimator lgbm's best error=0.3282\n", "[flaml.automl.automl: 04-09 03:12:07] {3334} INFO - iteration 67, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:12:09] {3519} INFO - at 55.9s,\testimator lgbm's best error=0.3274,\tbest estimator lgbm's best error=0.3274\n", "[flaml.automl.automl: 04-09 03:12:09] {3334} INFO - iteration 68, current learner xgboost\n", "[flaml.automl.automl: 04-09 03:12:10] {3519} INFO - at 56.9s,\testimator xgboost's best error=0.3561,\tbest estimator lgbm's best error=0.3274\n", "[flaml.automl.automl: 04-09 03:12:10] {3334} INFO - iteration 69, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:12:11] {3519} INFO - at 58.3s,\testimator lgbm's best error=0.3274,\tbest estimator lgbm's best error=0.3274\n", "[flaml.automl.automl: 04-09 03:12:11] {3334} INFO - iteration 70, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:12:12] {3519} INFO - at 59.2s,\testimator lgbm's best error=0.3274,\tbest estimator lgbm's best error=0.3274\n", "[flaml.automl.automl: 04-09 03:12:12] {3334} INFO - iteration 71, current learner rf\n", "[flaml.automl.automl: 04-09 03:12:12] {3519} INFO - at 59.4s,\testimator rf's best error=0.3781,\tbest estimator lgbm's best error=0.3274\n", "[flaml.automl.automl: 04-09 03:12:12] {3334} INFO - iteration 72, current learner rf\n", "[flaml.automl.automl: 04-09 03:12:13] {3519} INFO - at 59.4s,\testimator rf's best error=0.3781,\tbest estimator lgbm's best error=0.3274\n", "[flaml.automl.automl: 04-09 03:12:13] {3334} INFO - iteration 73, current learner rf\n", "[flaml.automl.automl: 04-09 03:12:13] {3519} INFO - at 59.5s,\testimator rf's best error=0.3725,\tbest estimator lgbm's best error=0.3274\n", "[flaml.automl.automl: 04-09 03:12:13] {3334} INFO - iteration 74, current learner rf\n", "[flaml.automl.automl: 04-09 03:12:13] {3519} INFO - at 59.6s,\testimator rf's best error=0.3725,\tbest estimator lgbm's best error=0.3274\n", "[flaml.automl.automl: 04-09 03:12:13] {3334} INFO - iteration 75, current learner rf\n", "[flaml.automl.automl: 04-09 03:12:13] {3519} INFO - at 59.7s,\testimator rf's best error=0.3725,\tbest estimator lgbm's best error=0.3274\n", "[flaml.automl.automl: 04-09 03:12:13] {3334} INFO - iteration 76, current learner rf\n", "[flaml.automl.automl: 04-09 03:12:13] {3519} INFO - at 59.7s,\testimator rf's best error=0.3706,\tbest estimator lgbm's best error=0.3274\n", "[flaml.automl.automl: 04-09 03:12:13] {3334} INFO - iteration 77, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:12:18] {3519} INFO - at 65.4s,\testimator lgbm's best error=0.3268,\tbest estimator lgbm's best error=0.3268\n", "[flaml.automl.automl: 04-09 03:12:18] {3334} INFO - iteration 78, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:12:21] {3519} INFO - at 68.1s,\testimator lgbm's best error=0.3268,\tbest estimator lgbm's best error=0.3268\n", "[flaml.automl.automl: 04-09 03:12:21] {3334} INFO - iteration 79, current learner rf\n", "[flaml.automl.automl: 04-09 03:12:21] {3519} INFO - at 68.3s,\testimator rf's best error=0.3706,\tbest estimator lgbm's best error=0.3268\n", "[flaml.automl.automl: 04-09 03:12:21] {3334} INFO - iteration 80, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:12:27] {3519} INFO - at 74.4s,\testimator lgbm's best error=0.3250,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:12:27] {3334} INFO - iteration 81, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:12:30] {3519} INFO - at 77.0s,\testimator lgbm's best error=0.3250,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:12:30] {3334} INFO - iteration 82, current learner extra_tree\n", "[flaml.automl.automl: 04-09 03:12:30] {3519} INFO - at 77.2s,\testimator extra_tree's best error=0.3787,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:12:30] {3334} INFO - iteration 83, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:12:50] {3519} INFO - at 96.7s,\testimator lgbm's best error=0.3250,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:12:50] {3334} INFO - iteration 84, current learner rf\n", "[flaml.automl.automl: 04-09 03:12:50] {3519} INFO - at 96.8s,\testimator rf's best error=0.3706,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:12:50] {3334} INFO - iteration 85, current learner rf\n", "[flaml.automl.automl: 04-09 03:12:50] {3519} INFO - at 97.0s,\testimator rf's best error=0.3678,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:12:50] {3334} INFO - iteration 86, current learner xgboost\n", "[flaml.automl.automl: 04-09 03:12:50] {3519} INFO - at 97.3s,\testimator xgboost's best error=0.3561,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:12:50] {3334} INFO - iteration 87, current learner rf\n", "[flaml.automl.automl: 04-09 03:12:51] {3519} INFO - at 97.4s,\testimator rf's best error=0.3678,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:12:51] {3334} INFO - iteration 88, current learner rf\n", "[flaml.automl.automl: 04-09 03:12:51] {3519} INFO - at 97.5s,\testimator rf's best error=0.3666,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:12:51] {3334} INFO - iteration 89, current learner rf\n", "[flaml.automl.automl: 04-09 03:12:51] {3519} INFO - at 97.7s,\testimator rf's best error=0.3645,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:12:51] {3334} INFO - iteration 90, current learner rf\n", "[flaml.automl.automl: 04-09 03:12:51] {3519} INFO - at 97.8s,\testimator rf's best error=0.3645,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:12:51] {3334} INFO - iteration 91, current learner xgboost\n", "[flaml.automl.automl: 04-09 03:12:51] {3519} INFO - at 98.2s,\testimator xgboost's best error=0.3561,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:12:51] {3334} INFO - iteration 92, current learner rf\n", "[flaml.automl.automl: 04-09 03:12:51] {3519} INFO - at 98.3s,\testimator rf's best error=0.3645,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:12:51] {3334} INFO - iteration 93, current learner xgb_limitdepth\n", "[flaml.automl.automl: 04-09 03:12:51] {3519} INFO - at 98.3s,\testimator xgb_limitdepth's best error=0.3612,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:12:51] {3334} INFO - iteration 94, current learner xgb_limitdepth\n", "[flaml.automl.automl: 04-09 03:12:52] {3519} INFO - at 98.5s,\testimator xgb_limitdepth's best error=0.3612,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:12:52] {3334} INFO - iteration 95, current learner xgboost\n", "[flaml.automl.automl: 04-09 03:12:52] {3519} INFO - at 98.8s,\testimator xgboost's best error=0.3561,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:12:52] {3334} INFO - iteration 96, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:12:58] {3519} INFO - at 105.1s,\testimator lgbm's best error=0.3250,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:12:58] {3334} INFO - iteration 97, current learner xgb_limitdepth\n", "[flaml.automl.automl: 04-09 03:12:58] {3519} INFO - at 105.3s,\testimator xgb_limitdepth's best error=0.3612,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:12:58] {3334} INFO - iteration 98, current learner rf\n", "[flaml.automl.automl: 04-09 03:12:59] {3519} INFO - at 105.5s,\testimator rf's best error=0.3560,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:12:59] {3334} INFO - iteration 99, current learner rf\n", "[flaml.automl.automl: 04-09 03:12:59] {3519} INFO - at 105.7s,\testimator rf's best error=0.3560,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:12:59] {3334} INFO - iteration 100, current learner rf\n", "[flaml.automl.automl: 04-09 03:12:59] {3519} INFO - at 106.0s,\testimator rf's best error=0.3560,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:12:59] {3334} INFO - iteration 101, current learner rf\n", "[flaml.automl.automl: 04-09 03:12:59] {3519} INFO - at 106.3s,\testimator rf's best error=0.3560,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:12:59] {3334} INFO - iteration 102, current learner xgb_limitdepth\n", "[flaml.automl.automl: 04-09 03:12:59] {3519} INFO - at 106.4s,\testimator xgb_limitdepth's best error=0.3604,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:12:59] {3334} INFO - iteration 103, current learner rf\n", "[flaml.automl.automl: 04-09 03:13:00] {3519} INFO - at 106.7s,\testimator rf's best error=0.3547,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:13:00] {3334} INFO - iteration 104, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:13:06] {3519} INFO - at 113.1s,\testimator lgbm's best error=0.3250,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:13:06] {3334} INFO - iteration 105, current learner xgboost\n", "[flaml.automl.automl: 04-09 03:13:07] {3519} INFO - at 113.5s,\testimator xgboost's best error=0.3561,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:13:07] {3334} INFO - iteration 106, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:13:09] {3519} INFO - at 116.2s,\testimator lgbm's best error=0.3250,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:13:09] {3334} INFO - iteration 107, current learner xgb_limitdepth\n", "[flaml.automl.automl: 04-09 03:13:10] {3519} INFO - at 116.4s,\testimator xgb_limitdepth's best error=0.3604,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:13:10] {3334} INFO - iteration 108, current learner xgb_limitdepth\n", "[flaml.automl.automl: 04-09 03:13:10] {3519} INFO - at 116.5s,\testimator xgb_limitdepth's best error=0.3584,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:13:10] {3334} INFO - iteration 109, current learner xgb_limitdepth\n", "[flaml.automl.automl: 