{ "cells": [ { "cell_type": "markdown", "id": "26efd514", "metadata": {}, "source": [ "## Example for CESM1 " ] }, { "cell_type": "markdown", "id": "dd3f8e4a-6a62-4434-a7aa-8cf913527976", "metadata": {}, "source": [ "Reference: \n", "- GitHub: https://github.com/ncar/cesm-lens-aws/ \n", "- Data/Variables Information: https://ncar.github.io/cesm-lens-aws/#data-catalog \n", "- Reproduce CESM-LENS: http://gallery.pangeo.io/repos/NCAR/cesm-lens-aws/notebooks/kay-et-al-2015.v3.html " ] }, { "cell_type": "markdown", "id": "insured-finnish", "metadata": {}, "source": [ "**Step 0: load necessary packages and define parameters (no need to change)**" ] }, { "cell_type": "code", "execution_count": 1, "id": "a851a64a-c5db-4fb8-b226-d249cc2679bf", "metadata": { "scrolled": false }, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "/glade/work/zhonghua/miniconda3/envs/aws_urban/lib/python3.8/site-packages/xgboost/compat.py:31: FutureWarning: pandas.Int64Index is deprecated and will be removed from pandas in a future version. Use pandas.Index with the appropriate dtype instead.\n", " from pandas import MultiIndex, Int64Index\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\n", "--> The keys in the returned dictionary of datasets are constructed as follows:\n", "\t'component.experiment.frequency'\n" ] }, { "data": { "text/html": [ "\n", "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "\n", "
\n", " \n", " 100.00% [2/2 00:01<00:00]\n", "
\n", " " ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "CPU times: user 51.6 s, sys: 35.5 s, total: 1min 27s\n", "Wall time: 59.9 s\n" ] } ], "source": [ "%%time\n", "# Display output of plots directly in Notebook\n", "%matplotlib inline\n", "import matplotlib.pyplot as plt\n", "import pandas as pd\n", "import json\n", "from flaml import AutoML\n", "from sklearn.metrics import mean_squared_error, r2_score\n", "import warnings\n", "warnings.filterwarnings(\"ignore\")\n", "import util\n", "\n", "with open(\"./config_cesm1.json\",'r') as load_f:\n", "# param = json.loads(json.load(load_f))\n", " param = json.load(load_f)\n", " \n", " model = param[\"model\"] # cesm1\n", " city_loc = param[\"city_loc\"] # {\"lat\": 40.0150, \"lon\": -105.2705}\n", " l_component = param[\"l_component\"]\n", " a_component = param[\"a_component\"]\n", " experiment = param[\"experiment\"]\n", " frequency = param[\"frequency\"]\n", " cam_ls = param[\"cam_ls\"]\n", " clm_ls = param[\"clm_ls\"]\n", " time = slice(param[\"time_start\"],param[\"time_end\"])\n", " member_id = param[\"member_id\"]\n", " estimator_list = param[\"estimator_list\"]\n", " time_budget = param[\"time_budget\"]\n", " features = param[\"features\"]\n", " label = param[\"label\"]\n", " clm_var_mask = param[\"label\"][0]\n", " \n", "# get a dataset\n", "ds = util.get_data(model, city_loc, experiment, frequency, member_id, time, cam_ls, clm_ls)\n", "\n", "# create a dataframe\n", "ds['time'] = ds.indexes['time'].to_datetimeindex()\n", "df = ds.to_dataframe().reset_index().dropna()\n", "\n", "if \"PRSN\" in features:\n", " df[\"PRSN\"] = df[\"PRECSC\"] + df[\"PRECSL\"]\n", " \n", "# setup for automl\n", "automl = AutoML()\n", "automl_settings = {\n", " \"time_budget\": time_budget, # in seconds\n", " \"metric\": 'r2',\n", " \"task\": 'regression',\n", " \"estimator_list\":estimator_list,\n", "}" ] }, { "cell_type": "markdown", "id": "productive-negative", "metadata": {}, "source": [ "### Step 1: data analysis" ] }, { "cell_type": "markdown", "id": "mechanical-metabolism", "metadata": {}, "source": [ "**xarray.Dataset**" ] }, { "cell_type": "code", "execution_count": 2, "id": "square-amateur", "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "
