diff --git a/inclusio/StarterNotebook.ipynb b/inclusio/StarterNotebook.ipynb index 3d80d62..b807cd5 100644 --- a/inclusio/StarterNotebook.ipynb +++ b/inclusio/StarterNotebook.ipynb @@ -31,7 +31,7 @@ }, { "cell_type": "code", - "execution_count": 419, + "execution_count": 177, "metadata": { "id": "8IWFJK2h22yc" }, @@ -64,7 +64,7 @@ }, { "cell_type": "code", - "execution_count": 420, + "execution_count": 178, "metadata": { "id": "MQCvC_XjDzyH" }, @@ -79,7 +79,7 @@ }, { "cell_type": "code", - "execution_count": 421, + "execution_count": 179, "metadata": { "colab": { "base_uri": "https://localhost:8080/" @@ -116,7 +116,7 @@ }, { "cell_type": "code", - "execution_count": 422, + "execution_count": 180, "metadata": { "colab": { "base_uri": "https://localhost:8080/", @@ -277,7 +277,7 @@ "4 Primary education Informally employed " ] }, - "execution_count": 422, + "execution_count": 180, "metadata": {}, "output_type": "execute_result" } @@ -289,7 +289,7 @@ }, { "cell_type": "code", - "execution_count": 423, + "execution_count": 181, "metadata": { "colab": { "base_uri": "https://localhost:8080/" @@ -334,141 +334,6 @@ "\n" ] }, - { - "cell_type": "code", - "execution_count": 424, - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/", - "height": 386 - }, - "id": "QW4mlprHj-Ir", - "outputId": "6089fa03-6eb3-4117-f694-b4b901bd57c0" - }, - "outputs": [ - { - "data": { - "text/plain": [ - "" - ] - }, - "execution_count": 424, - "metadata": {}, - "output_type": "execute_result" - }, - { - "data": { - "image/png": "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", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" - } - ], - "source": [ - "# Explore Target distribution \n", - "sns.catplot(x=\"bank_account\", kind=\"count\", data=train)" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "W0sPGiOnbTnh" - }, - "source": [ - "It is important to understand the meaning of each feature so you can really understand the dataset. You can read the VariableDefinition.csv file to understand the meaning of each variable presented in the dataset.\n", - "\n", - "The SampleSubmission.csv gives us an example of how our submission file should look. This file will contain the uniqueid column combined with the country name from the Test.csv file and the target we predict with our model. Once we have created this file, we will submit it to the competition page and obtain a position on the leaderboard.\n" - ] - }, - { - "cell_type": "code", - "execution_count": 425, - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/", - "height": 206 - }, - "id": "dPjsaC0_a0qY", - "outputId": "999100c2-5e84-42df-a3d9-4bbf99a9dd4a" - }, - "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", - "
unique_idbank_account
0uniqueid_1 x Kenya0
1uniqueid_2 x Kenya0
2uniqueid_3 x Kenya0
3uniqueid_4 x Kenya0
4uniqueid_5 x Kenya0
\n", - "
" - ], - "text/plain": [ - " unique_id bank_account\n", - "0 uniqueid_1 x Kenya 0\n", - "1 uniqueid_2 x Kenya 0\n", - "2 uniqueid_3 x Kenya 0\n", - "3 uniqueid_4 x Kenya 0\n", - "4 uniqueid_5 x Kenya 0" - ] - }, - "execution_count": 425, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "# view the submission file\n", - "ss.head()" - ] - }, { "cell_type": "markdown", "metadata": { @@ -481,7 +346,7 @@ }, { "cell_type": "code", - "execution_count": 426, + "execution_count": 182, "metadata": { "colab": { "base_uri": "https://localhost:8080/" @@ -540,7 +405,7 @@ }, { "cell_type": "code", - "execution_count": 427, + "execution_count": 183, "metadata": { "colab": { "base_uri": "https://localhost:8080/", @@ -656,7 +521,7 @@ "11 job_type Type of job interviewee has: Farming and Fishi..." ] }, - "execution_count": 427, + "execution_count": 183, "metadata": {}, "output_type": "execute_result" } @@ -667,22 +532,24 @@ ] }, { - "cell_type": "markdown", - "metadata": { - "id": "wK0NuOEOcAmC" - }, + "cell_type": "code", + "execution_count": 184, + "metadata": {}, + "outputs": [], "source": [ - "### 3. Data preparation for machine learning\n", - "Before you train the model for prediction, you need to perform data [cleaning](https://www.dataquest.io/blog/machine-learning-preparing-data/) and [preprocessing](https://towardsdatascience.com/data-preprocessing-concepts-fa946d11c825). This is a very important step; your model will not perform well without these steps.\n", - "\n", - "![48f91a8a-d9c4-4853-ab32-ccb9f143a238.png](data:image/png;base64,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)\n", - "\n", - "The first step is to separate the independent variables and target(bank_account) from the train data. Then transform the target values from the object data type into numerical by using [LabelEncoder](https://towardsdatascience.com/categorical-encoding-using-label-encoding-and-one-hot-encoder-911ef77fb5bd).