{ "cells": [ { "cell_type": "markdown", "id": "aa74dbe0-f974-4b5c-94f4-4dba9fbc64fa", "metadata": {}, "source": [ "# Business Data Challenge - Team 1" ] }, { "cell_type": "code", "execution_count": 1, "id": "94c498e7-7c50-45f9-b3f4-a1ab19b7ccc4", "metadata": {}, "outputs": [], "source": [ "import pandas as pd\n", "import numpy as np\n", "\n", "\n" ] }, { "cell_type": "markdown", "id": "7a3b50ac-b1ff-4f3d-9938-e048fdc8e027", "metadata": {}, "source": [ "Configuration de l'accès aux données" ] }, { "cell_type": "code", "execution_count": 2, "id": "0b029d42-fb02-481e-a407-7e41886198a6", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "['bdc2324-data/1',\n", " 'bdc2324-data/10',\n", " 'bdc2324-data/101',\n", " 'bdc2324-data/11',\n", " 'bdc2324-data/12',\n", " 'bdc2324-data/13',\n", " 'bdc2324-data/14',\n", " 'bdc2324-data/2',\n", " 'bdc2324-data/3',\n", " 'bdc2324-data/4',\n", " 'bdc2324-data/5',\n", " 'bdc2324-data/6',\n", " 'bdc2324-data/7',\n", " 'bdc2324-data/8',\n", " 'bdc2324-data/9']" ] }, "execution_count": 2, "metadata": {}, "output_type": "execute_result" } ], "source": [ "import os\n", "import s3fs\n", "# Create filesystem object\n", "S3_ENDPOINT_URL = \"https://\" + os.environ[\"AWS_S3_ENDPOINT\"]\n", "fs = s3fs.S3FileSystem(client_kwargs={'endpoint_url': S3_ENDPOINT_URL})\n", "\n", "BUCKET = \"bdc2324-data\"\n", "fs.ls(BUCKET)" ] }, { "cell_type": "code", "execution_count": 4, "id": "fbaf9aa7-ff70-4dbe-a969-b801c593510b", "metadata": {}, "outputs": [], "source": [ "# Chargement des fichiers campaign_stats.csv\n", "FILE_PATH_S3 = 'bdc2324-data/1/1campaign_stats.csv'\n", "\n", "with fs.open(FILE_PATH_S3, mode=\"rb\") as file_in:\n", " campaign_stats_1 = pd.read_csv(file_in, sep=\",\")\n", "\n", "FILE_PATH_S3 = 'bdc2324-data/2/2campaign_stats.csv'\n", "\n", "with fs.open(FILE_PATH_S3, mode=\"rb\") as file_in:\n", " campaign_stats_2 = pd.read_csv(file_in, sep=\",\")\n", "\n", "FILE_PATH_S3 = 'bdc2324-data/3/3campaign_stats.csv'\n", "\n", "with fs.open(FILE_PATH_S3, mode=\"rb\") as file_in:\n", " campaign_stats_3 = pd.read_csv(file_in, sep=\",\")" ] }, { "cell_type": "code", "execution_count": 5, "id": "1e0418bc-8e97-4a04-b7f3-bda3bef7d36e", "metadata": {}, "outputs": [], "source": [ "# Conversion des dates 'sent_at'\n", "campaign_stats_1['sent_at'] = pd.to_datetime(campaign_stats_1['sent_at'], format = 'ISO8601', utc = True)\n", "campaign_stats_2['sent_at'] = pd.to_datetime(campaign_stats_2['sent_at'], format = 'ISO8601', utc = True)\n", "campaign_stats_3['sent_at'] = pd.to_datetime(campaign_stats_3['sent_at'], format = 'ISO8601', utc = True)" ] }, { "cell_type": "code", "execution_count": 6, "id": "cc5c20ba-e827-4e5a-97a5-7f3947e0621c", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "2023-11-09 18:10:45+00:00\n", "2020-06-02 08:24:08+00:00\n", "2023-10-12 01:39:48+00:00\n", "2023-10-10 17:06:29+00:00\n", "2023-11-01 09:20:48+00:00\n", "2021-03-31 14:59:02+00:00\n" ] } ], "source": [ "# Chaque unites correspond à une période ? --> Non, les dossiers ont juste pour but de réduire la taille des fichiers\n", "print(campaign_stats_1['sent_at'].max())\n", "print(campaign_stats_1['sent_at'].min())\n", "\n", "print(campaign_stats_2['sent_at'].max())\n", "print(campaign_stats_2['sent_at'].min())\n", "\n", "print(campaign_stats_3['sent_at'].max())\n", "print(campaign_stats_3['sent_at'].min())" ] }, { "cell_type": "code", "execution_count": 7, "id": "c75632df-b018-4bb8-a99d-83f15af94369", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "0 2021-03-28 16:01:09+00:00\n", "1 