04-09 03:13:10] {3519} INFO - at 116.6s,\testimator xgb_limitdepth's best error=0.3584,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:13:10] {3334} INFO - iteration 110, current learner xgb_limitdepth\n", "[flaml.automl.automl: 04-09 03:13:10] {3519} INFO - at 116.8s,\testimator xgb_limitdepth's best error=0.3575,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:13:10] {3334} INFO - iteration 111, current learner xgb_limitdepth\n", "[flaml.automl.automl: 04-09 03:13:10] {3519} INFO - at 116.9s,\testimator xgb_limitdepth's best error=0.3575,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:13:10] {3334} INFO - iteration 112, current learner rf\n", "[flaml.automl.automl: 04-09 03:13:10] {3519} INFO - at 117.1s,\testimator rf's best error=0.3547,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:13:10] {3334} INFO - iteration 113, current learner xgb_limitdepth\n", "[flaml.automl.automl: 04-09 03:13:10] {3519} INFO - at 117.2s,\testimator xgb_limitdepth's best error=0.3575,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:13:10] {3334} INFO - iteration 114, current learner xgb_limitdepth\n", "[flaml.automl.automl: 04-09 03:13:10] {3519} INFO - at 117.3s,\testimator xgb_limitdepth's best error=0.3575,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:13:10] {3334} INFO - iteration 115, current learner lrl1\n", "[flaml.automl.automl: 04-09 03:13:11] {3519} INFO - at 118.0s,\testimator lrl1's best error=0.4334,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:13:11] {3334} INFO - iteration 116, current learner rf\n", "[flaml.automl.automl: 04-09 03:13:11] {3519} INFO - at 118.1s,\testimator rf's best error=0.3547,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:13:11] {3334} INFO - iteration 117, current learner rf\n", "[flaml.automl.automl: 04-09 03:13:11] {3519} INFO - at 118.3s,\testimator rf's best error=0.3547,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:13:11] {3334} INFO - iteration 118, current learner xgb_limitdepth\n", "[flaml.automl.automl: 04-09 03:13:11] {3519} INFO - at 118.4s,\testimator xgb_limitdepth's best error=0.3575,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:13:11] {3334} INFO - iteration 119, current learner xgb_limitdepth\n", "[flaml.automl.automl: 04-09 03:13:12] {3519} INFO - at 118.5s,\testimator xgb_limitdepth's best error=0.3575,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:13:12] {3334} INFO - iteration 120, current learner rf\n", "[flaml.automl.automl: 04-09 03:13:12] {3519} INFO - at 118.6s,\testimator rf's best error=0.3547,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:13:12] {3334} INFO - iteration 121, current learner xgb_limitdepth\n", "[flaml.automl.automl: 04-09 03:13:12] {3519} INFO - at 119.2s,\testimator xgb_limitdepth's best error=0.3520,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:13:12] {3334} INFO - iteration 122, current learner xgb_limitdepth\n", "[flaml.automl.automl: 04-09 03:13:13] {3519} INFO - at 119.8s,\testimator xgb_limitdepth's best error=0.3481,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:13:13] {3334} INFO - iteration 123, current learner extra_tree\n", "[flaml.automl.automl: 04-09 03:13:13] {3519} INFO - at 119.8s,\testimator extra_tree's best error=0.3787,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:13:13] {3334} INFO - iteration 124, current learner extra_tree\n", "[flaml.automl.automl: 04-09 03:13:13] {3519} INFO - at 119.8s,\testimator extra_tree's best error=0.3787,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:13:13] {3334} INFO - iteration 125, current learner extra_tree\n", "[flaml.automl.automl: 04-09 03:13:13] {3519} INFO - at 119.9s,\testimator extra_tree's best error=0.3787,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:13:13] {3334} INFO - iteration 126, current learner extra_tree\n", "[flaml.automl.automl: 04-09 03:13:13] {3519} INFO - at 119.9s,\testimator extra_tree's best error=0.3787,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:13:13] {3334} INFO - iteration 127, current learner extra_tree\n", "[flaml.automl.automl: 04-09 03:13:13] {3519} INFO - at 119.9s,\testimator extra_tree's best error=0.3787,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:13:13] {3334} INFO - iteration 128, current learner extra_tree\n", "[flaml.automl.automl: 04-09 03:13:13] {3519} INFO - at 119.9s,\testimator extra_tree's best error=0.3787,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:13:13] {3334} INFO - iteration 129, current learner extra_tree\n", "[flaml.automl.automl: 04-09 03:13:13] {3519} INFO - at 120.0s,\testimator extra_tree's best error=0.3787,\tbest estimator lgbm's best error=0.3250\n", "[flaml.automl.automl: 04-09 03:13:19] {3783} INFO - retrain lgbm for 5.8s\n", "[flaml.automl.automl: 04-09 03:13:19] {3790} INFO - retrained model: LGBMClassifier(colsample_bytree=0.763983850698587,\n", " learning_rate=0.087493667994037, max_bin=127,\n", " min_child_samples=128, n_estimators=302, num_leaves=466,\n", " reg_alpha=0.09968008477303378, reg_lambda=23.227419343318914,\n", " verbose=-1)\n", "[flaml.automl.automl: 04-09 03:13:19] {3034} INFO - fit succeeded\n", "[flaml.automl.automl: 04-09 03:13:19] {3035} INFO - Time taken to find the best model: 74.35051536560059\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ "/nfs4/pyenv-bfada21f-d1ed-44b9-a41d-4ff480d237e7/lib/python3.8/site-packages/sklearn/linear_model/_sag.py:350: ConvergenceWarning: The max_iter was reached which means the coef_ did not converge\n", " warnings.warn(\n", "/nfs4/pyenv-bfada21f-d1ed-44b9-a41d-4ff480d237e7/lib/python3.8/site-packages/sklearn/linear_model/_sag.py:350: ConvergenceWarning: The max_iter was reached which means the coef_ did not converge\n", " warnings.warn(\n", "/nfs4/pyenv-bfada21f-d1ed-44b9-a41d-4ff480d237e7/lib/python3.8/site-packages/sklearn/linear_model/_sag.py:350: ConvergenceWarning: The max_iter was reached which means the coef_ did not converge\n", " warnings.warn(\n", "/nfs4/pyenv-bfada21f-d1ed-44b9-a41d-4ff480d237e7/lib/python3.8/site-packages/sklearn/linear_model/_sag.py:350: ConvergenceWarning: The max_iter was reached which means the coef_ did not converge\n", " warnings.warn(\n", "/nfs4/pyenv-bfada21f-d1ed-44b9-a41d-4ff480d237e7/lib/python3.8/site-packages/sklearn/linear_model/_sag.py:350: ConvergenceWarning: The max_iter was reached which means the coef_ did not converge\n", " warnings.warn(\n" ] } ], "source": [ "'''The main flaml automl API'''\n", "automl.fit(X_train=X_train, y_train=y_train, **settings)" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "### Best model and metric" ] }, { "cell_type": "code", "execution_count": 46, "metadata": { "slideshow": { "slide_type": "slide" }, "tags": [] }, "outputs": [ { "data": { "application/vnd.livy.statement-meta+json": { "execution_finish_time": "2023-04-09T03:13:21.4301236Z", "execution_start_time": "2023-04-09T03:13:21.0903825Z", "livy_statement_state": "available", "parent_msg_id": "7d9a796c-9ca5-415d-9dab-de06e4170216", "queued_time": "2023-04-09T03:10:34.5888418Z", "session_id": "7", "session_start_time": null, "spark_jobs": null, "spark_pool": "automl", "state": "finished", "statement_id": 72 }, "text/plain": [ "StatementMeta(automl, 7, 72, Finished, Available)" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "Best ML leaner: lgbm\n", "Best hyperparmeter config: {'n_estimators': 302, 'num_leaves': 466, 'min_child_samples': 128, 'learning_rate': 0.087493667994037, 'log_max_bin': 7, 'colsample_bytree': 0.763983850698587, 'reg_alpha': 0.09968008477303378, 'reg_lambda': 23.227419343318914}\n", "Best accuracy on validation data: 0.675\n", "Training duration of best run: 5.756 s\n" ] } ], "source": [ "'''retrieve best config and best learner'''\n", "print('Best ML leaner:', automl.best_estimator)\n", "print('Best hyperparmeter config:', automl.best_config)\n", "print('Best accuracy on validation data: {0:.4g}'.format(1-automl.best_loss))\n", "print('Training duration of best run: {0:.4g} s'.format(automl.best_config_train_time))" ] }, { "cell_type": "code", "execution_count": 47, "metadata": { "slideshow": { "slide_type": "slide" } }, "outputs": [ { "data": { "application/vnd.livy.statement-meta+json": { "execution_finish_time": "2023-04-09T03:13:22.00515Z", "execution_start_time": "2023-04-09T03:13:21.668468Z", "livy_statement_state": "available", "parent_msg_id": "69be3bb6-08bb-40d8-bfbd-bfd3eabd2abf", "queued_time": "2023-04-09T03:10:34.6939373Z", "session_id": "7", "session_start_time": null, "spark_jobs": null, "spark_pool": "automl", "state": "finished", "statement_id": 73 }, "text/plain": [ "StatementMeta(automl, 7, 73, Finished, Available)" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "
LGBMClassifier(colsample_bytree=0.763983850698587,\n",
              "               learning_rate=0.087493667994037, max_bin=127,\n",
              "               min_child_samples=128, n_estimators=302, num_leaves=466,\n",
              "               reg_alpha=0.09968008477303378, reg_lambda=23.227419343318914,\n",
              "               verbose=-1)
In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.
