{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Identify money laundering with anomaly detection\n", "\n", "This notebook demonstrates how DataRobot performs outlier detection with a use case that prevents money laundering: the process of hiding illicitly obtained money. The notebook uses a historical money transactions dataset and trains anomaly detection models to detect outliers. \n", "\n", "In the sample dataset, fraudulent transactions are identified in the `SAR` column; however, in this use case, that information will not be used to train the model. This is because, in most cases, money laundering goes undetected and [anomaly detection in DataRobot](https://docs.datarobot.com/en/docs/modeling/special-workflows/unsupervised/anomaly-detection.html) can help identify when it occurs. This notebook uses a small subset of the data to evaluate how well the unsupervised approach works, as you can compare the results to the data that's already labeled as fraud (the `SAR` column).\n", "\n", "## Requirements\n", "\n", "- Python version 3.7.3\n", "- DataRobot API version 2.19.0\n", "\n", "Note that small adjustments may be required depending on the Python and DataRobot API versions you are using.\n", "\n", "Full documentation of the Python package can be found [here](https://datarobot-public-api-client.readthedocs-hosted.com)." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Import libraries" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "import datarobot as dr\n", "import matplotlib.pyplot as plt\n", "import numpy as np\n", "import pandas as pd\n", "from sklearn.metrics import confusion_matrix\n", "from sklearn.model_selection import train_test_split\n", "\n", "%matplotlib inline\n", "import time\n", "\n", "import seaborn as sns\n", "from sklearn.metrics import f1_score" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Upload data\n", "\n", "In the sample dataset, [available for download](aml.csv), `SAR` is the target feature. For the purposes of this notebook, the target is not used for training, but instead to evaluate the accuracy of the anomaly detection models built in later steps. This is because you do not want to cause [target leakage](https://docs.datarobot.com/en/docs/glossary/index.html#target-leakage)." ] }, { "cell_type": "code", "execution_count": 2, "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", " \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", "
ALERTSARkycRiskScoreincometenureMonthscreditScorestatenbrPurchases90davgTxnSize90dtotalSpend90d...indCustReqRefund90dtotalRefundsToCust90dnbrPaymentsCashLike90dmaxRevolveLineindOwnsHomenbrInquiries1ynbrCollections3ynbrWebLogins90dnbrPointRed90dPEP
0103110300.05757PA10153.801538.00...145.8256000030610
1102107800.06715NY221.5934.98...167.400100001308700
210174000.013751MA757.64403.48...1450.69010000030600
310057700.01659NJ1429.52413.28...171.4308000150720
410159800.03709PA54115.776251.58...12731.3937000110810
\n", "

5 rows × 31 columns

\n", "
" ], "text/plain": [ " ALERT SAR kycRiskScore income tenureMonths creditScore state \\\n", "0 1 0 3 110300.0 5 757 PA \n", "1 1 0 2 107800.0 6 715 NY \n", "2 1 0 1 74000.0 13 751 MA \n", "3 1 0 0 57700.0 1 659 NJ \n", "4 1 0 1 59800.0 3 709 PA \n", "\n", " nbrPurchases90d avgTxnSize90d totalSpend90d ... indCustReqRefund90d \\\n", "0 10 153.80 1538.00 ... 1 \n", "1 22 1.59 34.98 ... 1 \n", "2 7 57.64 403.48 ... 1 \n", "3 14 29.52 413.28 ... 1 \n", "4 54 115.77 6251.58 ... 