<xarray.Dataset>\n",
       "Dimensions:     (member_id: 1, time: 7299)\n",
       "Coordinates:\n",
       "  * member_id   (member_id) int64 2\n",
       "    lat         float64 40.05\n",
       "    lon         float64 255.0\n",
       "  * time        (time) datetime64[ns] 2081-01-02T12:00:00 ... 2100-12-31T12:0...\n",
       "Data variables:\n",
       "    TREFHT      (member_id, time) float32 255.8 266.8 271.1 ... 277.6 276.1\n",
       "    TREFHTMX    (member_id, time) float32 269.3 278.4 281.2 ... 283.9 277.5\n",
       "    FLNS        (member_id, time) float32 77.09 67.4 77.49 ... 64.89 37.45 38.35\n",
       "    FSNS        (member_id, time) float32 83.31 88.83 90.25 ... 67.98 80.27\n",
       "    PRECSC      (member_id, time) float32 0.0 0.0 0.0 0.0 ... 0.0 0.0 8.927e-12\n",
       "    PRECSL      (member_id, time) float32 4.887e-10 6.665e-10 ... 9.722e-10\n",
       "    PRECT       (member_id, time) float32 4.887e-10 6.665e-10 ... 2.32e-08\n",
       "    QBOT        (member_id, time) float32 0.000913 0.001889 ... 0.005432 0.00492\n",
       "    UBOT        (member_id, time) float32 5.461 4.815 4.506 ... 2.865 3.255\n",
       "    VBOT        (member_id, time) float32 1.27 3.189 3.691 ... 1.215 0.8704\n",
       "    TREFMXAV_U  (member_id, time) float32 259.6 270.0 279.8 ... 284.9 285.3\n",
       "Attributes: (12/14)\n",
       "    intake_esm_varname:        FLNS\\nFSNS\\nPRECSC\\nPRECSL\\nPRECT\\nQBOT\\nTREFH...\n",
       "    topography_file:           /scratch/p/pjk/mudryk/cesm1_1_2_LENS/inputdata...\n",
       "    title:                     UNSET\n",
       "    Version:                   $Name$\n",
       "    NCO:                       4.4.2\n",
       "    host:                      tcs-f02n07\n",
       "    ...                        ...\n",
       "    important_note:            This data is part of the project 'Blind Evalua...\n",
       "    initial_file:              b.e11.B20TRC5CNBDRD.f09_g16.105.cam.i.2006-01-...\n",
       "    source:                    CAM\n",
       "    revision_Id:               $Id$\n",
       "    logname:                   mudryk\n",
       "    intake_esm_dataset_key:    atm.RCP85.daily
" ], "text/plain": [ "\n", "Dimensions: (member_id: 1, time: 7299)\n", "Coordinates:\n", " * member_id (member_id) int64 2\n", " lat float64 40.05\n", " lon float64 255.0\n", " * time (time) datetime64[ns] 2081-01-02T12:00:00 ... 2100-12-31T12:0...\n", "Data variables:\n", " TREFHT (member_id, time) float32 255.8 266.8 271.1 ... 277.6 276.1\n", " TREFHTMX (member_id, time) float32 269.3 278.4 281.2 ... 283.9 277.5\n", " FLNS (member_id, time) float32 77.09 67.4 77.49 ... 64.89 37.45 38.35\n", " FSNS (member_id, time) float32 83.31 88.83 90.25 ... 67.98 80.27\n", " PRECSC (member_id, time) float32 0.0 0.0 0.0 0.0 ... 0.0 0.0 8.927e-12\n", " PRECSL (member_id, time) float32 4.887e-10 6.665e-10 ... 9.722e-10\n", " PRECT (member_id, time) float32 4.887e-10 6.665e-10 ... 2.32e-08\n", " QBOT (member_id, time) float32 0.000913 0.001889 ... 0.005432 0.00492\n", " UBOT (member_id, time) float32 5.461 4.815 4.506 ... 2.865 3.255\n", " VBOT (member_id, time) float32 1.27 3.189 3.691 ... 1.215 0.8704\n", " TREFMXAV_U (member_id, time) float32 259.6 270.0 279.8 ... 284.9 285.3\n", "Attributes: (12/14)\n", " intake_esm_varname: FLNS\\nFSNS\\nPRECSC\\nPRECSL\\nPRECT\\nQBOT\\nTREFH...\n", " topography_file: /scratch/p/pjk/mudryk/cesm1_1_2_LENS/inputdata...\n", " title: UNSET\n", " Version: $Name$\n", " NCO: 4.4.2\n", " host: tcs-f02n07\n", " ... ...