2021-03-28 16:01:09+00:00\n", "2 2021-03-28 16:00:59+00:00\n", "3 2021-03-28 16:00:59+00:00\n", "4 2021-03-28 16:01:06+00:00\n", " ... \n", "6214803 2023-10-23 09:32:33+00:00\n", "6214804 2023-10-23 09:32:49+00:00\n", "6214805 2023-10-23 09:33:28+00:00\n", "6214806 2023-10-23 09:31:53+00:00\n", "6214807 2023-10-23 09:33:54+00:00\n", "Name: sent_at, Length: 6214808, dtype: datetime64[ns, UTC]" ] }, "execution_count": 7, "metadata": {}, "output_type": "execute_result" } ], "source": [ "campaign_stats_1['sent_at']" ] }, { "cell_type": "markdown", "id": "f4c0c63e-0418-4cfe-a57d-7af57bca0c22", "metadata": {}, "source": [ "### Customersplus.csv" ] }, { "cell_type": "code", "execution_count": 8, "id": "d3bf880d-1065-4d5b-9954-1830aa5081af", "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "/tmp/ipykernel_1362/4118060109.py:9: DtypeWarning: Columns (20) have mixed types. Specify dtype option on import or set low_memory=False.\n", " customers_plus_2 = pd.read_csv(file_in, sep=\",\")\n" ] } ], "source": [ "FILE_PATH_S3 = 'bdc2324-data/1/1customersplus.csv'\n", "\n", "with fs.open(FILE_PATH_S3, mode=\"rb\") as file_in:\n", " customers_plus_1 = pd.read_csv(file_in, sep=\",\")\n", "\n", "FILE_PATH_S3 = 'bdc2324-data/2/2customersplus.csv'\n", "\n", "with fs.open(FILE_PATH_S3, mode=\"rb\") as file_in:\n", " customers_plus_2 = pd.read_csv(file_in, sep=\",\")" ] }, { "cell_type": "code", "execution_count": 10, "id": "7368f381-db8e-4a4d-9fe2-5947eb55be58", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "Index(['id', 'lastname', 'firstname', 'birthdate', 'email', 'street_id',\n", " 'created_at', 'updated_at', 'civility', 'is_partner', 'extra',\n", " 'deleted_at', 'reference', 'gender', 'is_email_true', 'extra_field',\n", " 'identifier', 'opt_in', 'structure_id', 'note', 'profession',\n", " 'language', 'mcp_contact_id', 'need_reload', 'last_buying_date',\n", " 'max_price', 'ticket_sum', 'average_price', 'fidelity',\n", " 'average_purchase_delay', 'average_price_basket',\n", " 'average_ticket_basket', 'total_price', 'preferred_category',\n", " 'preferred_supplier', 'preferred_formula', 'purchase_count',\n", " 'first_buying_date', 'last_visiting_date', 'zipcode', 'country', 'age',\n", " 'tenant_id'],\n", " dtype='object')" ] }, "execution_count": 10, "metadata": {}, "output_type": "execute_result" } ], "source": [ "customers_plus_1.columns" ] }, { "cell_type": "code", "execution_count": null, "id": "08091935-b159-47fa-806c-e1444f3b227e", "metadata": {}, "outputs": [], "source": [ "customers_plus_1.shape" ] }, { "cell_type": "code", "execution_count": null, "id": "9f8c8868-c1ac-4cee-af08-533d928f6764", "metadata": {}, "outputs": [], "source": [ "customers_plus_1['id'].nunique()" ] }, { "cell_type": "code", "execution_count": null, "id": "bf95daf2-4852-4718-b474-207a1ebd8ac4", "metadata": {}, "outputs": [], "source": [ "customers_plus_2['id'].nunique()" ] }, { "cell_type": "code", "execution_count": null, "id": "1425c385-3216-4e4f-ae8f-a121624721ba", "metadata": {}, "outputs": [], "source": [ "common_id = set(customers_plus_2['id']).intersection(customers_plus_1['id'])" ] }, { "cell_type": "code", "execution_count": 61, "id": "92533026-e27c-4f1f-81ca-64eda32a34c0", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "1" ] }, "execution_count": 61, "metadata": {}, "output_type": "execute_result" } ], "source": [ "common_id = set(customers_plus_2['id']).intersection(customers_plus_1['id'])\n", "# Exemple id commun = caractéristiques