" ], "text/plain": [ "LGBMClassifier(colsample_bytree=0.763983850698587,\n", " learning_rate=0.087493667994037, max_bin=127,\n", " min_child_samples=128, n_estimators=302, num_leaves=466,\n", " reg_alpha=0.09968008477303378, reg_lambda=23.227419343318914,\n", " verbose=-1)" ] }, "execution_count": 19, "metadata": {}, "output_type": "execute_result" } ], "source": [ "automl.model.estimator" ] }, { "cell_type": "code", "execution_count": 48, "metadata": { "slideshow": { "slide_type": "slide" } }, "outputs": [ { "data": { "application/vnd.livy.statement-meta+json": { "execution_finish_time": "2023-04-09T03:13:22.565239Z", "execution_start_time": "2023-04-09T03:13:22.2540989Z", "livy_statement_state": "available", "parent_msg_id": "75ef8b8e-a50b-4f56-9d25-5fc985379c27", "queued_time": "2023-04-09T03:10:34.7945603Z", "session_id": "7", "session_start_time": null, "spark_jobs": null, "spark_pool": "automl", "state": "finished", "statement_id": 74 }, "text/plain": [ "StatementMeta(automl, 7, 74, Finished, Available)" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "'''pickle and save the automl object'''\n", "import pickle\n", "with open('automl.pkl', 'wb') as f:\n", " pickle.dump(automl, f, pickle.HIGHEST_PROTOCOL)\n", "'''load pickled automl object'''\n", "with open('automl.pkl', 'rb') as f:\n", " automl = pickle.load(f)" ] }, { "cell_type": "code", "execution_count": 49, "metadata": { "slideshow": { "slide_type": "slide" }, "tags": [] }, "outputs": [ { "data": { "application/vnd.livy.statement-meta+json": { "execution_finish_time": "2023-04-09T03:13:25.1592289Z", "execution_start_time": "2023-04-09T03:13:22.8210504Z", "livy_statement_state": "available", "parent_msg_id": "32c71506-0598-4e00-aea9-cb84387ecc5b", "queued_time": "2023-04-09T03:10:34.9144997Z", "session_id": "7", "session_start_time": null, "spark_jobs": null, "spark_pool": "automl", "state": "finished", "statement_id": 75 }, "text/plain": [ "StatementMeta(automl, 7, 75, Finished, Available)" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "Predicted labels ['1' '0' '1' ... '1' '0' '0']\n", "True labels 118331 0\n", "328182 0\n", "335454 0\n", "520591 1\n", "344651 0\n", " ..\n", "367080 0\n", "203510 1\n", "254894 0\n", "296512 1\n", "362444 0\n", "Name: Delay, Length: 134846, dtype: category\n", "Categories (2, object): ['0' < '1']\n" ] } ], "source": [ "'''compute predictions of testing dataset''' \n", "y_pred = automl.predict(X_test)\n", "print('Predicted labels', y_pred)\n", "print('True labels', y_test)\n", "y_pred_proba = automl.predict_proba(X_test)[:,1]" ] }, { "cell_type": "code", "execution_count": 50, "metadata": { "slideshow": { "slide_type": "slide" }, "tags": [] }, "outputs": [ { "data": { "application/vnd.livy.statement-meta+json": { "execution_finish_time": "2023-04-09T03:13:26.1850094Z", "execution_start_time": "2023-04-09T03:13:25.4270376Z", "livy_statement_state": "available", "parent_msg_id": "5c1b0a67-28a7-4155-84e2-e732fb48b37d", "queued_time": "2023-04-09T03:10:35.0461186Z", "session_id": "7", "session_start_time": null, "spark_jobs": null, "spark_pool": "automl", "state": "finished", "statement_id": 76 }, "text/plain": [ "StatementMeta(automl, 7, 76, Finished, Available)" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "accuracy = 0.6732939797991784\n", "roc_auc = 0.7276250346550404\n", "log_loss = 0.6014655432027879\n" ] } ], "source": [ "''' compute different metric values on testing dataset'''\n", "from flaml.ml import sklearn_metric_loss_score\n", "print('accuracy', '=', 1 - sklearn_metric_loss_score('accuracy', y_pred, y_test))\n", "print('roc_auc', '=', 1 - sklearn_metric_loss_score('roc_auc', y_pred_proba, y_test))\n", "print('log_loss', '=', sklearn_metric_loss_score('log_loss', y_pred_proba, y_test))" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "See Section 4 for an accuracy comparison with default LightGBM and XGBoost.\n", "\n", "### Log history" ] }, { "cell_type": "code", "execution_count": 51, "metadata": { "slideshow": { "slide_type": "subslide" }, "tags": [] }, "outputs": [ { "data": { "application/vnd.livy.statement-meta+json": { "execution_finish_time": "2023-04-09T03:13:26.7290827Z", "execution_start_time": "2023-04-09T03:13:26.4652129Z", "livy_statement_state": "available", "parent_msg_id": "74e2927e-2fe9-4956-9e67-1246b2b24c66", "queued_time": "2023-04-09T03:10:35.1554934Z", "session_id": "7", "session_start_time": null, "spark_jobs": null, "spark_pool": "automl", "state": "finished", "statement_id": 77 }, "text/plain": [ "StatementMeta(automl, 7, 77, Finished, Available)" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "{'Current Learner': 'lgbm', 'Current Sample': 10000, 'Current Hyper-parameters': {'n_estimators': 4, 'num_leaves': 4, 'min_child_samples': 20, 'learning_rate': 0.09999999999999995, 'log_max_bin': 8, 'colsample_bytree': 1.0, 'reg_alpha': 0.0009765625, 'reg_lambda': 1.0, 'FLAML_sample_size': 10000}, 'Best Learner': 'lgbm', 'Best Hyper-parameters': {'n_estimators': 4, 'num_leaves': 4, 'min_child_samples': 20, 'learning_rate': 0.09999999999999995, 'log_max_bin': 8, 'colsample_bytree': 1.0, 'reg_alpha': 0.0009765625, 'reg_lambda': 1.0, 'FLAML_sample_size': 10000}}\n", "{'Current Learner': 'lgbm', 'Current Sample': 10000, 'Current Hyper-parameters': {'n_estimators': 26, 'num_leaves': 4, 'min_child_samples': 18, 'learning_rate': 0.2293009676418639, 'log_max_bin': 9, 'colsample_bytree': 0.9086551727646448, 'reg_alpha': 0.0015561782752413472, 'reg_lambda': 0.33127416269768944, 'FLAML_sample_size': 10000}, 'Best Learner': 'lgbm', 'Best Hyper-parameters': {'n_estimators': 26, 'num_leaves': 4, 'min_child_samples': 18, 'learning_rate': 0.2293009676418639, 'log_max_bin': 9, 'colsample_bytree': 0.9086551727646448, 'reg_alpha': 0.0015561782752413472, 'reg_lambda': 0.33127416269768944, 'FLAML_sample_size': 10000}}\n", "{'Current Learner': 'lgbm', 'Current Sample': 40000, 'Current Hyper-parameters': {'n_estimators': 55, 'num_leaves': 4, 'min_child_samples': 20, 'learning_rate': 0.43653962213332903, 'log_max_bin': 10, 'colsample_bytree': 0.8048558760626646, 'reg_alpha': 0.0009765625, 'reg_lambda': 0.23010605579846408, 'FLAML_sample_size': 40000}, 'Best Learner': 'lgbm', 'Best Hyper-parameters': {'n_estimators': 55, 'num_leaves': 4, 'min_child_samples': 20, 'learning_rate': 0.43653962213332903, 'log_max_bin': 10, 'colsample_bytree': 0.8048558760626646, 'reg_alpha': 0.0009765625, 'reg_lambda': 0.23010605579846408, 'FLAML_sample_size': 40000}}\n", "{'Current Learner': 'lgbm', 'Current Sample': 40000, 'Current Hyper-parameters': {'n_estimators': 90, 'num_leaves': 18, 'min_child_samples': 34, 'learning_rate': 0.3572626620529719, 'log_max_bin': 10, 'colsample_bytree': 0.9295656128173544, 'reg_alpha': 0.0009765625, 'reg_lambda': 0.1981463604305675, 'FLAML_sample_size': 40000}, 'Best Learner': 'lgbm', 'Best Hyper-parameters': {'n_estimators': 90, 'num_leaves': 18, 'min_child_samples': 34, 'learning_rate': 0.3572626620529719, 'log_max_bin': 10, 'colsample_bytree': 0.9295656128173544, 'reg_alpha': 0.0009765625, 'reg_lambda': 0.1981463604305675, 'FLAML_sample_size': 40000}}\n", "{'Current Learner': 'lgbm', 'Current Sample': 40000, 'Current Hyper-parameters': {'n_estimators': 56, 'num_leaves': 7, 'min_child_samples': 92, 'learning_rate': 0.23536463281405412, 'log_max_bin': 10, 'colsample_bytree': 0.9898009552962395, 'reg_alpha': 0.0009765625, 'reg_lambda': 0.143294261726433, 'FLAML_sample_size': 40000}, 'Best Learner': 'lgbm', 'Best Hyper-parameters': {'n_estimators': 56, 'num_leaves': 7, 'min_child_samples': 92, 'learning_rate': 0.23536463281405412, 'log_max_bin': 10, 'colsample_bytree': 0.9898009552962395, 'reg_alpha': 0.0009765625, 'reg_lambda': 0.143294261726433, 'FLAML_sample_size': 40000}}\n", "{'Current Learner': 'lgbm', 'Current Sample': 364083, 'Current Hyper-parameters': {'n_estimators': 56, 'num_leaves': 7, 'min_child_samples': 92, 'learning_rate': 0.23536463281405412, 'log_max_bin': 10, 'colsample_bytree': 0.9898009552962395, 'reg_alpha': 0.0009765625, 'reg_lambda': 0.143294261726433, 'FLAML_sample_size': 364083}, 'Best Learner': 'lgbm', 'Best Hyper-parameters': {'n_estimators': 56, 'num_leaves': 7, 'min_child_samples': 92, 'learning_rate': 0.23536463281405412, 'log_max_bin': 10, 'colsample_bytree': 0.9898009552962395, 'reg_alpha': 0.0009765625, 'reg_lambda': 0.143294261726433, 'FLAML_sample_size': 364083}}\n", "{'Current Learner': 'lgbm', 'Current Sample': 364083, 'Current Hyper-parameters': {'n_estimators': 179, 'num_leaves': 27, 'min_child_samples': 75, 'learning_rate': 0.09744966359309021, 'log_max_bin': 10, 'colsample_bytree': 1.0, 'reg_alpha': 0.002826104794043855, 'reg_lambda': 0.145731823715616, 'FLAML_sample_size': 364083}, 'Best Learner': 'lgbm', 'Best Hyper-parameters': {'n_estimators': 179, 'num_leaves': 27, 'min_child_samples': 75, 'learning_rate': 0.09744966359309021, 'log_max_bin': 10, 'colsample_bytree': 1.0, 'reg_alpha': 0.002826104794043855, 'reg_lambda': 0.145731823715616, 'FLAML_sample_size': 364083}}\n", "{'Current Learner': 'lgbm', 'Current Sample': 364083, 'Current Hyper-parameters': {'n_estimators': 180, 'num_leaves': 31, 'min_child_samples': 112, 'learning_rate': 0.14172261747380863, 'log_max_bin': 8, 'colsample_bytree': 0.9882716197099741, 'reg_alpha': 0.004676080321450302, 'reg_lambda': 2.7048628270368136, 'FLAML_sample_size': 364083}, 'Best Learner': 'lgbm', 'Best Hyper-parameters': {'n_estimators': 180, 'num_leaves': 31, 'min_child_samples': 112, 'learning_rate': 0.14172261747380863, 'log_max_bin': 8, 'colsample_bytree': 0.9882716197099741, 'reg_alpha': 0.004676080321450302, 'reg_lambda': 2.7048628270368136, 'FLAML_sample_size': 364083}}\n", "{'Current Learner': 'lgbm', 'Current Sample': 364083, 'Current Hyper-parameters': {'n_estimators': 