\n", " important_note: This data is part of the project 'Blind Evalua...\n", " initial_file: b.e11.B20TRC5CNBDRD.f09_g16.105.cam.i.2006-01-...\n", " source: CAM\n", " revision_Id: $Id$\n", " logname: mudryk\n", " intake_esm_dataset_key: atm.RCP85.daily" ] }, "execution_count": 2, "metadata": {}, "output_type": "execute_result" } ], "source": [ "ds" ] }, { "cell_type": "markdown", "id": "breeding-junior", "metadata": {}, "source": [ "**pandas dataframe**" ] }, { "cell_type": "code", "execution_count": 3, "id": "86c130d8", "metadata": {}, "outputs": [ { "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", " \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", "
member_idtimeTREFHTTREFHTMXFLNSFSNSPRECSCPRECSLPRECTQBOTUBOTVBOTlatlonTREFMXAV_UPRSN
022081-01-02 12:00:00255.849716269.32785077.08580083.3110660.04.886532e-104.886532e-100.0009135.4612931.27023640.052357255.0259.6419684.886532e-10
122081-01-03 12:00:00266.790833278.39654567.40265788.8311920.06.664702e-106.665211e-100.0018894.8145453.18950040.052357255.0269.9728096.664702e-10
222081-01-04 12:00:00271.122040281.18194677.49319590.2474820.08.030515e-142.538048e-110.0033314.5064593.69069840.052357255.0279.8288278.030515e-14
322081-01-05 12:00:00276.329895281.84454364.78974993.3431930.01.896001e-202.508245e-090.0042095.1629414.96315740.052357255.0282.6741031.896001e-20
422081-01-06 12:00:00275.347229280.98046969.64704180.9937060.05.442079e-177.844814e-090.0040144.4784844.27258640.052357255.0283.5187075.442079e-17
\n", "
" ], "text/plain": [ " member_id time TREFHT TREFHTMX FLNS \\\n", "0 2 2081-01-02 12:00:00 255.849716 269.327850 77.085800 \n", "1 2 2081-01-03 12:00:00 266.790833 278.396545 67.402657 \n", "2 2 2081-01-04 12:00:00 271.122040 281.181946 77.493195 \n", "3 2 2081-01-05 12:00:00 276.329895 281.844543 64.789749 \n", "4 2 2081-01-06 12:00:00 275.347229 280.980469 69.647041 \n", "\n", " FSNS PRECSC PRECSL PRECT QBOT UBOT \\\n", "0 83.311066 0.0 4.886532e-10 4.886532e-10 0.000913 5.461293 \n", "1 88.831192 0.0 6.664702e-10 6.665211e-10 0.001889 4.814545 \n", "2 90.247482 0.0 8.030515e-14 2.538048e-11 0.003331 4.506459 \n", "3 93.343193 0.0 1.896001e-20 2.508245e-09 0.004209 5.162941 \n", "4 80.993706 0.0 5.442079e-17 7.844814e-09 0.004014 4.478484 \n", "\n", " VBOT lat lon TREFMXAV_U PRSN \n", "0 1.270236 40.052357 255.0 259.641968 4.886532e-10 \n", "1 3.189500 40.052357 255.0 269.972809 6.664702e-10 \n", "2 3.690698 40.052357 255.0 279.828827 8.030515e-14 \n", "3 4.963157 40.052357 255.0 282.674103 1.896001e-20 \n", "4 4.272586 40.052357 255.0 283.518707 5.442079e-17 " ] }, "execution_count": 3, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.head()" ] }, { "cell_type": "markdown", "id": "designing-asbestos", "metadata": {}, "source": [ "**data visualization**" ] }, { "cell_type": "code", "execution_count": 4, "id": "standing-infrared", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "[]" ] }, "execution_count": 4, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "ds[\"TREFMXAV_U\"].plot()" ] }, { "cell_type": "markdown", "id": "confused-affect", "metadata": {}, "source": [ "### Step 2: automated machine learning" ] }, { "cell_type": "markdown", "id": "defensive-footage", "metadata": {}, "source": [ "**train a model (emulator)**" ] }, { "cell_type": "code", "execution_count": 