1 \n", "\n", " totalRefundsToCust90d nbrPaymentsCashLike90d maxRevolveLine indOwnsHome \\\n", "0 45.82 5 6000 0 \n", "1 67.40 0 10000 1 \n", "2 450.69 0 10000 0 \n", "3 71.43 0 8000 1 \n", "4 2731.39 3 7000 1 \n", "\n", " nbrInquiries1y nbrCollections3y nbrWebLogins90d nbrPointRed90d PEP \n", "0 3 0 6 1 0 \n", "1 3 0 87 0 0 \n", "2 3 0 6 0 0 \n", "3 5 0 7 2 0 \n", "4 1 0 8 1 0 \n", "\n", "[5 rows x 31 columns]" ] }, "execution_count": 2, "metadata": {}, "output_type": "execute_result" } ], "source": [ "data_path = \"https://docs.datarobot.com/en/docs/api/guide/common-case/aml.csv\"\n", "\n", "df = pd.read_csv(data_path) # Add your dataset here\n", "df.head()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Connect to DataRobot\n", "\n", "To use this notebook, you first need to establish a connection between your machine and the DataRobot instance. Read more about different options for [connecting to DataRobot](https://docs.datarobot.com/en/docs/api/api-quickstart/api-qs.html)." ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "# If the config file is not in the default location described in the API Quickstart guide, '~/.config/datarobot/drconfig.yaml', then you will need to call\n", "# dr.Client(config_path='path-to-drconfig.yaml')" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Clean and split the data\n", "\n", "Use the snippets below to split the data into a training set and an external test set. The second line removes the target from the training dataset." ] }, { "cell_type": "code", "execution_count": 14, "metadata": {}, "outputs": [], "source": [ "train, external_test = train_test_split(df, test_size=0.15)\n", "\n", "train = train.drop(\"SAR\", axis=1)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Create a project\n", "\n", "Once connected to DataRobot, you can create a project using the `dr.Project.create` and `dr.Project.set_target()` methods. Include `unsupervised_mode = True` to enable Autopilot without a target feature." ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "project = dr.Project.create(project_name=\"Outlier Detection\", sourcedata=train)\n", "\n", "project.set_target(unsupervised_mode=True, worker_count=-1)\n", "project.wait_for_autopilot()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Unlock holdout\n", "\n", "DataRobot uses [Synthetic AUC](https://docs.datarobot.com/en/docs/modeling/special-workflows/unsupervised/anomaly-detection.html#synthetic-auc-metric) to evaluate model performance. Synthetic AUC is a useful metric to see if a model can detect anomalies, but it is not an actual representation of accuracy. Use the snippet below to train all models to 100% and evaluate them all against the labeled test data. The top-performing model will be bound to `best_model`." ] }, { "cell_type": "code", "execution_count": 144, "metadata": {}, "outputs": [], "source": [ "project.unlock_holdout()\n", "\n", "for model in project.get_models():\n", " model.train(100)\n", "\n", "# Sleep while waiting for the modeling jobs to complete\n", "while len(project.get_model_jobs()) != 0:\n", " time.sleep(5)\n", "\n", "# Overwrite best_model with the newly trained top-performing model\n", "models = []\n", "for model in project.get_models():\n", " if model.sample_pct == 100:\n", " models.append(model)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Upload the test dataset\n", "\n", "When working with anomaly detection, note that for training data, the most anomalous value will have a score of 1. When you score data, if an observation is even more anomalous than the training data's values, it will have a score greater than 1.