284, 'num_leaves': 24, 'min_child_samples': 57, 'learning_rate': 0.34506374431782616, 'log_max_bin': 8, 'colsample_bytree': 0.9661606582789269, 'reg_alpha': 0.05708594148438563, 'reg_lambda': 3.080643548412343, 'FLAML_sample_size': 364083}, 'Best Learner': 'lgbm', 'Best Hyper-parameters': {'n_estimators': 284, 'num_leaves': 24, 'min_child_samples': 57, 'learning_rate': 0.34506374431782616, 'log_max_bin': 8, 'colsample_bytree': 0.9661606582789269, 'reg_alpha': 0.05708594148438563, 'reg_lambda': 3.080643548412343, 'FLAML_sample_size': 364083}}\n", "{'Current Learner': 'lgbm', 'Current Sample': 364083, 'Current Hyper-parameters': {'n_estimators': 150, 'num_leaves': 176, 'min_child_samples': 62, 'learning_rate': 0.2607939951456863, 'log_max_bin': 8, 'colsample_bytree': 1.0, 'reg_alpha': 0.015973158305354472, 'reg_lambda': 1.1581244082992237, 'FLAML_sample_size': 364083}, 'Best Learner': 'lgbm', 'Best Hyper-parameters': {'n_estimators': 150, 'num_leaves': 176, 'min_child_samples': 62, 'learning_rate': 0.2607939951456863, 'log_max_bin': 8, 'colsample_bytree': 1.0, 'reg_alpha': 0.015973158305354472, 'reg_lambda': 1.1581244082992237, 'FLAML_sample_size': 364083}}\n", "{'Current Learner': 'lgbm', 'Current Sample': 364083, 'Current Hyper-parameters': {'n_estimators': 100, 'num_leaves': 380, 'min_child_samples': 83, 'learning_rate': 0.1439688182217924, 'log_max_bin': 7, 'colsample_bytree': 0.9365250834556608, 'reg_alpha': 0.07492795084698504, 'reg_lambda': 10.854898771631566, 'FLAML_sample_size': 364083}, 'Best Learner': 'lgbm', 'Best Hyper-parameters': {'n_estimators': 100, 'num_leaves': 380, 'min_child_samples': 83, 'learning_rate': 0.1439688182217924, 'log_max_bin': 7, 'colsample_bytree': 0.9365250834556608, 'reg_alpha': 0.07492795084698504, 'reg_lambda': 10.854898771631566, 'FLAML_sample_size': 364083}}\n", "{'Current Learner': 'lgbm', 'Current Sample': 364083, 'Current Hyper-parameters': {'n_estimators': 157, 'num_leaves': 985, 'min_child_samples': 115, 'learning_rate': 0.15986853540486204, 'log_max_bin': 6, 'colsample_bytree': 0.8905312088154893, 'reg_alpha': 0.17376372850615002, 'reg_lambda': 196.8899439847594, 'FLAML_sample_size': 364083}, 'Best Learner': 'lgbm', 'Best Hyper-parameters': {'n_estimators': 157, 'num_leaves': 985, 'min_child_samples': 115, 'learning_rate': 0.15986853540486204, 'log_max_bin': 6, 'colsample_bytree': 0.8905312088154893, 'reg_alpha': 0.17376372850615002, 'reg_lambda': 196.8899439847594, 'FLAML_sample_size': 364083}}\n", "{'Current Learner': 'lgbm', 'Current Sample': 364083, 'Current Hyper-parameters': {'n_estimators': 302, 'num_leaves': 466, 'min_child_samples': 128, 'learning_rate': 0.087493667994037, 'log_max_bin': 7, 'colsample_bytree': 0.763983850698587, 'reg_alpha': 0.09968008477303378, 'reg_lambda': 23.227419343318914, 'FLAML_sample_size': 364083}, 'Best Learner': 'lgbm', 'Best Hyper-parameters': {'n_estimators': 302, 'num_leaves': 466, 'min_child_samples': 128, 'learning_rate': 0.087493667994037, 'log_max_bin': 7, 'colsample_bytree': 0.763983850698587, 'reg_alpha': 0.09968008477303378, 'reg_lambda': 23.227419343318914, 'FLAML_sample_size': 364083}}\n" ] } ], "source": [ "from flaml.data import get_output_from_log\n", "time_history, best_valid_loss_history, valid_loss_history, config_history, metric_history = \\\n", " get_output_from_log(filename=settings['log_file_name'], time_budget=240)\n", "for config in config_history:\n", " print(config)" ] }, { "cell_type": "code", "execution_count": 52, "metadata": { "slideshow": { "slide_type": "slide" } }, "outputs": [ { "data": { "application/vnd.livy.statement-meta+json": { "execution_finish_time": "2023-04-09T03:13:27.2414306Z", "execution_start_time": "2023-04-09T03:13:26.9671462Z", "livy_statement_state": "available", "parent_msg_id": "5e00da90-af15-4ffd-b1b5-b946fabfc565", "queued_time": "2023-04-09T03:10:35.2740852Z", "session_id": "7", "session_start_time": null, "spark_jobs": null, "spark_pool": "automl", "state": "finished", "statement_id": 78 }, "text/plain": [ "StatementMeta(automl, 7, 78, Finished, Available)" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "import matplotlib.pyplot as plt\n", "import numpy as np\n", "\n", "plt.title('Learning Curve')\n", "plt.xlabel('Wall Clock Time (s)')\n", "plt.ylabel('Validation Accuracy')\n", "plt.scatter(time_history, 1 - np.array(valid_loss_history))\n", "plt.step(time_history, 1 - np.array(best_valid_loss_history), where='post')\n", "plt.show()" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "## 3. Comparison with alternatives\n" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "### Default LightGBM" ] }, { "cell_type": "code", "execution_count": 53, "metadata": {}, "outputs": [ { "data": { "application/vnd.livy.statement-meta+json": { "execution_finish_time": "2023-04-09T03:13:27.7753221Z", "execution_start_time": "2023-04-09T03:13:27.4870777Z", "livy_statement_state": "available", "parent_msg_id": "249fba84-ec7c-4801-9dac-861ffa0d0290", "queued_time": "2023-04-09T03:10:35.4112806Z", "session_id": "7", "session_start_time": null, "spark_jobs": null, "spark_pool": "automl", "state": "finished", "statement_id": 79 }, "text/plain": [ "StatementMeta(automl, 7, 79, Finished, Available)" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "from lightgbm import LGBMClassifier\n", "lgbm = LGBMClassifier()" ] }, { "cell_type": "code", "execution_count": 54, "metadata": {}, "outputs": [ { "data": { "application/vnd.livy.statement-meta+json": { "execution_finish_time": "2023-04-09T03:13:29.4430851Z", "execution_start_time": "2023-04-09T03:13:28.0142422Z", "livy_statement_state": "available", "parent_msg_id": "635ca27a-7ae7-44e9-9d57-f81b36236398", "queued_time": "2023-04-09T03:10:35.511851Z", "session_id": "7", "session_start_time": null, "spark_jobs": null, "spark_pool": "automl", "state": "finished", "statement_id": 80 }, "text/plain": [ "StatementMeta(automl, 7, 80, Finished, Available)" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "
LGBMClassifier()
In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.
" ], "text/plain": [ "LGBMClassifier()" ] }, "execution_count": 33, "metadata": {}, "output_type": "execute_result" } ], "source": [ "lgbm.fit(X_train, y_train)" ] }, { "cell_type": "code", "execution_count": 55, "metadata": {}, "outputs": [ { "data": { "application/vnd.livy.statement-meta+json": { "execution_finish_time": "2023-04-09T03:13:30.0093622Z", "execution_start_time": "2023-04-09T03:13:29.7202855Z", "livy_statement_state": "available", "parent_msg_id": "608a77ce-d7b2-4921-adff-d1618a8316ad", "queued_time": "2023-04-09T03:10:35.6550041Z", "session_id": "7", "session_start_time": null, "spark_jobs": null, "spark_pool": "automl", "state": "finished", "statement_id": 81 }, "text/plain": [ "StatementMeta(automl, 7, 81, Finished, Available)" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "y_pred_lgbm = lgbm.predict(X_test)" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "### Default XGBoost" ] }, { "cell_type": "code", "execution_count": 56, "metadata": {}, "outputs": [ { "data": { "application/vnd.livy.statement-meta+json": { "execution_finish_time": "2023-04-09T03:13:30.5721373Z", "execution_start_time": "2023-04-09T03:13:30.2846919Z", "livy_statement_state": "available", "parent_msg_id": "4b08eacb-4745-48d9-b223-ec5fbdab69ab", "queued_time": "2023-04-09T03:10:35.7535047Z", "session_id": "7", "session_start_time": null, "spark_jobs": null, "spark_pool": "automl", "state": "finished", "statement_id": 82 }, "text/plain": [ "StatementMeta(automl, 7, 82, Finished, Available)" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "from xgboost import XGBClassifier\n", "xgb = XGBClassifier()\n", "cat_columns = X_train.select_dtypes(include=['category']).columns\n", "X = X_train.copy()\n", "X[cat_columns] = X[cat_columns].apply(lambda x: x.cat.codes)\n", "y_train_xgb = y_train.astype(\"int\")" ] }, { "cell_type": "code", "execution_count": 57, "metadata": {}, "outputs": [ { "data": { "application/vnd.livy.statement-meta+json": { "execution_finish_time": "2023-04-09T03:13:38.5603565Z", "execution_start_time": "2023-04-09T03:13:30.8138989Z", "livy_statement_state": "available", "parent_msg_id": "7536603f-0254-4f00-aac1-73d67d529a05", "queued_time": "2023-04-09T03:10:35.8542308Z", "session_id": "7", "session_start_time": null, "spark_jobs": null, "spark_pool": "automl", "state": "finished", "statement_id": 83 }, "text/plain": [ "StatementMeta(automl, 7, 83, Finished, Available)" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "
XGBClassifier(base_score=0.5, booster='gbtree', callbacks=None,\n",
              "              colsample_bylevel=1, colsample_bynode=1, colsample_bytree=1,\n",
              "              early_stopping_rounds=None, enable_categorical=False,\n",
              "              eval_metric=None, gamma=0, gpu_id=-1, grow_policy='depthwise',\n",
              "              importance_type=None, interaction_constraints='',\n",
              "              learning_rate=0.300000012, max_bin=256, max_cat_to_onehot=4,\n",
              "              max_delta_step=0, max_depth=6, max_leaves=0, min_child_weight=1,\n",
              "              missing=nan, monotone_constraints='()', n_estimators=100,\n",
              "              n_jobs=0, num_parallel_tree=1, predictor='auto', random_state=0,\n",
              "              reg_alpha=0, reg_lambda=1, ...)
In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.