\n" + "#Convert to datetime features\n", + "train[['year']] = train[['year']].apply(lambda x:pd.to_datetime(x,format='%Y-%m-%d',errors='coerce'))\n", + "test[['year']] = test[['year']].apply(lambda x:pd.to_datetime(x,format='%Y-%m-%d',errors='coerce'))\n", + "#Create datetime features\n", + "from feature_engine.datetime import DatetimeFeatures\n", + "dtf = DatetimeFeatures(features_to_extract = [\"month\"])\n", + "train = dtf.fit_transform(train)\n", + "test = dtf.fit_transform(test)" ] }, { "cell_type": "code", - "execution_count": 428, + "execution_count": 185, "metadata": { "colab": { "base_uri": "https://localhost:8080/" @@ -728,7 +595,7 @@ }, { "cell_type": "code", - "execution_count": 429, + "execution_count": 186, "metadata": {}, "outputs": [], "source": [ @@ -755,7 +622,7 @@ }, { "cell_type": "code", - "execution_count": 430, + "execution_count": 187, "metadata": { "id": "5nLXAW5hdYOk" }, @@ -765,7 +632,7 @@ "def preprocessing_data(data):\n", "\n", " # Convert the following numerical labels from interger to float\n", - " float_array = data[[\"household_size\", \"age_of_respondent\", \"year\"]].values.astype(float)\n", + " float_array = data[[\"household_size\", \"age_of_respondent\", \"year_month\"]].values.astype(float)\n", " \n", " # categorical features to be onverted to One Hot Encoding\n", " categ = [\"relationship_with_head\",\n", @@ -803,7 +670,7 @@ }, { "cell_type": "code", - "execution_count": 431, + "execution_count": 188, "metadata": { "id": "eNdoBGhgdYdQ" }, @@ -825,7 +692,7 @@ }, { "cell_type": "code", - "execution_count": 432, + "execution_count": 189, "metadata": { "colab": { "base_uri": "https://localhost:8080/" @@ -838,7 +705,7 @@ "name": "stdout", "output_type": "stream", "text": [ - "[[1. 0. 1. 0.1 0.0952381 0. 0.\n", + "[[0. 1. 0.1 0.0952381 0. 0. 0.\n", " 0. 0. 0. 0. 1. 0. 0.\n", " 1. 0. 0. 0. 0. 0. 1.\n", " 0. 0. 0. 0. 0. 0. 0.\n", @@ -863,7 +730,7 @@ }, { "cell_type": "code", - "execution_count": 433, + "execution_count": 190, "metadata": { "colab": { "base_uri": "https://localhost:8080/" @@ -887,22 +754,22 @@ }, { "cell_type": "code", - "execution_count": 434, + "execution_count": 191, "metadata": {}, "outputs": [ { "data": { "text/plain": [ - "array([[1., 0., 1., ..., 0., 0., 0.],\n", - " [1., 0., 0., ..., 0., 0., 0.],\n", - " [1., 1., 1., ..., 0., 0., 0.],\n", + "array([[0. , 1. , 0.1 , ..., 0. , 0. , 0. ],\n", + " [0. , 0. , 0.2 , ..., 0. , 0. , 0. ],\n", + " [1. , 1. , 0.2 , ..., 0. , 0. , 0. ],\n", " ...,\n", - " [1., 0., 1., ..., 0., 0., 1.],\n", - " [1., 1., 1., ..., 0., 0., 1.],\n", - " [1., 0., 1., ..., 0., 0., 1.]])" + " [0. , 1. , 0.2 , ..., 0. , 0. , 1. ],\n", + " [1. , 1. , 0.3 , ..., 0. , 0. , 1. ],\n", + " [0. , 1. , 0.45, ..., 0. , 0. , 1. ]])" ] }, - "execution_count": 434, + "execution_count": 191, "metadata": {}, "output_type": "execute_result" } @@ -913,26 +780,26 @@ }, { "cell_type": "code", - "execution_count": 435, + "execution_count": 192, "metadata": {}, "outputs": [], "source": [ - "processed_train = np.delete(processed_train,[23,13,18,29,8,10], axis=1)\n", - "processed_test = np.delete(processed_test,[23,13,18,29,8,10], axis=1) " + "processed_train = np.delete(processed_train,[23,13,18,29,8,10,5], axis=1)\n", + "processed_test = np.delete(processed_test,[23,13,18,29,8,10,5], axis=1) " ] }, { "cell_type": "code", - "execution_count": 436, + "execution_count": 193, "metadata": {}, "outputs": [ { "data": { "text/plain": [ - "(23524, 31)" + "(23524, 30)" ] }, - "execution_count": 436, + "execution_count": 193, "metadata": {}, "output_type": "execute_result" } @@ -968,7 +835,7 @@ }, { "cell_type": "code", - "execution_count": 437, + "execution_count": 194, "metadata": { "id": "2MSEu48clFhH" }, @@ -979,7 +846,7 @@ }, { "cell_type": "code", - "execution_count": 438, + "execution_count": 195, "metadata": { "id": "6HXDwTMtlB8-" }, @@ -1013,7 +880,7 @@ }, { "cell_type": "code", - "execution_count": 439, + "execution_count": 196, "metadata": { "colab": { "base_uri": "https://localhost:8080/" @@ -1025,7 +892,7 @@ { "data": { "text/html": [ - "
XGBClassifier(base_score=None, booster=None, callbacks=None,\n",
+              "
XGBClassifier(base_score=None, booster=None, callbacks=None,\n",
               "              colsample_bylevel=None, colsample_bynode=None,\n",
               "              colsample_bytree=None, device=None, early_stopping_rounds=None,\n",
               "              enable_categorical=False, eval_metric=None, feature_types=None,\n",
@@ -1035,7 +902,7 @@
               "              max_delta_step=None, max_depth=None, max_leaves=None,\n",
               "              min_child_weight=None, missing=nan, monotone_constraints=None,\n",
               "              multi_strategy=None, n_estimators=None, n_jobs=None,\n",
-              "              num_parallel_tree=None, random_state=None, ...)
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.