communes\n", "print(customers_plus_2[customers_plus_2['id'] == list(common_id)[0]])\n", "\n", "print(customers_plus_1[customers_plus_1['id'] == list(common_id)[0]])" ] }, { "cell_type": "code", "execution_count": 49, "id": "bf9ebc94-0ba6-443d-8e53-22477a6e79a7", "metadata": { "scrolled": true }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "id 0.000000\n", "lastname 43.461341\n", "firstname 44.995588\n", "birthdate 96.419870\n", "email 8.622075\n", "street_id 0.000000\n", "created_at 0.000000\n", "updated_at 0.000000\n", "civility 100.000000\n", "is_partner 0.000000\n", "extra 100.000000\n", "deleted_at 100.000000\n", "reference 100.000000\n", "gender 0.000000\n", "is_email_true 0.000000\n", "extra_field 100.000000\n", "identifier 0.000000\n", "opt_in 0.000000\n", "structure_id 88.072380\n", "note 99.403421\n", "profession 95.913503\n", "language 99.280945\n", "mcp_contact_id 34.876141\n", "need_reload 0.000000\n", "last_buying_date 51.653431\n", "max_price 51.653431\n", "ticket_sum 0.000000\n", "average_price 8.639195\n", "fidelity 0.000000\n", "average_purchase_delay 51.653431\n", "average_price_basket 51.653431\n", "average_ticket_basket 51.653431\n", "total_price 43.014236\n", "preferred_category 100.000000\n", "preferred_supplier 100.000000\n", "preferred_formula 100.000000\n", "purchase_count 0.000000\n", "first_buying_date 51.653431\n", "last_visiting_date 100.000000\n", "zipcode 71.176564\n", "country 5.459418\n", "age 96.419870\n", "tenant_id 0.000000\n", "dtype: float64\n" ] } ], "source": [ "pd.DataFrame(customers_plus_1.isna().mean()*100)" ] }, { "cell_type": "code", "execution_count": 11, "id": "6d62e73f-3925-490f-9fd4-d0e838903cb2", "metadata": {}, "outputs": [], "source": [ "# Chargement de toutes les données\n", "liste_base = ['customer_target_mappings', 'customersplus', 'target_types', 'tags', 'events', 'tickets', 'representations', 'purchases', 'products']\n", "\n", "for nom_base in liste_base:\n", " FILE_PATH_S3 = 'bdc2324-data/11/11' + nom_base + '.csv'\n", " with fs.open(FILE_PATH_S3, mode=\"rb\") as file_in:\n", " globals()[nom_base] = pd.read_csv(file_in, sep=\",\")" ] }, { "cell_type": "code", "execution_count": 12, "id": "12b24f1c-eb3e-45be-aaf3-b9273180caa3", "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", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
idlastnamefirstnamebirthdateemailstreet_idcreated_atupdated_atcivilityis_partner...tenant_idid_xcustomer_idpurchase_datetype_ofis_from_subscriptionamountis_full_pricestart_date_timeevent_name
0405082lastname405082NaNNaNNaN62023-01-12 06:30:31.197484+01:002023-01-12 06:30:31.197484+01:00NaNFalse...15569924234050822023-01-11 17:08:41+01:003False13.0False2023-02-06 20:00:00+01:00zaide
1405082lastname405082NaNNaNNaN62023-01-12 06:30:31.197484+01:002023-01-12 06:30:31.197484+01:00NaNFalse...15569924234050822023-01-11 17:08:41+01:003False13.0False2023-02-06 20:00:00+01:00zaide
2411168lastname411168NaNNaNNaN62023-03-17 06:30:35.431967+01:002023-03-17 06:30:35.431967+01:00NaNFalse...155610539344111682023-03-16 16:23:10+01:003False62.0False2023-03-19 16:00:00+01:00luisa miller
3411168lastname411168NaNNaNNaN62023-03-17 06:30:35.431967+01:002023-03-17 06:30:35.431967+01:00NaNFalse...155610539344111682023-03-16 16:23:10+01:003False62.0False2023-03-19 16:00:00+01:00luisa miller
44380lastname4380firstname4380NaNNaN12021-04-22 14:51:55.432952+02:002022-04-14 11:41:33.738500+02:00NaNFalse...1556118914143802020-11-26 13:12:53+01:003False51.3False2020-12-01 20:00:00+01:00iphigenie en tauride
..................................................................