\n", "\n", "Anomalous values are *normalized*, which requires applying business logic to get an accurate result. Experts suggest that you can expect 10% of transactions to be fraudulent; therefore, you can assume that 10% of the most anomalous values represent fraud." ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "# Upload external test data for predictions\n", "dataset_from_path = project.upload_dataset(external_test)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Make predictions\n", "\n", "Use the function below to test predictions with the external test dataset. This function includes a prediction threshold based on quantile." ] }, { "cell_type": "code", "execution_count": 146, "metadata": {}, "outputs": [], "source": [ "def get_predictions(model, q=95):\n", " predict_job = model.request_predictions(dataset_from_path.id)\n", " predictions = predict_job.get_result_when_complete()\n", " p = predictions.join(external_test.reset_index())[[\"prediction\", \"SAR\"]]\n", " p_threshold = np.quantile(p[\"prediction\"], 0.95)\n", "\n", " p[\"prediction_binary\"] = p[\"prediction\"].apply(lambda x: 1 if x >= p_threshold else 0)\n", "\n", " return p" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Plot a confusion matrix for each model\n", "\n", "The function below plots a [confusion matrix](https://docs.datarobot.com/en/docs/modeling/analyze-models/evaluate/roc-curve-tab/confusion-matrix.html) for each model based on the prediction threshold. You can experiment with both accuracy metrics and prediction thresholds and analyze the new results." ] }, { "cell_type": "code", "execution_count": 153, "metadata": { "scrolled": true }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "F1 Score: 0.53007208859048 Model('Mahalanobis Distance Ranked Anomaly Detection with PCA and Calibration')\n", "F1 Score: 0.5251254789966957 Model('Mahalanobis Distance Ranked Anomaly Detection with PCA and Calibration')\n", "F1 Score: 0.6240576708723841 Model('Isolation Forest Anomaly Detection with Calibration')\n", "F1 Score: 0.6685771572164438 Model('Isolation Forest Anomaly Detection with Calibration')\n", "F1 Score: 0.6240576708723841 Model('Anomaly Detection Blender')\n", "F1 Score: 0.6290042804661685 Model('Anomaly Detection Blender')\n", "F1 Score: 0.6388974996537373 Model('Anomaly Detection with Supervised Learning (XGB) and Calibration')\n", "F1 Score: 0.5993246229034619 Model('Anomaly Detection with Supervised Learning (XGB) and Calibration')\n", "F1 Score: 0.6537373284350906 Model('Anomaly Detection with Supervised Learning (XGB) and Calibration')\n", "F1 Score: 0.6042712324972463 Model('Anomaly Detection with Supervised Learning (XGB) and Calibration')\n", "F1 Score: 0.09937560287698714 Model('One-Class SVM Anomaly Detection with Calibration')\n", "F1 Score: 0.6092178420910308 Model('Local Outlier Factor Anomaly Detection with Calibration')\n", "F1 Score: 0.5943780133096775 Model('Local Outlier Factor Anomaly Detection with Calibration')\n", "F1 Score: 0.6317668260899703 Model('Double Median Absolute Deviation Anomaly Detection with Calibration')\n", "F1 Score: 0.6326081632002365 Model('Double Median Absolute Deviation Anomaly Detection with Calibration')\n", "F1 Score: 0.09587568850476916 Model('One-Class SVM Anomaly Detection with