" ], "text/plain": [ "XGBClassifier(base_score=0.5, booster='gbtree', callbacks=None,\n", " colsample_bylevel=1, colsample_bynode=1, colsample_bytree=1,\n", " early_stopping_rounds=None, enable_categorical=False,\n", " eval_metric=None, gamma=0, gpu_id=-1, grow_policy='depthwise',\n", " importance_type=None, interaction_constraints='',\n", " learning_rate=0.300000012, max_bin=256, max_cat_to_onehot=4,\n", " max_delta_step=0, max_depth=6, max_leaves=0, min_child_weight=1,\n", " missing=nan, monotone_constraints='()', n_estimators=100,\n", " n_jobs=0, num_parallel_tree=1, predictor='auto', random_state=0,\n", " reg_alpha=0, reg_lambda=1, ...)" ] }, "execution_count": 39, "metadata": {}, "output_type": "execute_result" } ], "source": [ "xgb.fit(X, y_train_xgb)" ] }, { "cell_type": "code", "execution_count": 58, "metadata": {}, "outputs": [ { "data": { "application/vnd.livy.statement-meta+json": { "execution_finish_time": "2023-04-09T03:13:39.158293Z", "execution_start_time": "2023-04-09T03:13:38.8646861Z", "livy_statement_state": "available", "parent_msg_id": "6cc9c9ae-70a1-4233-8d7e-87b0f49cfe84", "queued_time": "2023-04-09T03:10:35.9526459Z", "session_id": "7", "session_start_time": null, "spark_jobs": null, "spark_pool": "automl", "state": "finished", "statement_id": 84 }, "text/plain": [ "StatementMeta(automl, 7, 84, Finished, Available)" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "X = X_test.copy()\n", "X[cat_columns] = X[cat_columns].apply(lambda x: x.cat.codes)\n", "y_pred_xgb = xgb.predict(X)\n", "y_test_xgb = y_test.astype(\"int\")\n" ] }, { "cell_type": "code", "execution_count": 59, "metadata": {}, "outputs": [ { "data": { "application/vnd.livy.statement-meta+json": { "execution_finish_time": "2023-04-09T03:13:40.1931477Z", "execution_start_time": "2023-04-09T03:13:39.4172862Z", "livy_statement_state": "available", "parent_msg_id": "ce07a96a-a8a2-43f1-b7fc-c76eb204382e", "queued_time": "2023-04-09T03:10:36.0501561Z", "session_id": "7", "session_start_time": null, "spark_jobs": null, "spark_pool": "automl", "state": "finished", "statement_id": 85 }, "text/plain": [ "StatementMeta(automl, 7, 85, Finished, Available)" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "default xgboost accuracy = 0.6676060098186078\n", "default lgbm accuracy = 0.6602346380315323\n", "flaml (10 min) accuracy = 0.6732939797991784\n" ] } ], "source": [ "print('default xgboost accuracy', '=', 1 - sklearn_metric_loss_score('accuracy', y_pred_xgb, y_test_xgb))\n", "print('default lgbm accuracy', '=', 1 - sklearn_metric_loss_score('accuracy', y_pred_lgbm, y_test))\n", "print('flaml (2 min) accuracy', '=', 1 - sklearn_metric_loss_score('accuracy', y_pred, y_test))" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "## 4. Customized Learner" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "Some experienced automl users may have a preferred model to tune or may already have a reasonably by-hand-tuned model before launching the automl experiment. They need to select optimal configurations for the customized model mixed with standard built-in learners. \n", "\n", "FLAML can easily incorporate customized/new learners (preferably with sklearn API) provided by users in a real-time manner, as demonstrated below." ] }, { "attachments": {}, "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "### Example of Regularized Greedy Forest\n", "\n", "[Regularized Greedy Forest](https://arxiv.org/abs/1109.0887) (RGF) is a machine learning method currently not included in FLAML. The RGF has many tuning parameters, the most critical of which are: `[max_leaf, n_iter, n_tree_search, opt_interval, min_samples_leaf]`. To run a customized/new learner, the user needs to provide the following information:\n", "* an implementation of the customized/new learner\n", "* a list of hyperparameter names and types\n", "* rough ranges of hyperparameters (i.e., upper/lower bounds)\n", "* choose initial value corresponding to low cost for cost-related hyperparameters (e.g., initial value for max_leaf and n_iter should be small)\n", "\n", "In this example, the above information for RGF is wrapped in a python class called *MyRegularizedGreedyForest* that exposes the hyperparameters." ] }, { "cell_type": "code", "execution_count": 60, "metadata": {}, "outputs": [ { "data": { "application/vnd.livy.statement-meta+json": { "execution_finish_time": "2023-04-09T03:13:50.122632Z", "execution_start_time": "2023-04-09T03:13:40.4359303Z", "livy_statement_state": "available", "parent_msg_id": "4855a514-2527-4852-95e2-743f509bf2c7", "queued_time": "2023-04-09T03:10:36.1656825Z", "session_id": "7", "session_start_time": null, "spark_jobs": null, "spark_pool": "automl", "state": "finished", "statement_id": 86 }, "text/plain": [ "StatementMeta(automl, 7, 86, Finished, Available)" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "Collecting rgf-python\n", " Using cached rgf_python-3.12.0-py3-none-manylinux1_x86_64.whl (757 kB)\n", "Requirement already satisfied: joblib in /home/trusted-service-user/cluster-env/env/lib/python3.8/site-packages (from rgf-python) (1.0.1)\n", "Requirement already satisfied: scikit-learn>=0.18 in /home/trusted-service-user/cluster-env/env/lib/python3.8/site-packages (from rgf-python) (0.23.2)\n", "Requirement already satisfied: numpy>=1.13.3 in /home/trusted-service-user/cluster-env/env/lib/python3.8/site-packages (from scikit-learn>=0.18->rgf-python) (1.19.4)\n", "Requirement already satisfied: threadpoolctl>=2.0.0 in /home/trusted-service-user/cluster-env/env/lib/python3.8/site-packages (from scikit-learn>=0.18->rgf-python) (2.1.0)\n", "Requirement already satisfied: scipy>=0.19.1 in /home/trusted-service-user/cluster-env/env/lib/python3.8/site-packages (from scikit-learn>=0.18->rgf-python) (1.5.3)\n", "Installing collected packages: rgf-python\n", "Successfully installed rgf-python-3.12.0\n" ] } ], "source": [ "!pip install rgf-python " ] }, { "cell_type": "code", "execution_count": 61, "metadata": { "slideshow": { "slide_type": "slide" } }, "outputs": [ { "data": { "application/vnd.livy.statement-meta+json": { "execution_finish_time": "2023-04-09T03:13:50.6337005Z", "execution_start_time": "2023-04-09T03:13:50.3672163Z", "livy_statement_state": "available", "parent_msg_id": "6f475eea-c02b-491f-a85e-e696dfdf6882", "queued_time": "2023-04-09T03:10:36.2639428Z", "session_id": "7", "session_start_time": null, "spark_jobs": null, "spark_pool": "automl", "state": "finished", "statement_id": 87 }, "text/plain": [ "StatementMeta(automl, 7, 87, Finished, Available)" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "''' SKLearnEstimator is the super class for a sklearn learner '''\n", "from flaml.model import SKLearnEstimator\n", "from flaml import tune\n", "from flaml.data import CLASSIFICATION\n", "\n", "\n", "class MyRegularizedGreedyForest(SKLearnEstimator):\n", " def __init__(self, task='binary', **config):\n", " '''Constructor\n", " \n", " Args:\n", " task: A string of the task type, one of\n", " 'binary', 'multiclass', 'regression'\n", " config: A dictionary containing the hyperparameter names\n", " and 'n_jobs' as keys. n_jobs is the number of parallel threads.\n", " '''\n", "\n", " super().__init__(task, **config)\n", "\n", " '''task=binary or multi for classification task'''\n", " if task in CLASSIFICATION:\n", " from rgf.sklearn import RGFClassifier\n", "\n", " self.estimator_class = RGFClassifier\n", " else:\n", " from rgf.sklearn import RGFRegressor\n", " \n", " self.estimator_class = RGFRegressor\n", "\n", " @classmethod\n", " def search_space(cls, data_size, task):\n", " '''[required method] search space\n", "\n", " Returns:\n", " A dictionary of the search space. \n", " Each key is the name of a hyperparameter, and value is a dict with\n", " its domain (required) and low_cost_init_value, init_value,\n", " cat_hp_cost (if applicable).\n", " e.g.,\n", " {'domain': tune.randint(lower=1, upper=10), 'init_value': 1}.\n", " '''\n", " space = { \n", " 'max_leaf': {'domain': tune.lograndint(lower=4, upper=data_size[0]), 'init_value': 4, 'low_cost_init_value': 4},\n", " 'n_iter': {'domain': tune.lograndint(lower=1, upper=data_size[0]), 'init_value': 1, 'low_cost_init_value': 1},\n", " 'n_tree_search': {'domain': tune.lograndint(lower=1, upper=32768), 'init_value': 1, 'low_cost_init_value': 1},\n", " 'opt_interval': {'domain': tune.lograndint(lower=1, upper=10000), 'init_value': 100},\n", " 'learning_rate': {'domain': tune.loguniform(lower=0.01, upper=20.0)},\n", " 'min_samples_leaf': {'domain': tune.lograndint(lower=1, upper=20), 'init_value': 20},\n", " }\n", " return space\n", "\n", " @classmethod\n", " def size(cls, config):\n", " '''[optional method] memory size of the estimator in bytes\n", " \n", " Args:\n", " config - the dict of the hyperparameter config\n", "\n", " Returns:\n", " A float of the memory size required by the estimator to train the\n", " given config\n", " '''\n", " max_leaves = int(round(config['max_leaf']))\n", " n_estimators = int(round(config['n_iter']))\n", " return (max_leaves * 3 + (max_leaves - 1) * 4 + 1.0) * n_estimators * 8\n", "\n", " @classmethod\n", " def cost_relative2lgbm(cls):\n", " '''[optional method] relative cost compared to lightgbm\n", " '''\n", " return 1.0\n" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "### Add Customized Learner and Run FLAML AutoML\n", "\n", "After adding RGF into the list of learners, we run automl by tuning hyperpameters of RGF as well as the default learners. " ] }, { "cell_type": "code", "execution_count": 62, "metadata": { "slideshow": { "slide_type": "slide" } }, "outputs": [ { "data": { "application/vnd.livy.statement-meta+json": { "execution_finish_time": "2023-04-09T03:13:51.1287115Z", "execution_start_time": "2023-04-09T03:13:50.8741632Z", "livy_statement_state": "available", "parent_msg_id": "702a9e5c-a880-483b-985c-4ebbcbde5e07", "queued_time": "2023-04-09T03:10:36.3578919Z", "session_id": "7", "session_start_time": null, "spark_jobs": null, "spark_pool": "automl", "state": "finished", "statement_id": 88 }, "text/plain": [ "StatementMeta(automl, 7, 88, Finished, Available)" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "automl = AutoML()\n", "automl.add_learner(learner_name='RGF', learner_class=MyRegularizedGreedyForest)" ] }, { "cell_type": "code", "execution_count": 63, "metadata": { "slideshow": { "slide_type": "slide" }, "tags": [] }, "outputs": [ { "data": { "application/vnd.livy.statement-meta+json": { "execution_finish_time": "2023-04-09T03:14:03.5802415Z", "execution_start_time": "2023-04-09T03:13:51.3699652Z", "livy_statement_state": "available", "parent_msg_id": "2e5e85aa-8e78-4d78-a275-c6a160a7b415", "queued_time": "2023-04-09T03:10:36.4663752Z", "session_id": "7", "session_start_time": null, "spark_jobs": null, "spark_pool": "automl", "state": "finished", "statement_id": 89 }, "text/plain": [ "StatementMeta(automl, 7, 89, Finished, Available)" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "[flaml.automl.automl: 04-09 03:13:51] {2726} INFO - task = classification\n", "[flaml.automl.automl: 04-09 03:13:51] {2728} INFO - Data split method: stratified\n", "[flaml.automl.automl: 04-09 03:13:51] {2731} INFO - Evaluation method: holdout\n", "[flaml.automl.automl: 04-09 03:13:51] {2858} INFO - Minimizing error metric: 1-accuracy\n", "[flaml.automl.automl: 04-09 03:13:51] {3004} INFO - List of ML learners in AutoML Run: ['RGF', 'lgbm', 'rf', 'xgboost']\n", "[flaml.automl.automl: 04-09 03:13:51] {3334} INFO - iteration 0, current learner RGF\n", "[flaml.automl.automl: 04-09 03:13:52] {3472} INFO - Estimated sufficient time budget=173368s. Estimated necessary time budget=173s.