31896419095lastname19095firstname190951979-07-16email1909562021-04-22 15:06:30.120537+02:002023-09-12 18:27:36.904104+02:00NaNFalse...15561090839190952019-05-19 21:18:36+02:001False4.5False2019-05-27 20:00:00+02:00entre femmes
31896519095lastname19095firstname190951979-07-16email1909562021-04-22 15:06:30.120537+02:002023-09-12 18:27:36.904104+02:00NaNFalse...15561090839190952019-05-19 21:18:36+02:001False4.5False2019-05-27 20:00:00+02:00entre femmes
31896619095lastname19095firstname190951979-07-16email1909562021-04-22 15:06:30.120537+02:002023-09-12 18:27:36.904104+02:00NaNFalse...15561090839190952019-05-19 21:18:36+02:001False4.5False2019-05-27 20:00:00+02:00entre femmes
31896719095lastname19095firstname190951979-07-16email1909562021-04-22 15:06:30.120537+02:002023-09-12 18:27:36.904104+02:00NaNFalse...15561244277190952019-12-31 11:04:07+01:001False5.5False2020-02-03 20:00:00+01:00a boire et a manger
31896819095lastname19095firstname190951979-07-16email1909562021-04-22 15:06:30.120537+02:002023-09-12 18:27:36.904104+02:00NaNFalse...15561244277190952019-12-31 11:04:07+01:001False5.5False2020-02-03 20:00:00+01:00a boire et a manger
\n", "

318969 rows × 52 columns

\n", "
" ], "text/plain": [ " id lastname firstname birthdate email \\\n", "0 405082 lastname405082 NaN NaN NaN \n", "1 405082 lastname405082 NaN NaN NaN \n", "2 411168 lastname411168 NaN NaN NaN \n", "3 411168 lastname411168 NaN NaN NaN \n", "4 4380 lastname4380 firstname4380 NaN NaN \n", "... ... ... ... ... ... \n", "318964 19095 lastname19095 firstname19095 1979-07-16 email19095 \n", "318965 19095 lastname19095 firstname19095 1979-07-16 email19095 \n", "318966 19095 lastname19095 firstname19095 1979-07-16 email19095 \n", "318967 19095 lastname19095 firstname19095 1979-07-16 email19095 \n", "318968 19095 lastname19095 firstname19095 1979-07-16 email19095 \n", "\n", " street_id created_at \\\n", "0 6 2023-01-12 06:30:31.197484+01:00 \n", "1 6 2023-01-12 06:30:31.197484+01:00 \n", "2 6 2023-03-17 06:30:35.431967+01:00 \n", "3 6 2023-03-17 06:30:35.431967+01:00 \n", "4 1 2021-04-22 14:51:55.432952+02:00 \n", "... ... ... \n", "318964 6 2021-04-22 15:06:30.120537+02:00 \n", "318965 6 2021-04-22 15:06:30.120537+02:00 \n", "318966 6 2021-04-22 15:06:30.120537+02:00 \n", "318967 6 2021-04-22 15:06:30.120537+02:00 \n", "318968 6 2021-04-22 15:06:30.120537+02:00 \n", "\n", " updated_at civility is_partner ... \\\n", "0 2023-01-12 06:30:31.197484+01:00 NaN False ... \n", "1 2023-01-12 06:30:31.197484+01:00 NaN False ... \n", "2 2023-03-17 06:30:35.431967+01:00 NaN False ... \n", "3 2023-03-17 06:30:35.431967+01:00 NaN False ... \n", "4 2022-04-14 11:41:33.738500+02:00 NaN False ... \n", "... ... ... ... ... \n", "318964 2023-09-12 18:27:36.904104+02:00 NaN False ... \n", "318965 2023-09-12 18:27:36.904104+02:00 NaN False ... \n", "318966 2023-09-12 18:27:36.904104+02:00 NaN False ... \n", "318967 2023-09-12 18:27:36.904104+02:00 NaN False ... \n", "318968 2023-09-12 18:27:36.904104+02:00 NaN False ... \n", "\n", " tenant_id id_x customer_id purchase_date type_of \\\n", "0 1556 992423 405082 2023-01-11 17:08:41+01:00 3 \n", "1 1556 992423 405082 2023-01-11 17:08:41+01:00 3 \n", "2 1556 1053934 411168 2023-03-16 16:23:10+01:00 3 \n", "3 1556 1053934 411168 2023-03-16 16:23:10+01:00 3 \n", "4 1556 1189141 4380 2020-11-26 13:12:53+01:00 3 \n", "... ... ... ... ... ... \n", "318964 1556 1090839 19095 2019-05-19 21:18:36+02:00 1 \n", "318965 1556 1090839 19095 2019-05-19 21:18:36+02:00 1 \n", "318966 1556 1090839 19095 2019-05-19 21:18:36+02:00 1 \n", "318967 1556 1244277 19095 2019-12-31 11:04:07+01:00 1 \n", "318968 1556 1244277 19095 2019-12-31 11:04:07+01:00 1 \n", "\n", " is_from_subscription amount is_full_price start_date_time \\\n", "0 False 13.0 False 2023-02-06 20:00:00+01:00 \n", "1 False 13.0 False 2023-02-06 20:00:00+01:00 \n", "2 False 62.0 False 2023-03-19 16:00:00+01:00 \n", "3 False 62.0 False 2023-03-19 16:00:00+01:00 \n", "4 False 51.3 False 2020-12-01 20:00:00+01:00 \n", "... ... ... ... ... \n", "318964 False 4.5 False 2019-05-27 20:00:00+02:00 \n", "318965 False 4.5 False 2019-05-27 20:00:00+02:00 \n", "318966 False 4.5 False 2019-05-27 20:00:00+02:00 \n", "318967 False 5.5 False 2020-02-03 20:00:00+01:00 \n", "318968 False 5.5 False 2020-02-03 20:00:00+01:00 \n", "\n", " event_name \n", "0 zaide \n", "1 zaide \n", "2 luisa miller \n", "3 luisa miller \n", "4 iphigenie en tauride \n", "... ... \n", "318964 entre femmes \n", "318965 entre femmes \n", "318966 entre femmes \n", "318967 a boire et a manger \n", "318968 a boire et a manger \n", "\n", "[318969 rows x 52 columns]" ] }, "execution_count": 12, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Jointure\n", "merge_1 = pd.merge(purchases, tickets, left_on='id', right_on='purchase_id', how='inner')[['id_x', 'customer_id','product_id', 'purchase_date', 'type_of', 'is_from_subscription']]\n", "merge_2 = pd.merge(products, merge_1, left_on='id', right_on='product_id', how='inner')[['id_x', 'customer_id', 'representation_id', 'purchase_date', 'type_of', 'is_from_subscription', 'amount', 'is_full_price']]\n", "merge_3 = pd.merge(representations, merge_2, left_on='id', right_on='representation_id', how='inner')[['id_x', 'customer_id', 'event_id', 'purchase_date', 'type_of', 'is_from_subscription', 'amount', 'is_full_price', 'start_date_time']]\n", "merge_4 = pd.merge(events, merge_3, left_on='id', right_on='event_id', how='inner')[['id_x', 'customer_id', 'purchase_date', 'type_of', 'is_from_subscription', 'amount', 'is_full_price', 'start_date_time', 'name']]\n", "merge_4 = merge_4.rename(columns={'name': 'event_name'})\n", "df_customer_event = pd.merge(customersplus, merge_4, left_on = 'id', right_on = 'customer_id', how = 'inner')[['id_x', 'purchase_date', 'type_of', 'is_from_subscription', 'amount', 'is_full_price', 'start_date_time', 'event_name']]\n", "df_customer_event" ] } ], "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": "3.10.13" } }, "nbformat": 4, "nbformat_minor": 5 }