Calibration')\n" ] }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXkAAAD/CAYAAAAUnaZMAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjAsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+17YcXAAAYPUlEQVR4nO3deZRU1bXH8W8DCkJrwFkQR8xeUUEUpygo+NTEIevFKeZFMSZoFMgTIqIIigRFeSRGQNEgcYoTiqjPViPqExSNsyIqup2IIoNxAJGpobvq/XFvY1lW01U03Zc+9/dZ666yzrl17mlNdu/e99S5ZdlsFhERCVOzpCcgIiINR0FeRCRgCvIiIgFTkBcRCZiCvIhIwBTkRUQC1iLJi6/54iOt35TvKN/x8KSnIBupylXzyurz+VLizSZb71ava21MEg3yIiKNJlOd9AwSoSAvIumQzSQ9g0QoyItIOmQU5EVEgpVVJi8iErDqqga/hJldDBzn7t1z2vYDRgP7A2uAGcAF7j4v55yFwPZ5w93p7qfH/VsB44FjgDJgCnC+uy+ra05aQiki6ZCpLv5YD2bWDxiV19YR+D9gAdAdOA7YFnjczDaLz9maKMAfCeyQc/TPGeo+oFN8zgnx68Ri5qVMXkTSoYHKNWbWnijg9gI8r/s0YCVwlrtXxeefAXxCFPSfALoAWeB5d19RYPxDgJ7AXu4+J247G3jCzC5290/WNT9l8iKSDplM8UdpugHfEAXrF/P6pgKn1gT4WM16/S3j1y7AvwoF+FgP4N81AT42Mx6nR12TUyYvIqlQyo1XM2sLtC3QtcTdl+Q2uHsFUBF/jry+94H388YYSpTdPxW/7wysMrMHgYOAz4BbgGvdPQN0AD7NG3e1mX0BdKzrZ1EmLyLpUFomPxCYW+AYWJ8pmNn5QF/gInf/PG7emyirvxP4CTCJqLb/x7i/NVBZYLhKoFVd11QmLyLpUL2mlLPHArcWaF9SoK1OZtYMuAq4EBjp7tfmdPcANs1ZKTPbzLYALjWzEURZf8sCw7YE6lxdoyAvIulQQrkmLsmsV0DPZ2atgDuIVsX0d/fr8661Glid97HZwGbANsA8oH3emJsCW5NXxilE5RoRSYeGu/FaqziDnwocD5yUH+DNrKWZLTCzC/I+eiDwpbsvAp4BtrfvFvwPi19n1jUHZfIikg7JfON1IHAsUR3+BTPL/cLT1+6+0sweAoaZ2VxgFnAUUVmnJvC/CDwH3G1m5xJl+BOB2919fl0TUCYvIumQQCYPnB6/3gAszDtOi/vOA64F/gTMIfrF8Ad3nwDg7lngROBDohU5U4nW1/ctZgJl2WxyW7prP3nJp/3kpTb13U9+1ayHi443rboer/3kRUSaFO1CKSISMO1CKSISMD0ZSkQkYMrkRUQCppq8iEjAGuGhIRsjBXkRSQdl8iIi4cpmdeNVRCRcyuRFRAKm1TUiIgFTJi8iEjCtrhERCZjKNSIiAVO5RkQkYAryIiIBU7lGRCRguvEqIhIwlWtERAKmco2ISMCUyYuIBExBXkQkYNls0jNIhIK8iKRDlVbXiIiESzdeRUQCppq8iEjAVJMXEQmYMnkRkYApyIuIhCtbrQd5i4iES5m8iEjAGmEJpZldDBzn7t1z2nYBrgMOA1YCtwLD3L0q55z+wCBgB2AWcJ67v1zKGLVpVs+fSUSkachkiz/Wg5n1A0bltW0KPA5kgUOAs4E+wB9zzjkTGANcAnQDHJhmZtsWO8a6KMiLSDpkMsUfJTCz9mZWQRSoPa/7ZGBn4Ax3f8vdHwIuAgaY2WbxOUOBCe5+l7vPIQrgS4FzShijVgryjWD22+9y5u8vBODDuR/Tu+8gTj93EJf/+Tqqc24GfbV4Ccee2ofKytUArKqsZODQKzij7wX0HXQpXy1eksj8pfFss81WfPDBi9gPd+f2v0/g8cfv5fHH78X9n9z+9wlJT69pq64u/ihNN+AboAvwYl5fD2CWuy/OaZsBtAH2M7PtgD2A6TWd7l4NzCQqzdQ5Rl2Tq7MmH/+pcEp8wY5AK2AZ8Gl8ofuLqQul1c13TqHisafYrFVLAMZNvI0B55zJ/l07M+yKq5n+7AscefihPPfiq1xzw818+dW3/x3veeAR9th9F/r3OZ1Hn5zBxNsmc/HAc5P6UaSBtWjRggnXjWbVylUA9D6jPwBt2/6Ax6fdw+ALi/rrXGpTQoZuZm2BtgW6lrj7d7Itd68AKuLP5Z/fgShW5loQv3Ykqq9TyzkHFDnGOq0zkzez3YE5wETgR0R/QswDVgB7AzcDb8Y3BaSAju13YOyVl6x9f82oYezftTNr1qzhi68Ws9WW7QAoKyvjb+Ou4gdbbL723NfeeJvuB3UDoMfBB/DCy6837uSlUf3P6EuYNOkOFiz87Dvtwy89n+tvuIVFi/6d0MwCUVpNfiAwt8AxsMSrtgYq89pq3reK+6nlnFZFjrFOdWXy1xPVmPZ192/yO81sC2Ay0V3f4+u6WBod1as783P+T9u8eXMWLPqMswYMpbxNG3bdaUcADjnw+391LV+xgvLyNgC0ab0Zy5Yvb5xJS6Pr3fsUPv/iK5548mkGX9h/bfs222xFr16HcsFgZfH1VtrqmrFEK1jylVozXQm0zGureb+MbzP5QucsK3KMdaoryHcHDioU4AHcfamZDQGeretC8q3222/Ho/fcxH0PPcaY8Tdy5aUXFDyvTevWrFgR/W9g+YqVbF5e3pjTlEb061+fSjab5YgjurNPlz256aaxnHTyb/n5fx7D5Hv+l0xK13hvUCWsmolLMhviJtg8YN+8tvbx66fAJzltb+adU1OiqWuMdarrxutionrQuuwCKMUs0u8vHMHH8+YDUXberFnt/wn27bInzzwfLZWd+cLL7LfPXo0yR2l8Rx55MkcddQpHH/0L3pg9hz59BvLZZ59zxBHdmTZtet0DSJ2ymUzRxwb0DNA1rvHX6EV0o/Y1d/+cqFrSs6bTzJoT3Wx9upgx6ppAXZn8TcBtZnYZ0U3WT4lqQS2JfpP0BK4EJtV1IYn06f0Lho36C5u0aEGrVi0ZOaT2Et+pJxzHsCuupnffQWzSYhPGjLiwEWcqG4Mf/nA35s79pO4TpW7JbGvwIHAFcI+ZDSZaCnkV8Bd3Xx2fczUw3swceAkYDJTzbVwtZoxalWXXsf2mmZUBlwHnEy3XyfcNUT1+uLuX/OtvzRcfpXPvT6lV+Y6HJz0F2UhVrppXVp/PLx95WtHxps3wO9frWmZ2K9Ap7xuvnYAJRCsUFxMtWPlOzDSzQUQ3dbcCXgUGuPtrpYxRm3UG+ZwLbAp0JSrdtCa6ETAPeKOY3yS1UZCXfAryUpt6B/kR/1V8kB9xd72utTEpau+aOJC/1MBzERFpOOu5XUFTpw3KRCQd9IxXEZGAKZMXEQlXtkoPDRERCZcyeRGRgKkmLyISMGXyIiLhyirIi4gETDdeRUQCpkxeRCRgCvIiIuEqZp+uECnIi0g6KJMXEQmYgryISLiyVfoylIhIuNIZ4xXkRSQd9GUoEZGQKciLiARM5RoRkXCpXCMiErBslYK8iEi4VK4REQlXSp8ZoiAvIimhIC8iEi5l8iIiActWJT2DZCjIi0gqKJMXEQmYgryISMiyZUnPIBEK8iKSCsrkRUQCls0okxcRCVamesMHeTPrCUyvpXuuu+9mZlcBQwr0b+LuVfE4/YFBwA7ALOA8d395Q8yx2YYYRERkY5fNFH+U4J9EgTn3OBqoAq6Kz+kCTMo/LyfAnwmMAS4BugEOTDOzbev3E0eUyYtIKjREucbdVwOLat6b2SbAWOB+d58UN3cGKtx9UYEhAIYCE9z9rniMPsCHwDnA5fWdozJ5EUmFbLb4ox7OAzoCfwAws7bx+zmFTjaz7YA9yCn5uHs1MBM4rF4ziSmTF5FUKCWTj4Nz2wJdS9x9SS2faU2UlY9z9wVxc+f49TQzuwnYFJgBDHH3hUCHuP/TvOEWAAcUPeF1UCYvIqmQqS4r+gAGAnMLHAPXcYneQCtgXE7b3vHr18BJwO/ithnxL4XWcX9l3liV8Vj1pkxeRFKhxJr8WODWAu0Fs/hYb6Ja/Bc5bX8FJrv74vj9bDN7C5gH/JzoJitAy7yxWgLLSplwbRTkRSQVsiV84zUuyawroH+HmW0D/Bi4Mm+cLLA4r22+mX0J7AQ8ETe3B97MOa093y/hrBeVa0QkFRpoCWWNQ4Es8HRuo5mNM7PX89p2BbYG3nb3z4my+Z45/c2BHvljrS9l8iKSCpmG3btmX+Ajd1+e1z4F6Gdm44DriDL0ccBLwCPxOVcD483M4/bBQDnR2vp6UyYvIqmQzZYVfayHHYCv8hvd/VngeOBA4HVgKvAqcKy7Z+JzJhF9Eepy4BWgE3B0Xm1/vZVl67kotD7WfPFROh+fLrUq3/HwpKcgG6nKVfPqlYq/s8exRcebH73/aDAb3ahcIyKpoA3KREQC1sA1+Y2WgryIpMJ61tqbPAV5EUmFBG8/JkpBXkRSQeUaEZGAZXTjtfHtssfPkry8bISqMyl9EKc0OGXyIiIB041XEZGAKZMXEQlYShfXKMiLSDpUZ9K5VZeCvIikQlpv6SvIi0gqZFFNXkQkWJmUFuUV5EUkFTLK5EVEwqVyjYhIwKoV5EVEwqXVNSIiAVOQFxEJmGryIiIBS+lOwwryIpIOWkIpIhKw6qQnkBAFeRFJhUyZMnkRkWCldFcDBXkRSQctoRQRCZhW14iIBEzbGoiIBEyZvIhIwFSTFxEJWEOtrjEzA94t0HW2u//NzLoCY4EDgC+B8e7+55zPNwMuA84C2gHPAv3c/YMNMb90PtlWRFInU1b8UaIuwFJgh7zjTjPbGngSeB/YH7gUGGlmZ+d8fjjQFzgbOBioAqaZWav1/2m/pUxeRFKhAcs1nYF33H1RfoeZ/QFYDfR19yrgHTPrBAwBJplZS2AQcJG7Pxp/5pfAQuAU4Pb6Tk6ZvIikQnVZ8UeJugBzaunrAcyMA3yNGcBuZtYB6AqUA9NrOt19KfAacFjJMylAmbyIpEIpmbyZtQXaFuha4u5L8to6A++b2XNAJ+A94HJ3fxzowPd/ASyIXzsC7eN//rTAOR1LmHKtlMmLSCpkSjiAgcDcAsfA3DHNrA2wK9EvhGHAscArwGNmdiTQGqjMm0rN+1ZxP7Wco5q8iEixSlxdMxa4tUD7d7J4d19uZlsAq919ddz8qpntCVwIrARa5o1R835Z3F/TtjrvnGWlTbkwBXkRSYVSVs3EJZn8skxt5xYKxrOBnwEf8W1JpkZuiaZZTpvnnfN2sfNdF5VrRCQVSizXFMXMfmxm35jZAXldBxAF6WeA7maWm1D3At6LV+O8QbT8smfOmFsA+wFPlzCVWimTF5FUaKCHhrxCVKufZGb9ib7s1Bc4BDgImE9UtrnZzEYD3YDzgX4A7l5pZtcBV5rZonis0fHnpm6ICSrIi0gqNMTeNe6+xsx+ShSYpxLdgH0VOMrdXwcws6OB8UTLIhcBQ9z91pxhhgPNgRuBNsBM4Kc5Nf56Kctmk9tKv0O7vdK6j7/U4rPlRZVBJYWqVs+vV5gevfPpRcebIR/fEcx2ZsrkRSQV0ppRKsiLSCpkUhrmFeRFJBUa6MbrRk9BXkRSQfvJi4gETE+GEhEJmGryIiIBS2eIV5AXkZRQTV5EJGDVKc3lFeRFJBWUyYuIBEw3XkVEApbOEK8gLyIpoXKNiEjAdONVRCRgaa3J6/F/jWjfbp2ZUnHLd9pGjLqI3r/5xdr3I0dfzD+m38uUiluYUnELm29R3tjTlAS0aNGCW28Zz4yn7uf55x7m+OOPYp999uK5mRU8Pf0BJt14NWVlKf1e/gaSLeEIiTL5RtL3vN9y0qk/Y+Xy6OHsW27VjnF/vYrddt+ZD6+du/a8zvvsya9O+h2Lv9LDM9LktF+dyJdfLubM35zHllu245WXpvHa67O5YtQ1/OOxp/j7bddy3LFH8vAjTyQ91SZLmbw0qI/nzuPs3gPWvm/TpjV/GT2BqfdUrG0rKytj1912YszYETz42B2cetoJSUxVEnDf1Ie5bMSYte+rqqqYNest2m3ZFoDNNy9nzZo1SU0vCA3xIO+moM5M3sxmUuRfMO5+WL1nFKhHK55gx47t176f98l85n0yn15H9ljb1rrNZtxy411MvP42mjdvxpSHbmH2rLd55+33kpiyNKLly1cAUF7ehnsn38jwEWPIZrNcO24UQy8ewNKvlzLj6ecTnmXTllUmX6tHgUOBrYAP6zikHlauWMXfJt7OqpWrWL5sBc/NfIk997akpyWNZMcd2/PkE1O44877mDz5Qa65eiQ9jziRvTsfzu133MefxgxPeopNWjXZoo+Q1JnJu/tVZvY10dPIj3P3fzX4rFJqt067cMNNf+Ynh59Ms2bNOPDgfZly94NJT0sawbbbbs0/Hr2LAQMu4anpzwLw1eIlLF26DIAFCz/jkEMOSHKKTV5oZZhiFXXj1d2vN7NjgCuA0xt2Sun1wXsfcf+Uh6l44m6qqtZw3+SHeO9d/YGUBkMu+m/atf0Bw4YOYNjQ6N7NuecO5q47rqeqqorVq9dwTt/BCc+yactkw8rQi1WWLfIHN7MdgG7u/vCGuniHdnul89+61Oqz5VpVJIVVrZ5frzWkp+98YtHx5o6P7w9mvWrRSyjdfSGwwQK8iEhjSusSSq2TF5FUSOvqGgV5EUmFKgV5EZFwKZMXEQmYllCKiASs2JWEoVGQF5FU0OoaEZGANdR2BWa2OTASOAHYGngXGOnuD8X9VwFDCnx0E3evis/pDwwCdgBmAee5+8sbYn7ahVJEUiFDtuijRLcCxwNnAV2B+4EHzOyIuL8LMIkogK89cgL8mcAY4BKgG+DANDPbtj4/bw1l8iKSCg1Rkzez7YETgePd/cm4+Uoz+w+gD/AU0BmocPdFtQwzFJjg7nfFY/Yh2vDxHODy+s5RmbyIpEID7Se/HDgGeCavPQtsaWZtgY7AnEIfNrPtgD2A6TVt7l4NzAQ2yNbtyuRFJBUaYp28u38DPJbbZmYHA0cA5xFl8QCnmdlNwKbADGBIvFVMh7j/07yhFwAbZNtRBXkRSYVSau1xBt62QNcSd691Fz0z+xHwAPAiMJGoTg/wNXASUT3+SmCGme0LtI77K/OGqgRaFT3hdVCQF5FUqM6WVIgZCFxWoP2PwIhCHzCzw4gC/MdEz95YY2Z/BSa7++L4tNlm9hYwD/g50U1WgJZ5w7UElpUy4dooyItIKpRYrhlLtGomX8Es3sxOA24GngZOiss4uHsWWJx7rrvPN7MvgZ2AmieztwfezDmtPd8v4awXBXkRSYVSHhoSl2SKeriBmf0KuB24E/itu6/J6RsHHObu++a07Uq0nv5td//czBzoCUyL+5sDPYjKPfWmIC8iqdAQX4Uysx2J1sBPBy4EtjJb+1zm1cAUoF8c7K8jytDHAS8Bj8TnXQ2Mj4P9S8BgoDwet960hFJEUqGBvgx1ItHN0yOIVsQszDkecvdnib4odSDwOjAVeBU41t0zAO4+ieiLUJcDrwCdgKPd/YsN8XMX/fi/hqDH/0k+Pf5PalPfx//9uEOvouPN8/Onp+/xfyIiTVmJq2uCoSAvIqmgh4aIiARM+8mLiARM+8mLiARMmbyISMCqU/qUVwV5EUmFUr7xGhIFeRFJBa2uEREJmDJ5EZGAKZMXEQmYMnkRkYBpWwMRkYCpXCMiErCsMnkRkXBpWwMRkYBpWwMRkYApkxcRCVh1RjV5EZFgaXWNiEjAVJMXEQmYavIiIgFTJi8iEjDdeBURCZjKNSIiAVO5RkQkYNpqWEQkYFonLyISMGXyIiIBy2irYRGRcOnGq4hIwNIa5MvS+oOLiKRBs6QnICIiDUdBXkQkYAryIiIBU5AXEQmYgryISMAU5EVEAqYgLyISMAV5EZGAKciLiARM2xokyMyaAZcBZwHtgGeBfu7+QaITk42GmV0MHOfu3ZOeizRNyuSTNRzoC5wNHAxUAdPMrFWis5KNgpn1A0YlPQ9p2pTJJ8TMWgKDgIvc/dG47ZfAQuAU4PYEpycJMrP2wESgF+AJT0eaOGXyyekKlAPTaxrcfSnwGnBYUpOSjUI34BugC/BiwnORJk6ZfHI6xK+f5rUvADo28lxkI+LuFUAFgJklPBtp6pTJJ6d1/FqZ114JqCYvIhuEgnxyVsavLfPaWwLLGnkuIhIoBfnkzItf2+e1t+f7JRwRkfWiIJ+cN4ClQM+aBjPbAtgPeDqhOYlIYHTjNSHuXmlm1wFXmtkiYC4wGpgPTE10ciISDAX5ZA0HmgM3Am2AmcBP3X11orMSkWDoQd4iIgFTTV5EJGAK8iIiAVOQFxEJmIK8iEjAFORFRAKmIC8iEjAFeRGRgCnIi4gETEFeRCRg/w8BMIIZiFpNrAAAAABJRU5ErkJggg==", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXkAAAD/CAYAAAAUnaZMAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjAsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+17YcXAAAYjklEQVR4nO3deZQU1dnH8e8M4rAYQ6JGGUQiYp4YhaDEjQDiGhFy4pJEDZrXxBU0ZiKKKCooKi6o4BYV9TUxGhODcACNaBQEJYr7RvLEKFEEXF8QEMIw0/3+UTXatj0z3Qwzxdz6fTx12r5Vfev2OZynn3nqVt2ybDaLiIiEqTzpAYiISPNRkBcRCZiCvIhIwBTkRUQCpiAvIhIwBXkRkYBtluTJ13/0luZvyhe0r+yf9BBkE1VTvaSsKZ8vJd603bp7k861KUk0yIuItJhMbdIjSISCvIikQzaT9AgSoSAvIumQUZAXEQlWVpm8iEjAamuSHkEiFORFJB104VVEJGAtUK4xs/OAwe7eL6dtD+AK4HvAemAOcLa7L845ZhmwXV5397j7cfH+rYDrgUFAGXA/cJa7r25sTLoZSkTSIZMpftsAZjYcuCyvrSvwGLAU6AcMBr4BPGJm7eNjtiYK8AcBnXO203O6+gvQIz7miPj11mLGpUxeRFKhuS68mlklUcDdH/C83UOBtcBJ7l4TH/9z4B2ioP8o0AvIAn939zUF+u8LDAR2dfeFcdvJwKNmdp67v9PQ+JTJi0g6NF8m3wdYRRSsn8nbNwU4ui7Ax+ruvP16/NoL+E+hAB/rD3xQF+Bj8+J+Gr1FXJm8iKRD7fqiDzWzTkCnArtWuPuK3AZ3nwHMiD9H3r43gDfy+jifKLt/PH7fE/ivmU0D9gbeB/4XuMHdM0AX4N28fqvN7COga2PfRZm8iKRDNlP8BlXAogJbVVOGYGZnAcOAc939w7h5N6Ks/h7gB8Bkotr+xfH+DsC6At2tA9o1dk5l8iKSDqWVYSYCdxVoX1GgrVFmVg6MB0YCl7j7DTm7+wOb58yUecXMtgQuNLOxRFl/RYFuK4BGZ9coyItIOpRw4TUuyWxQQM9nZu2APxDNijnd3W/OO1c1UJ33sVeA9sA2wGKgMq/PzYGtySvjFKJyjYikQzNPoSwkzuCnAEOAo/IDvJlVmNlSMzs776N7AR+7+3vAXGA7+2LBf0D8Oq+xMSiTF5FUyGaKv/C6EVUBhxHV4Z82s9wbnj5x97VmNh0YbWaLgJeAg4nKOnWB/xngKeCPZnYaUYZ/K3C3uy9pbADK5EUkHRLI5IHj4tffAsvytqHxvjOBG4CrgYVEPwy/cfebANw9CxwJvEk0I2cK0fz6YcUMoCybTW5xJq0MJfm0MpTUp6krQ/33+WlFx5t2fQ7XylAiIq2KHlAmIhIwPU9eRCRgWhlKRCRgWjRERCRgyuRFRMKVzerCq4hIuJTJi4gETLNrREQCpkxeRCRgml0jIhIwlWtERAKmco2ISMAU5EVEAqZyjYhIwHThVUQkYCrXiIgETOUaEZGAKZMXEQmYgryISMASXM86SQryIpIONZpdIyISLl14FREJmGryIiIBU01eRCRgyuRFRAKmIC8iEq5srRbyFhEJlzJ5EZGAaQqliEjAMppdIyISrhYo15jZecBgd++X0/ZN4EZgALAWuAsY7e41OcecDowAOgMvAWe6+7Ol9FGf8iZ+JynCK6//kxPOGAnAm4ve5vhhIzjutBGMm3AjtfHFoN/fN5VjT67i2JOruPnOe77w+b898RQjx17Z4uOWllNeXs7k265h7pxpzH5sCt27d2ObbbbigSl3MvuxKcydM43u3bslPczWrba2+G0DmNlw4LK8ts2BR4As0Bc4GTgRuDjnmBOAq4ALgD6AA7PM7BvF9tGQRjP5+AQ/IfoF6Qq0A1YD7wJzgAeK+TVJqzvvuZ8ZDz9O+3YVAEy69Xf8+tQT+F7vnoy+9BpmP/k01qM7Mx+ZzR8nX0dZWRk/H34OBw7oi/XYkfETb2H+M89jO3dP+JtIcxoy5GAABgw8nP0G7MuEq8ewfPkn3PvHqfzlLzMYuF9fvm09eOuttxMeaSvWTJm8mVUCtwL7EwXoXD8GugF7u/ty4DUzOxeYZGaXuvta4HzgJne/N+7vROBN4FRgXJF91KvBTN7MdgIWxl9