\n", "[flaml.automl.automl: 04-09 03:13:52] {3519} INFO - at 0.9s,\testimator RGF's best error=0.3840,\tbest estimator RGF's best error=0.3840\n", "[flaml.automl.automl: 04-09 03:13:52] {3334} INFO - iteration 1, current learner RGF\n", "[flaml.automl.automl: 04-09 03:13:52] {3519} INFO - at 1.2s,\testimator RGF's best error=0.3840,\tbest estimator RGF's best error=0.3840\n", "[flaml.automl.automl: 04-09 03:13:52] {3334} INFO - iteration 2, current learner RGF\n", "[flaml.automl.automl: 04-09 03:13:52] {3519} INFO - at 1.6s,\testimator RGF's best error=0.3840,\tbest estimator RGF's best error=0.3840\n", "[flaml.automl.automl: 04-09 03:13:52] {3334} INFO - iteration 3, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:13:52] {3519} INFO - at 1.6s,\testimator lgbm's best error=0.3777,\tbest estimator lgbm's best error=0.3777\n", "[flaml.automl.automl: 04-09 03:13:52] {3334} INFO - iteration 4, current learner RGF\n", "[flaml.automl.automl: 04-09 03:13:53] {3519} INFO - at 2.0s,\testimator RGF's best error=0.3840,\tbest estimator lgbm's best error=0.3777\n", "[flaml.automl.automl: 04-09 03:13:53] {3334} INFO - iteration 5, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:13:53] {3519} INFO - at 2.1s,\testimator lgbm's best error=0.3777,\tbest estimator lgbm's best error=0.3777\n", "[flaml.automl.automl: 04-09 03:13:53] {3334} INFO - iteration 6, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:13:53] {3519} INFO - at 2.1s,\testimator lgbm's best error=0.3777,\tbest estimator lgbm's best error=0.3777\n", "[flaml.automl.automl: 04-09 03:13:53] {3334} INFO - iteration 7, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:13:53] {3519} INFO - at 2.1s,\testimator lgbm's best error=0.3661,\tbest estimator lgbm's best error=0.3661\n", "[flaml.automl.automl: 04-09 03:13:53] {3334} INFO - iteration 8, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:13:53] {3519} INFO - at 2.2s,\testimator lgbm's best error=0.3661,\tbest estimator lgbm's best error=0.3661\n", "[flaml.automl.automl: 04-09 03:13:53] {3334} INFO - iteration 9, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:13:53] {3519} INFO - at 2.2s,\testimator lgbm's best error=0.3633,\tbest estimator lgbm's best error=0.3633\n", "[flaml.automl.automl: 04-09 03:13:53] {3334} INFO - iteration 10, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:13:53] {3519} INFO - at 2.2s,\testimator lgbm's best error=0.3633,\tbest estimator lgbm's best error=0.3633\n", "[flaml.automl.automl: 04-09 03:13:53] {3334} INFO - iteration 11, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:13:53] {3519} INFO - at 2.3s,\testimator lgbm's best error=0.3633,\tbest estimator lgbm's best error=0.3633\n", "[flaml.automl.automl: 04-09 03:13:53] {3334} INFO - iteration 12, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:13:53] {3519} INFO - at 2.3s,\testimator lgbm's best error=0.3613,\tbest estimator lgbm's best error=0.3613\n", "[flaml.automl.automl: 04-09 03:13:53] {3334} INFO - iteration 13, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:13:53] {3519} INFO - at 2.4s,\testimator lgbm's best error=0.3613,\tbest estimator lgbm's best error=0.3613\n", "[flaml.automl.automl: 04-09 03:13:53] {3334} INFO - iteration 14, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:13:53] {3519} INFO - at 2.5s,\testimator lgbm's best error=0.3591,\tbest estimator lgbm's best error=0.3591\n", "[flaml.automl.automl: 04-09 03:13:53] {3334} INFO - iteration 15, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:13:54] {3519} INFO - at 2.7s,\testimator lgbm's best error=0.3591,\tbest estimator lgbm's best error=0.3591\n", "[flaml.automl.automl: 04-09 03:13:54] {3334} INFO - iteration 16, current learner RGF\n", "[flaml.automl.automl: 04-09 03:13:54] {3519} INFO - at 3.1s,\testimator RGF's best error=0.3840,\tbest estimator lgbm's best error=0.3591\n", "[flaml.automl.automl: 04-09 03:13:54] {3334} INFO - iteration 17, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:13:54] {3519} INFO - at 3.2s,\testimator lgbm's best error=0.3591,\tbest estimator lgbm's best error=0.3591\n", "[flaml.automl.automl: 04-09 03:13:54] {3334} INFO - iteration 18, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:13:54] {3519} INFO - at 3.4s,\testimator lgbm's best error=0.3591,\tbest estimator lgbm's best error=0.3591\n", "[flaml.automl.automl: 04-09 03:13:54] {3334} INFO - iteration 19, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:13:54] {3519} INFO - at 3.5s,\testimator lgbm's best error=0.3591,\tbest estimator lgbm's best error=0.3591\n", "[flaml.automl.automl: 04-09 03:13:54] {3334} INFO - iteration 20, current learner RGF\n", "[flaml.automl.automl: 04-09 03:13:55] {3519} INFO - at 4.0s,\testimator RGF's best error=0.3766,\tbest estimator lgbm's best error=0.3591\n", "[flaml.automl.automl: 04-09 03:13:55] {3334} INFO - iteration 21, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:13:55] {3519} INFO - at 4.1s,\testimator lgbm's best error=0.3591,\tbest estimator lgbm's best error=0.3591\n", "[flaml.automl.automl: 04-09 03:13:55] {3334} INFO - iteration 22, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:13:55] {3519} INFO - at 4.5s,\testimator lgbm's best error=0.3514,\tbest estimator lgbm's best error=0.3514\n", "[flaml.automl.automl: 04-09 03:13:55] {3334} INFO - iteration 23, current learner xgboost\n", "[flaml.automl.automl: 04-09 03:13:56] {3519} INFO - at 4.7s,\testimator xgboost's best error=0.3787,\tbest estimator lgbm's best error=0.3514\n", "[flaml.automl.automl: 04-09 03:13:56] {3334} INFO - iteration 24, current learner xgboost\n", "[flaml.automl.automl: 04-09 03:13:56] {3519} INFO - at 4.8s,\testimator xgboost's best error=0.3765,\tbest estimator lgbm's best error=0.3514\n", "[flaml.automl.automl: 04-09 03:13:56] {3334} INFO - iteration 25, current learner rf\n", "[flaml.automl.automl: 04-09 03:13:56] {3519} INFO - at 4.8s,\testimator rf's best error=0.3816,\tbest estimator lgbm's best error=0.3514\n", "[flaml.automl.automl: 04-09 03:13:56] {3334} INFO - iteration 26, current learner rf\n", "[flaml.automl.automl: 04-09 03:13:56] {3519} INFO - at 4.9s,\testimator rf's best error=0.3724,\tbest estimator lgbm's best error=0.3514\n", "[flaml.automl.automl: 04-09 03:13:56] {3334} INFO - iteration 27, current learner rf\n", "[flaml.automl.automl: 04-09 03:13:56] {3519} INFO - at 4.9s,\testimator rf's best error=0.3724,\tbest estimator lgbm's best error=0.3514\n", "[flaml.automl.automl: 04-09 03:13:56] {3334} INFO - iteration 28, current learner xgboost\n", "[flaml.automl.automl: 04-09 03:13:56] {3519} INFO - at 5.0s,\testimator xgboost's best error=0.3765,\tbest estimator lgbm's best error=0.3514\n", "[flaml.automl.automl: 04-09 03:13:56] {3334} INFO - iteration 29, current learner xgboost\n", "[flaml.automl.automl: 04-09 03:13:56] {3519} INFO - at 5.0s,\testimator xgboost's best error=0.3765,\tbest estimator lgbm's best error=0.3514\n", "[flaml.automl.automl: 04-09 03:13:56] {3334} INFO - iteration 30, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:13:56] {3519} INFO - at 5.4s,\testimator lgbm's best error=0.3511,\tbest estimator lgbm's best error=0.3511\n", "[flaml.automl.automl: 04-09 03:13:56] {3334} INFO - iteration 31, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:13:57] {3519} INFO - at 5.7s,\testimator lgbm's best error=0.3497,\tbest estimator lgbm's best error=0.3497\n", "[flaml.automl.automl: 04-09 03:13:57] {3334} INFO - iteration 32, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:13:57] {3519} INFO - at 5.9s,\testimator lgbm's best error=0.3497,\tbest estimator lgbm's best error=0.3497\n", "[flaml.automl.automl: 04-09 03:13:57] {3334} INFO - iteration 33, current learner rf\n", "[flaml.automl.automl: 04-09 03:13:57] {3519} INFO - at 6.0s,\testimator rf's best error=0.3724,\tbest estimator lgbm's best error=0.3497\n", "[flaml.automl.automl: 04-09 03:13:57] {3334} INFO - iteration 34, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:13:57] {3519} INFO - at 6.3s,\testimator lgbm's best error=0.3497,\tbest estimator lgbm's best error=0.3497\n", "[flaml.automl.automl: 04-09 03:13:57] {3334} INFO - iteration 35, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:13:57] {3519} INFO - at 6.6s,\testimator lgbm's best error=0.3497,\tbest estimator lgbm's best error=0.3497\n", "[flaml.automl.automl: 04-09 03:13:57] {3334} INFO - iteration 36, current learner xgboost\n", "[flaml.automl.automl: 04-09 03:13:57] {3519} INFO - at 6.7s,\testimator xgboost's best error=0.3699,\tbest estimator lgbm's best error=0.3497\n", "[flaml.automl.automl: 04-09 03:13:57] {3334} INFO - iteration 37, current learner rf\n", "[flaml.automl.automl: 04-09 03:13:58] {3519} INFO - at 6.7s,\testimator rf's best error=0.3724,\tbest estimator lgbm's best error=0.3497\n", "[flaml.automl.automl: 04-09 03:13:58] {3334} INFO - iteration 38, current learner xgboost\n", "[flaml.automl.automl: 04-09 03:13:58] {3519} INFO - at 6.8s,\testimator xgboost's best error=0.3699,\tbest estimator lgbm's best error=0.3497\n", "[flaml.automl.automl: 04-09 03:13:58] {3334} INFO - iteration 39, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:13:58] {3519} INFO - at 7.1s,\testimator lgbm's best error=0.3497,\tbest estimator lgbm's best error=0.3497\n", "[flaml.automl.automl: 04-09 03:13:58] {3334} INFO - iteration 40, current learner rf\n", "[flaml.automl.automl: 04-09 03:13:58] {3519} INFO - at 7.3s,\testimator rf's best error=0.3724,\tbest estimator lgbm's best error=0.3497\n", "[flaml.automl.automl: 04-09 03:13:58] {3334} INFO - iteration 41, current learner xgboost\n", "[flaml.automl.automl: 04-09 03:13:58] {3519} INFO - at 7.4s,\testimator