5, "id": "clear-driver", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "XGBRegressor(base_score=0.5, booster='gbtree',\n", " colsample_bylevel=0.8981353296453468, colsample_bynode=1,\n", " colsample_bytree=0.8589079860800738, gamma=0, gpu_id=-1,\n", " grow_policy='lossguide', importance_type='gain',\n", " interaction_constraints='', learning_rate=0.05211228610238813,\n", " max_delta_step=0, max_depth=0, max_leaves=29,\n", " min_child_weight=45.3846760978798, missing=nan,\n", " monotone_constraints='()', n_estimators=299, n_jobs=-1,\n", " num_parallel_tree=1, random_state=0, reg_alpha=0.01917865165509354,\n", " reg_lambda=2.2296607355174927, scale_pos_weight=1,\n", " subsample=0.9982731696185565, tree_method='hist',\n", " use_label_encoder=False, validate_parameters=1, verbosity=0)\n", "CPU times: user 3min 35s, sys: 2.67 s, total: 3min 38s\n", "Wall time: 15.7 s\n" ] } ], "source": [ "%%time\n", "# assume that we want to split the data into training data and testing data\n", "# let's use first 95% for training, and the remaining for testing \n", "idx = df.shape[0]\n", "train = df.iloc[:int(0.95*idx),:]\n", "test = df.iloc[int(0.95*idx):,:]\n", "(X_train, y_train) = (train[features], train[label].values)\n", "(X_test, y_test) = (test[features], test[label].values)\n", "\n", "# train the model\n", "automl.fit(X_train=X_train, y_train=y_train,\n", " **automl_settings, verbose=-1)\n", "print(automl.model.estimator)" ] }, { "cell_type": "markdown", "id": "secure-worst", "metadata": {}, "source": [ "**apply and test the machine learning model** \n", "use `automl.predict(X)` to apply the model" ] }, { "cell_type": "code", "execution_count": 6, "id": "sunrise-removal", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "model performance using training data:\n", "root mean square error: 1.7566193\n", "r2: 0.9681344385429989 \n", "\n", "model performance using testing data:\n", "root mean square error: 2.3287773\n", "r2: 0.9384653117109892\n" ] } ], "source": [ "# training data\n", "print(\"model performance using training data:\")\n", "y_pred = automl.predict(X_train)\n", "print(\"root mean square error:\", \n", " mean_squared_error(y_true=y_train, y_pred=y_pred, squared=False))\n", "print(\"r2:\", r2_score(y_true=y_train, y_pred=y_pred),\"\\n\")\n", "import pandas as pd\n", "d_train = {\"time\":train[\"time\"],\"y_train\":y_train.reshape(-1),\"y_pred\":y_pred.reshape(-1)}\n", "df_train = pd.DataFrame(d_train).set_index(\"time\")\n", "\n", "# testing data\n", "print(\"model performance using testing data:\")\n", "y_pred = automl.predict(X_test)\n", "print(\"root mean square error:\", \n", " mean_squared_error(y_true=y_test, y_pred=y_pred, squared=False))\n", "print(\"r2:\", r2_score(y_true=y_test, y_pred=y_pred))\n", "d_test = {\"time\":test[\"time\"],\"y_test\":y_test.reshape(-1),\"y_pred\":y_pred.reshape(-1)}\n", "df_test = pd.DataFrame(d_test).set_index(\"time\")" ] }, { "cell_type": "markdown", "id": "technical-summary", "metadata": {}, "source": [ "**visualization**" ] }, { "cell_type": "code", "execution_count": 7, "id": "supported-pension", "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "fig, (ax1,ax2) = plt.subplots(1,2,figsize=(12,3))\n", "fig.suptitle('emulator evaluation')\n", "df_train[\"y_train\"].plot(label=\"reference\",c=\"k\",ax=ax1)\n", "df_train[\"y_pred\"].plot(label=\"prediction\",c=\"r\",ax=ax1)\n", "ax1.set_title(\"training data\")\n", "ax1.set_ylabel(\"urban daily maximum temperature, K\")\n", "\n", "df_test[\"y_test\"].plot(label=\"reference\",c=\"k\",ax=ax2)\n", "df_test[\"y_pred\"].plot(label=\"prediction\",c=\"r\",ax=ax2)\n", "ax2.set_title(\"testing data\")\n", "plt.legend()\n", "plt.show()" ] } ], "metadata": { "kernelspec": { "display_name": "Python 3 (ipykernel)", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.8.0" } }, "nbformat": 4, "nbformat_minor": 5 }