gF2AlsBhYA+wG3Am8Gv8pIQV0rezMxMsv+Oz9dZeN5nu9e7J+/Xo++r/lbPX1r7Hdtttw67XjaNOmDeXl5dTU1FCxeVsAeu+2CxeefUZSw5cWMn36LE4bFv21t0O37Xn//Q/pu++ebN+lM7P+eh/HHnsEc56Yn/AoW7lMtvitNH2AVUAv4Jm8ff2Bl+LgXGcO0BHYw8y2BXYGZtftdPdaYB5RYt1oH40NrrFM/maiX6bd3X1V/k4z2xK4j6hWNKSxk6XRwfv3Y8my9z9736ZNG5a+9z4n/fp8tujYkR132J62m23G1zp9lWw2y4SbbmeXb+3EN3fYHoBBB+3HghdeSWr40oJqa2u5846JHP6jQzn6mFM44X+OZvnyFfxg0DFcMLqKkeecztiLJyQ9zNarhNk1ZtYJ6FRg1wp3X5Hb4O4zgBnx5/KP70JU9ci1NH7tSlRfp55j9iyyjwY1VpPvB5xbKMADuPtKYBSf/+JIESq325aH/nQHPz38MK66/jYA1q2r5tyLr2LNmrVcMOL0hEcoSfnliVXssmt/bvnt1axYsZIZMx8FYOaDj9Jnj14Jj66VKy2TrwIWFdiqSjxrB2BdXlvd+3bxfuo5pl2RfTSosSC/nOhXpCHfBD5t7EQSOWPkWN5evASAjh3aU15eTjab5VejLsZ67MiYkWfSpk2bhEcpLW3o0KM4d2RUlluzZi2ZTIa5855m0KADAOjfbx8WLvxXkkNs9bKZTNEbMBHYscA2scTTrgUq8trq3q/m80y+0DGri+yjQY2Va+4AfmdmY4hqQO8S/YJUAJXAQOByYHJjJ5LIicf/lNGXXUvbzTajXbsKLhlVxWNz5/PcS69SvX49855+DoCq035B7912SXi00lKmTn2IO26/jtmPTaFt27acdfYYXn75dW67ZQKnnfJzPvlkJcf9XNdmmqSEWTNxSWZFowc2bjGwe15bZfz6LvBOTturecfUlWga66NBjQX5sUTTdq4mKvLnW0VUj7+osROlWZfO23Lv5CgB2L3nd/jDLdd8Yf9B+32fF2ZPr/fze+3Ri730p3rQ1qxZy7E/O+1L7YcedmwCowlUMjdDzQV+YWadcmr5+xPFzhfcvdrMnChhngVgZm2ILrbeWkwfjQ2gwSDv7llgrJldDvQmKt10IPrzYTHwsrtXF/llRUSSk8yza6YBlwJ/MrNziKZCjgeuzYmd1wDXx8F+AXAOsAWfV0iK6aNeRd3xGne0oNhvJSKyyUkgk3f3/5rZocBNRNMrlxNl6JfkHDM5nqk4DtgKeB44xN0/KraPhpRlE1wtZf1Hb6XzYRJSr/aV/ZMegmyiaqqXlDXl859e+NOi403HcX9u0rk2JXp2jYikgx5QJiISrmyNFg0REQmXMnkRkYBp0RARkYApkxcRCVdWQV5EJGC68CoiEjBl8iIiAVOQFxEJV5J39ydJQV5E0kGZvIhIwBTkRUTCla3RzVAiIuFKZ4xXkBeRdNDNUCIiIVOQFxEJmMo1IiLhUrlGRCRg2RoFeRGRcKlcIyISrpSuGaIgLyIpoSAvIhIuZfIiIgHL1iQ9gmQoyItIKiiTFxEJmIK8iEjIsmVJjyARCvIikgrK5EVEApbNKJMXEQlWplZBXkQkWM1RrjGzgcDsenYvcvfuZjYeGFVgf1t3r4n7OR0YAXQGXgLOdPdnN8YYyzdGJyIim7pspqzorQTziQJz7nYIUAOMj4/pBUzOPy4nwJ8AXAVcAPQBHJhlZt9o4lcGlMmLSEpkm+EhlO5eDbxX997M2gITgQfcfXLc3BOY4e7vFegC4HzgJne/N+7jROBN4FRgXFPHqExeRFKhmTL5fGcCXYHfAJhZp/j9wkIHm9m2wM7klHzcvRaYBwxoykDqKJMXkVQo5cJrHJw7Fdi1wt1X1POZDkRZ+SR3Xxo394xfh5rZHcDmwBxglLsvA7rE+9/N624psGfRA26AMnkRSYUSM/kqYFGBraqBUxwPtAMm5bTtFr9+AhwFnBK3zYl/FDrE+9fl9bUu7qvJlMmLSCpkS7vjdSJwV4H2gll87HiiWvxHOW23APe5+/L4/Stm9hqwGDic6CIrQEVeXxXA6lIGXB8FeRFJhVKmUMYlmYYC+heY2TbAvsDlef1kgeV5bUvM7GNgB+DRuLkSeDXnsEq+XMLZICrXiEgqZLJlRW8b4PtAFngit9HMJpnZi3ltOwJbA6+7+4dE2fzAnP1tgP75fW0oZfIikgollmtKtTvwlrt/mtd+PzDczCYBNxJl6JOABcCD8THXANebmcft5wBbEM2tbzJl8iKSCpnasqK3DdAZ+L/8Rnd/EhgC7AW8CEwBngcOc/dMfMxkohuhxgHPAT2AQ/Jq+xusLNscdwgUaf1HbyV3ctkkta/sn/QQZBNVU72kSan4wp0GFx1vvvPmg8E86EblGhFJhQ2stbd6CvIikgrNXJPfZCnIi0gqJFiZTpSCvIikgso1IiIBy2hlqJZXudOgJE8vIimiTF5EJGC68CoiEjBl8iIiAUvp5BoFeRFJh9pMOp/ioiAvIqlQwpOGg6IgLyKpkEU1eRGRYGVSWpRXkBeRVMgokxcRCZfKNSIiAatVkBcRCZdm14iIBExBXkQkYKrJi4gELKVPGlaQF5F00BRKEZGA1SY9gIQoyItIKmTKlMmLiAQrpU81UJAXkXTQFEoRkYBpdo2ISMD0WAMRkYApkxcRCZhq8iIiAdPsGhGRgKlcIyISsOYq15iZAf8ssOtkd7/dzHoDE4E9gY+B6919Qs7ny4ExwEnA14AngeHu/u+NMb7yjdGJiMimrras+K1EvYCVQOe87R4z2xr4G/AG8D3gQuASMzs55/MXAcOAk4F9gBpglpm12/Bv+zll8iKSCs144bUn8A93fy9/h5n9BqgGhrl7DfAPM+sBjAImm1kFMAI4190fij9zDLAM+Alwd1MHp0xeRFIhU8JWol7Awnr29QfmxQG+zhygu5l1AXoDWwCz63a6+0rgBWBA6UP5MmXyIpIKpcyuMbNOQKcCu1a4+4q8tp7AG2b2FNAD+Bcwzt0fAbrw5R+ApfFrV6Ay/v93CxzTtYQh10uZvIikQqas+A2oAhYV2Kpy+zSzjsCORD8Io4HDgOeAh83sIKADsC5vKHXv28X7qecY1eRFRIpVYhlmInBXgfYvZPHu/qmZbQlUu3t13Py8mX0HGAmsBSry+qh7vzreX9dWnXfM6tKGXJiCvIikQimLhsQlmfyyTH3HFgrGrwA/BN7i85JMndwSTXlOm+cd83qx422IyjUikgollmuKYmb7mtkqM9szb9eeREF6LtDPzHIT6v2Bf8WzcV4mmn45MKfPLYE9gCc25HvmUyYvIqnQTFMonyOq1U82s9OJbnYaBvQF9gaWEJVt7jSzK4A+wFnAcAB3X2dmNwKXm9l7cV9XxJ+bsjEGqExeRFIhW8JWLHdfDxxKVJ6ZArxEdNPTwe7+ort/ABxCNOvmBWAcMMrd78rp5iJgMnAbMB8oAw7NqfE