xgboost's best error=0.3657,\tbest estimator lgbm's best error=0.3497\n", "[flaml.automl.automl: 04-09 03:13:58] {3334} INFO - iteration 42, current learner RGF\n", "[flaml.automl.automl: 04-09 03:13:59] {3519} INFO - at 7.7s,\testimator RGF's best error=0.3766,\tbest estimator lgbm's best error=0.3497\n", "[flaml.automl.automl: 04-09 03:13:59] {3334} INFO - iteration 43, current learner xgboost\n", "[flaml.automl.automl: 04-09 03:13:59] {3519} INFO - at 7.8s,\testimator xgboost's best error=0.3657,\tbest estimator lgbm's best error=0.3497\n", "[flaml.automl.automl: 04-09 03:13:59] {3334} INFO - iteration 44, current learner xgboost\n", "[flaml.automl.automl: 04-09 03:13:59] {3519} INFO - at 7.8s,\testimator xgboost's best error=0.3657,\tbest estimator lgbm's best error=0.3497\n", "[flaml.automl.automl: 04-09 03:13:59] {3334} INFO - iteration 45, current learner xgboost\n", "[flaml.automl.automl: 04-09 03:13:59] {3519} INFO - at 7.9s,\testimator xgboost's best error=0.3657,\tbest estimator lgbm's best error=0.3497\n", "[flaml.automl.automl: 04-09 03:13:59] {3334} INFO - iteration 46, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:13:59] {3519} INFO - at 8.1s,\testimator lgbm's best error=0.3497,\tbest estimator lgbm's best error=0.3497\n", "[flaml.automl.automl: 04-09 03:13:59] {3334} INFO - iteration 47, current learner xgboost\n", "[flaml.automl.automl: 04-09 03:13:59] {3519} INFO - at 8.3s,\testimator xgboost's best error=0.3657,\tbest estimator lgbm's best error=0.3497\n", "[flaml.automl.automl: 04-09 03:13:59] {3334} INFO - iteration 48, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:13:59] {3519} INFO - at 8.4s,\testimator lgbm's best error=0.3497,\tbest estimator lgbm's best error=0.3497\n", "[flaml.automl.automl: 04-09 03:13:59] {3334} INFO - iteration 49, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:13:59] {3519} INFO - at 8.5s,\testimator lgbm's best error=0.3497,\tbest estimator lgbm's best error=0.3497\n", "[flaml.automl.automl: 04-09 03:13:59] {3334} INFO - iteration 50, current learner xgboost\n", "[flaml.automl.automl: 04-09 03:14:00] {3519} INFO - at 8.7s,\testimator xgboost's best error=0.3657,\tbest estimator lgbm's best error=0.3497\n", "[flaml.automl.automl: 04-09 03:14:00] {3334} INFO - iteration 51, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:14:01] {3519} INFO - at 10.5s,\testimator lgbm's best error=0.3448,\tbest estimator lgbm's best error=0.3448\n", "[flaml.automl.automl: 04-09 03:14:03] {3783} INFO - retrain lgbm for 1.6s\n", "[flaml.automl.automl: 04-09 03:14:03] {3790} INFO - retrained model: LGBMClassifier(colsample_bytree=0.6649148062238498,\n", " learning_rate=0.06500463168967066, max_bin=255,\n", " min_child_samples=5, n_estimators=190, num_leaves=20,\n", " reg_alpha=0.0017271108100233477, reg_lambda=0.00468154746700776,\n", " verbose=-1)\n", "[flaml.automl.automl: 04-09 03:14:03] {3034} INFO - fit succeeded\n", "[flaml.automl.automl: 04-09 03:14:03] {3035} INFO - Time taken to find the best model: 10.480074405670166\n" ] } ], "source": [ "settings = {\n", " \"time_budget\": 10, # total running time in seconds\n", " \"metric\": 'accuracy', \n", " \"estimator_list\": ['RGF', 'lgbm', 'rf', 'xgboost'], # list of ML learners\n", " \"task\": 'classification', # task type \n", " \"log_file_name\": 'airlines_experiment_custom_learner.log', # flaml log file \n", " \"log_training_metric\": True, # whether to log training metric\n", "}\n", "\n", "automl.fit(X_train=X_train, y_train=y_train, **settings)" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "## 5. Customized Metric\n", "\n", "It's also easy to customize the optimization metric. As an example, we demonstrate with a custom metric function which combines training loss and validation loss as the final loss to minimize." ] }, { "cell_type": "code", "execution_count": 64, "metadata": {}, "outputs": [ { "data": { "application/vnd.livy.statement-meta+json": { "execution_finish_time": "2023-04-09T03:14:04.1303148Z", "execution_start_time": "2023-04-09T03:14:03.8308127Z", "livy_statement_state": "available", "parent_msg_id": "e1ced49a-d49a-4496-8ded-58deb936d247", "queued_time": "2023-04-09T03:10:36.6448318Z", "session_id": "7", "session_start_time": null, "spark_jobs": null, "spark_pool": "automl", "state": "finished", "statement_id": 90 }, "text/plain": [ "StatementMeta(automl, 7, 90, Finished, Available)" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "def custom_metric(X_val, y_val, estimator, labels, X_train, y_train,\n", " weight_val=None, weight_train=None, config=None,\n", " groups_val=None, groups_train=None):\n", " from sklearn.metrics import log_loss\n", " import time\n", " start = time.time()\n", " y_pred = estimator.predict_proba(X_val)\n", " pred_time = (time.time() - start) / len(X_val)\n", " val_loss = log_loss(y_val, y_pred, labels=labels,\n", " sample_weight=weight_val)\n", " y_pred = estimator.predict_proba(X_train)\n", " train_loss = log_loss(y_train, y_pred, labels=labels,\n", " sample_weight=weight_train)\n", " alpha = 0.5\n", " return val_loss * (1 + alpha) - alpha * train_loss, {\n", " \"val_loss\": val_loss, \"train_loss\": train_loss, \"pred_time\": pred_time\n", " }\n", " # two elements are returned:\n", " # the first element is the metric to minimize as a float number,\n", " # the second element is a dictionary of the metrics to log" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "We can then pass this custom metric function to automl's `fit` method." ] }, { "cell_type": "code", "execution_count": 65, "metadata": {}, "outputs": [ { "data": { "application/vnd.livy.statement-meta+json": { "execution_finish_time": "2023-04-09T03:14:16.3791532Z", "execution_start_time": "2023-04-09T03:14:04.3643576Z", "livy_statement_state": "available", "parent_msg_id": "e472943a-3204-41fc-a723-5f39f302b04c", "queued_time": "2023-04-09T03:10:36.8448553Z", "session_id": "7", "session_start_time": null, "spark_jobs": null, "spark_pool": "automl", "state": "finished", "statement_id": 91 }, "text/plain": [ "StatementMeta(automl, 7, 91, Finished, Available)" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "[flaml.automl.automl: 04-09 03:14:04] {2726} INFO - task = classification\n", "[flaml.automl.automl: 04-09 03:14:04] {2728} INFO - Data split method: stratified\n", "[flaml.automl.automl: 04-09 03:14:04] {2731} INFO - Evaluation method: holdout\n", "[flaml.automl.automl: 04-09 03:14:04] {2858} INFO - Minimizing error metric: customized metric\n", "[flaml.automl.automl: 04-09 03:14:04] {3004} INFO - List of ML learners in AutoML Run: ['lgbm', 'rf', 'xgboost', 'extra_tree', 'xgb_limitdepth', 'lrl1']\n", "[flaml.automl.automl: 04-09 03:14:04] {3334} INFO - iteration 0, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:14:04] {3472} INFO - Estimated sufficient time budget=11191s. Estimated necessary time budget=258s.\n", "[flaml.automl.automl: 04-09 03:14:04] {3519} INFO - at 0.4s,\testimator lgbm's best error=0.6647,\tbest estimator lgbm's best error=0.6647\n", "[flaml.automl.automl: 04-09 03:14:04] {3334} INFO - iteration 1, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:14:04] {3519} INFO - at 0.5s,\testimator lgbm's best error=0.6647,\tbest estimator lgbm's best error=0.6647\n", "[flaml.automl.automl: 04-09 03:14:04] {3334} INFO - iteration 2, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:14:04] {3519} INFO - at 0.5s,\testimator lgbm's best error=0.6491,\tbest estimator lgbm's best error=0.6491\n", "[flaml.automl.automl: 04-09 03:14:04] {3334} INFO - iteration 3, current learner xgboost\n", "[flaml.automl.automl: 04-09 03:14:05] {3519} INFO - at 0.7s,\testimator xgboost's best error=0.6845,\tbest estimator lgbm's best error=0.6491\n", "[flaml.automl.automl: 04-09 03:14:05] {3334} INFO - iteration 4, current learner extra_tree\n", "[flaml.automl.automl: 04-09 03:14:05] {3519} INFO - at 0.8s,\testimator extra_tree's best error=0.6678,\tbest estimator lgbm's best error=0.6491\n", "[flaml.automl.automl: 04-09 03:14:05] {3334} INFO - iteration 5, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:14:05] {3519} INFO - at 0.8s,\testimator lgbm's best error=0.6423,\tbest estimator lgbm's best error=0.6423\n", "[flaml.automl.automl: 04-09 03:14:05] {3334} INFO - iteration 6, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:14:05] {3519} INFO - at 0.9s,\testimator lgbm's best error=0.6423,\tbest estimator lgbm's best error=0.6423\n", "[flaml.automl.automl: 04-09 03:14:05] {3334} INFO - iteration 7, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:14:05] {3519} INFO - at 0.9s,\testimator lgbm's best error=0.6423,\tbest estimator lgbm's best error=0.6423\n", "[flaml.automl.automl: 04-09 03:14:05] {3334} INFO - iteration 8, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:14:05] {3519} INFO - at 0.9s,\testimator lgbm's best error=0.6400,\tbest estimator lgbm's best error=0.6400\n", "[flaml.automl.automl: 04-09 03:14:05] {3334} INFO - iteration 9, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:14:05] {3519} INFO - at 1.0s,\testimator lgbm's best error=0.6400,\tbest estimator lgbm's best error=0.6400\n", "[flaml.automl.automl: 04-09 03:14:05] {3334} INFO - iteration 10, current learner xgboost\n", "[flaml.automl.automl: 04-09 03:14:05] {3519} INFO - at 1.2s,\testimator xgboost's best error=0.6845,\tbest estimator lgbm's best error=0.6400\n", "[flaml.automl.automl: 04-09 03:14:05] {3334} INFO - iteration 11, current learner extra_tree\n", "[flaml.automl.automl: 04-09 03:14:05] {3519} INFO - at 1.3s,\testimator extra_tree's best error=0.6576,\tbest estimator lgbm's best error=0.6400\n", "[flaml.automl.automl: 04-09 03:14:05] {3334} INFO - iteration 12, current learner rf\n", "[flaml.automl.automl: 04-09 03:14:05] {3519} INFO - at 1.3s,\testimator rf's best error=0.6614,\tbest estimator lgbm's best error=0.6400\n", "[flaml.automl.automl: 04-09 03:14:05] {3334} INFO - iteration 13, current learner rf\n", "[flaml.automl.automl: 04-09 03:14:05] {3519} INFO - at 1.4s,\testimator rf's best error=0.6523,\tbest estimator lgbm's best error=0.6400\n", "[flaml.automl.automl: 04-09 03:14:05] {3334} INFO - iteration 14, current learner rf\n", "[flaml.automl.automl: 04-09 03:14:05] {3519} INFO - at 1.4s,\testimator rf's best error=0.6523,\tbest estimator lgbm's best error=0.6400\n", "[flaml.automl.automl: 04-09 03:14:05] {3334} INFO - iteration 15, current learner xgboost\n", "[flaml.automl.automl: 04-09 03:14:05] {3519} INFO - at 1.5s,\testimator xgboost's best error=0.6503,\tbest estimator lgbm's best error=0.6400\n", "[flaml.automl.automl: 04-09 03:14:05] {3334} INFO - iteration 16, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:14:05] {3519} INFO - at 1.6s,\testimator lgbm's best error=0.6400,\tbest estimator lgbm's best error=0.6400\n", "[flaml.automl.automl: 04-09 03:14:05] {3334} INFO - iteration 17, current learner extra_tree\n", "[flaml.automl.automl: 04-09 03:14:06] {3519} INFO - at 1.8s,\testimator extra_tree's best error=0.6576,\tbest estimator lgbm's best error=0.6400\n", "[flaml.automl.automl: 04-09 03:14:06] {3334} INFO - iteration 18, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:14:06] {3519} INFO - at 1.8s,\testimator lgbm's best error=0.6400,\tbest estimator lgbm's best error=0.6400\n", "[flaml.automl.automl: 04-09 03:14:06] {3334} INFO - iteration 19, current learner xgboost\n", "[flaml.automl.automl: 04-09 03:14:06] {3519} INFO - at 2.0s,\testimator xgboost's best error=0.6486,\tbest estimator lgbm's best error=0.6400\n", "[flaml.automl.automl: 04-09 03:14:06] {3334} INFO - iteration 20, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:14:06] {3519} INFO - at 2.1s,\testimator lgbm's best error=0.6335,\tbest estimator lgbm's best error=0.6335\n", "[flaml.automl.automl: 04-09 03:14:06] {3334} INFO - iteration 21, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:14:06] {3519} INFO - at 2.1s,\testimator lgbm's best error=0.6335,\tbest estimator lgbm's best error=0.6335\n", "[flaml.automl.automl: 04-09 03:14:06] {3334} INFO - iteration 22, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:14:06] {3519} INFO - at 2.2s,\testimator lgbm's best error=0.6335,\tbest estimator lgbm's best error=0.6335\n", "[flaml.automl.automl: 04-09 03:14:06] {3334} INFO - iteration 23, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:14:06] {3519} INFO - at 2.3s,\testimator lgbm's best error=0.6335,\tbest estimator lgbm's best error=0.6335\n", "[flaml.automl.automl: 04-09 03:14:06] {3334} INFO - iteration 24, current learner rf\n", "[flaml.automl.automl: 04-09 03:14:06] {3519} INFO - at 2.4s,\testimator rf's best error=0.6523,\tbest estimator lgbm's best error=0.6335\n", "[flaml.automl.automl: 04-09 03:14:06] {3334} INFO - iteration 25, current learner extra_tree\n", "[flaml.automl.automl: 04-09 03:14:06] {3519} INFO - at 2.5s,\testimator extra_tree's best error=0.6576,\tbest estimator lgbm's best error=0.6335\n", "[flaml.automl.automl: 04-09 03:14:06] {3334} INFO - iteration 26, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:14:06] {3519} INFO - at 2.6s,\testimator lgbm's best error=0.6335,\tbest estimator lgbm's best error=0.6335\n", "[flaml.automl.automl: 04-09 03:14:06] {3334} INFO - iteration 27, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:14:07] {3519} INFO - at 2.9s,\testimator lgbm's best error=0.6328,\tbest estimator lgbm's best error=0.6328\n", "[flaml.automl.automl: 04-09 03:14:07] {3334} INFO - iteration 28, current learner extra_tree\n", "[flaml.automl.automl: 04-09 03:14:07] {3519} INFO - at 3.0s,\testimator extra_tree's best error=0.6576,\tbest estimator lgbm's best error=0.6328\n", "[flaml.automl.automl: 04-09 03:14:07] {3334} INFO - iteration 29, current learner extra_tree\n", "[flaml.automl.automl: 04-09 03:14:07] {3519} INFO - at 3.1s,\testimator extra_tree's best error=0.6443,\tbest estimator lgbm's best error=0.6328\n", "[flaml.automl.automl: 04-09 03:14:07] {3334} INFO - iteration 30, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:14:07] {3519} INFO - at 3.4s,\testimator lgbm's best error=0.6241,\tbest estimator lgbm's best error=0.6241\n", "[flaml.automl.automl: 04-09 03:14:07] {3334} INFO - iteration 31, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:14:07] {3519} INFO - at 3.7s,\testimator lgbm's best error=0.6241,\tbest estimator lgbm's best error=0.6241\n", "[flaml.automl.automl: 04-09 03:14:07] {3334} INFO - iteration 32, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:14:08] {3519} INFO - at 4.0s,\testimator lgbm's best error=0.6206,\tbest estimator lgbm's best error=0.6206\n", "[flaml.automl.automl: 04-09 03:14:08] {3334} INFO - iteration 33, current learner extra_tree\n", "[flaml.automl.automl: 04-09 03:14:08] {3519} INFO - at 4.1s,\testimator extra_tree's best error=0.6443,\tbest estimator lgbm's best error=0.6206\n", "[flaml.automl.automl: 04-09 03:14:08] {3334} INFO - iteration 34, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:14:08] {3519} INFO - at 4.4s,\testimator lgbm's best error=0.6206,\tbest estimator lgbm's best error=0.6206\n", "[flaml.automl.automl: 04-09 03:14:08] {3334} INFO - iteration 35, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:14:09] {3519} INFO - at 4.7s,\testimator lgbm's best error=0.6206,\tbest estimator lgbm's best error=0.6206\n", "[flaml.automl.automl: 04-09 03:14:09] {3334} INFO - iteration 36, current learner extra_tree\n", "[flaml.automl.automl: 04-09 03:14:09] {3519} INFO - at 4.8s,\testimator extra_tree's best error=0.6416,\tbest estimator lgbm's best error=0.6206\n", "[flaml.automl.automl: 04-09 03:14:09] {3334} INFO - iteration 37, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:14:09] {3519} INFO - at 5.3s,\testimator lgbm's best error=0.6185,\tbest estimator lgbm's best error=0.6185\n", "[flaml.automl.automl: 04-09 03:14:09] {3334} INFO - iteration 38, current learner rf\n", "[flaml.automl.automl: 04-09 03:14:09] {3519} INFO - at 5.4s,\testimator rf's best error=0.6458,\tbest estimator lgbm's best error=0.6185\n", "[flaml.automl.automl: 04-09 03:14:09] {3334} INFO - iteration 39, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:14:10] {3519} INFO - at 6.0s,\testimator lgbm's best error=0.6156,\tbest estimator lgbm's best error=0.6156\n", "[flaml.automl.automl: 04-09 03:14:10] {3334} INFO - iteration 40, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:14:10] {3519} INFO - at 6.4s,\testimator lgbm's best error=0.6156,\tbest estimator lgbm's best error=0.6156\n", "[flaml.automl.automl: 04-09 03:14:10] {3334} INFO - iteration 41, current learner rf\n", "[flaml.automl.automl: 04-09 03:14:10] {3519} INFO - at 6.6s,\testimator rf's best error=0.6458,\tbest estimator lgbm's best error=0.6156\n", "[flaml.automl.automl: 04-09 03:14:10] {3334} INFO - iteration 42, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:14:11] {3519} INFO - at 7.1s,\testimator lgbm's best error=0.6156,\tbest estimator lgbm's best error=0.6156\n", "[flaml.automl.automl: 04-09 03:14:11] {3334} INFO - iteration 43, current learner rf\n", "[flaml.automl.automl: 04-09 03:14:11] {3519} INFO - at 7.3s,\testimator rf's best error=0.6425,\tbest estimator lgbm's best error=0.6156\n", "[flaml.automl.automl: 04-09 03:14:11] {3334} INFO - iteration 44, current learner extra_tree\n", "[flaml.automl.automl: 04-09 03:14:11] {3519} INFO - at 7.4s,\testimator extra_tree's best error=0.6416,\tbest estimator lgbm's best error=0.6156\n", "[flaml.automl.automl: 04-09 03:14:11] {3334} INFO - iteration 45, current learner rf\n", "[flaml.automl.automl: 04-09 03:14:11] {3519} INFO - at 7.6s,\testimator rf's best error=0.6384,\tbest estimator lgbm's best error=0.6156\n", "[flaml.automl.automl: 04-09 03:14:11] {3334} INFO - iteration 46, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:14:12] {3519} INFO - at 8.1s,\testimator lgbm's best error=0.6156,\tbest estimator lgbm's best error=0.6156\n", "[flaml.automl.automl: 04-09 03:14:12] {3334} INFO - iteration 47, current learner rf\n", "[flaml.automl.automl: 04-09 03:14:12] {3519} INFO - at 8.3s,\testimator rf's best error=0.6384,\tbest estimator lgbm's best error=0.6156\n", "[flaml.automl.automl: 04-09 03:14:12] {3334} INFO - iteration 48, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:14:13] {3519} INFO - at 9.0s,\testimator lgbm's best error=0.6156,\tbest estimator lgbm's best error=0.6156\n", "[flaml.automl.automl: 04-09 03:14:13] {3334} INFO - iteration 49, current learner xgb_limitdepth\n", "[flaml.automl.automl: 04-09 03:14:13] {3519} INFO - at 9.1s,\testimator xgb_limitdepth's best error=0.6682,\tbest estimator lgbm's best error=0.6156\n", "[flaml.automl.automl: 04-09 03:14:13] {3334} INFO - iteration 50, current learner xgb_limitdepth\n", "[flaml.automl.automl: 04-09 03:14:13] {3519} INFO - at 9.2s,\testimator xgb_limitdepth's best error=0.6682,\tbest estimator lgbm's best error=0.6156\n", "[flaml.automl.automl: 04-09 03:14:13] {3334} INFO - iteration 51, current learner xgb_limitdepth\n", "[flaml.automl.automl: 04-09 03:14:13] {3519} INFO - at 9.3s,\testimator xgb_limitdepth's best error=0.6542,\tbest estimator lgbm's best error=0.6156\n", "[flaml.automl.automl: 04-09 03:14:13] {3334} INFO - iteration 52, current learner xgboost\n", "[flaml.automl.automl: 04-09 03:14:13] {3519} INFO - at 9.3s,\testimator xgboost's best error=0.6486,\tbest estimator lgbm's best error=0.6156\n", "[flaml.automl.automl: 04-09 03:14:13] {3334} INFO - iteration 53, current learner rf\n", "[flaml.automl.automl: 04-09 03:14:13] {3519} INFO - at 9.4s,\testimator rf's best error=0.6384,\tbest estimator lgbm's best error=0.6156\n", "[flaml.automl.automl: 04-09 03:14:13] {3334} INFO - iteration 54, current learner lgbm\n", "[flaml.automl.automl: 04-09 03:14:14] {3519} INFO - at 9.8s,\testimator lgbm's best error=0.6156,\tbest estimator lgbm's best error=0.6156\n", "[flaml.automl.automl: 04-09 03:14:14] {3334} INFO - iteration 55, current learner xgb_limitdepth\n", "[flaml.automl.automl: 04-09 03:14:14] {3519} INFO - at 10.0s,\testimator xgb_limitdepth's best error=0.6496,\tbest estimator lgbm's best error=0.6156\n", "[flaml.automl.automl: 04-09 03:14:14] {3783} INFO - retrain lgbm for 0.3s\n", "[flaml.automl.automl: 04-09 03:14:14] {3790} INFO - retrained model: LGBMClassifier(colsample_bytree=0.9031374907114736,\n", " learning_rate=0.3525398690474661, max_bin=1023,\n", " min_child_samples=4, n_estimators=22, num_leaves=69,\n", " reg_alpha=0.0060777294606297145, reg_lambda=37.65858370595088,\n", " verbose=-1)\n", "[flaml.automl.automl: 04-09 03:14:14] {3034} INFO - fit succeeded\n", "[flaml.automl.automl: 04-09 03:14:14] {3035} INFO - Time taken to find the best model: 5.982900142669678\n" ] } ], "source": [ "automl = AutoML()\n", "settings = {\n", " \"time_budget\": 10, # total running time in seconds\n", " \"metric\": custom_metric, # pass the custom metric funtion here\n", " \"task\": 'classification', # task type\n", " \"log_file_name\": 'airlines_experiment_custom_metric.log', # flaml log file\n", "}\n", "\n", "automl.fit(X_train=X_train, y_train=y_train, **settings)" ] } ], "metadata": { "description": null, "kernelspec": { "display_name": "Synapse PySpark", "name": "synapse_pyspark" }, "language_info": { "name": "python" }, "save_output": true, "synapse_widget": { "state": {}, "version": "0.1" } }, "nbformat": 4, "nbformat_minor": 2 }