3SVk2m9xje7b5qqX1mUFSj+VrN8q1JglQTfWSJj195rJuQ4uON6PfvieYJ92oXCMiqVDKhdeQKMiLSCroefIiIgHTo4ZFRAKWSemyIQryIpIK6QzxCvIikhKqyYuIBKw2pbm8gryIpIIyeRGRgOnCq4hIwNIZ4hXkRSQlVK4REQmYLryKiAQsrTV5PWq4Be3RpxfTZv4egB2778DMh+9lxl/v4aprx1JW9vk91+3bt2P2vGkccGD/pIYqLay8vJzJt13D3DnTmP3YFLp378buvXfj70/NZM7jDzDxunFf+DcipWuORw23BgryLeSMX5/EdTdcSkW7aHnHSy47j8svncgPBw2lrKyMQYMP/OzYKydcRDa4f2rSkCFDDgZgwMDDGXvxBCZcPYbf/vYqzhoxloEHHMnKlas49tgjEh5l65YhW/QWEgX5FvKfRe/wi+N+9dn77/belflPLgDgsUfnst/AvgAM/9UvWbDgRV5/9Z+JjFOSMX36LE4bNhKAHbptz/vvf8j2XTrz96efA2D+/Gf5ft+9khxiq5cpYQtJozV5M5tHkX/BuPuAJo8oUDOnP0LXHbp89j73T+/Vqz/lK1t+hf777UP3nbpxdtUY9t57jySGKQmqra3lzjsmcviPDuXoY05ht12/zYD++zB33tMMHnwwHTt2SHqIrVpa/zou5sLrQ8ClwD+BBc07nPTIZD7PF7bYoiMrP1nJ0ON/TNeuXZg28/fs/K3u9PzurnzwwYe8pqw+NX55YhXnnb8N85+cyRFH/YLxl53P2SOG89zzL1G9bqOsBpdaml1TD3cfb2afEC0uO9jd/9Pso0qBV19ZSN9+ezH/yQUcePAAnpr3NNMe+Otn+2+4eTxTpzykAJ8SQ4cexfZdOnPlVTeyZs1aMpkMhw06kJNOGcGyZe8z8bpxzJo1O+lhtmqhlWGKVdQUSne/2cwGEWX0xzXvkNLhotFXct3142i7eVve8LeYPm1W0kOSBE2d+hB33H4dsx+bQtu2bTnr7DFkM1lmTL+btWvWMueJ+fz14ceTHmarlklwPeskFb2Qt5l1Bvq4+8yNdXIt5C35tJC31KepC3kf1+3IouPNH95+IJj5qkXfDOXuy4CNFuBFRFpSaFMji6U7XkUkFTS7RkQkYDUK8iIi4VImLyISME2hFBEJWLEzCUOjIC8iqaDZNSIiAdNjDUREAqZMXkQkYM1VkzezrwCXAEcAWxM9zPESd58e7x8PjCrw0bbuXhMfczowAugMvASc6e7Pbozx6XnyIpIKzfg8+buAIcBJQG/gAWCqmR0Q7+8FTCYK4J9tOQH+BOAq4AKgD+DALDP7RulD+TJl8iKSCs0xT97MtgOOBIa4+9/i5svN7EDgROBxoCcww93fq6eb84Gb3P3euM8TgTeBU4FxTR2jMnkRSYVmWv7vU2AQMDevPQt83cw6AV2BhYU+bGbbAjsDnz1H2t1rgXnARlmESZm8iKRCbbb4QkwcnDsV2LXC3VfUvXH3VcDDeZ/dBzgAOJMoiwcYamZ3AJsDc4BR8UMf65aLezfvPEuBPYsecAOUyYtIKmRL+A+oAhYV2KoaOoeZ7QJMBZ4BbgV2i3d9AhwFnBK3zTGzDkDdmo7r8rpaB7Rr4lcGlMmLSEqUuGjIRKILqvlWFGgDwMwGEAX4t4lW0VtvZrcA97n78viwV8zsNWAxcDjRRVaAirzuKoCNsriCgryIpEIpIT4uydQb0POZ2VDgTuAJ4Ki4jIO7Z4Hluce6+xIz+xjYAXg0bq4EXs05rJIvl3A2iMo1IpIKzXThFTP7GXA38GeiDH5Vzr5JZvZi3vE7Es2nf93dPyTK5gfm7G8D9Cf6wWgyZfIikgrNccermW1PNAd+NjAS2MrM6nZXA/cDw81sEnAjUYY+CVgAPBgfdw1wvZl53H4OsEXcb5MpkxeRVKjNZoreSnAk0cXTA4hmxCzL2aa7+5NEN0rtBbwITAGeBw5z9wyAu08muhFqHPAc0AM4xN0/2hjfu+iFvJuDFvKWfFrIW+rT1IW896wcUHS8eXbp3PQt5C0i0prpefIiIgHTUyhFRAKmTF5EJGC1KV3lVUFeRFKhxDteg6EgLyKp0ByPGm4NFORFJBWUyYuIBEyZvIhIwJTJi4gErMTHFQRDQV5EUkHlGhGRgGWVyYuIhEuPNRARCZgeayAiEjBl8iIiAavNqCYvIhIsza4REQmYavIiIgFTTV5EJGDK5EVEAqYLryIiAVO5RkQkYCrXiIgETI8aFhEJmObJi4gETJm8iEjAMnrUsIhIuHThVUQkYGkN8mVp/eIiImlQnvQARESk+SjIi4gETEFeRCRgCvIiIgFTkBcRCZiCvIhIwBTkRUQCpiAvIhIwBXkRkYDpsQYJMrNyYAxwEvA14ElguLv/O9GBySbDzM4DBrt7v6THIq2TMvlkXQQMA04G9gFqgFlm1i7RUckmwcyGA5clPQ5p3ZTJJ8TMKoARwLnu/lDcdgywDPgJcHeCw5MEmVklcCuwP+AJD0daOWXyyekNbAHMrmtw95XAC8CApAYlm4Q+wCqgF/BMwmORVk6ZfHK6xK/v5rUvBbq28FhkE+LuM4AZAGaW8GiktVMmn5wO8eu6vPZ1gGryIrJRKMgnZ238WpHXXgGsbuGxiEigFOSTszh+rcxrr+TLJRwRkQ2iIJ+cl4GVwMC6BjPbEtgDeCKhMYlIYHThNSHuvs7MbgQuN7P3gEXAFcASYEqigxORYCjIJ+sioA1wG9ARmAcc6u7ViY5KRIKhhbxFRAKmmryISMAU5EVEAqYgLyISMAV5EZGAKciLiARMQV5EJGAK8iIiAVOQFxEJmIK8iEjA/h8XOrykFjmhhQAAAABJRU5ErkJggg==", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "for i, model in enumerate(project.get_models()):\n", " plt.figure(i)\n", " result = get_predictions(model)\n", " cm = confusion_matrix(result[\"SAR\"], result[\"prediction_binary\"])\n", " df_cm = pd.DataFrame(cm, range(2), range(2))\n", " print(\n", " \"F1 Score: \", f1_score(result[\"SAR\"], result[\"prediction_binary\"], average=\"macro\"), model\n", " )\n", " sns.heatmap(df_cm, annot=True, fmt=\"g\")" ] } ], "metadata": { "kernelspec": { "display_name": "Python 3", "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": "3.8.8" } }, "nbformat": 4, "nbformat_minor": 4 }