{ "cells": [ { "cell_type": "markdown", "metadata": { "id": "1fBtzpkhbCjc" }, "source": [ "## Prerequisites\n", "\n", "Run this cell to prepare the environment. This step is obligatory." ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "!pip install retentioneering" ] }, { "cell_type": "markdown", "metadata": { "id": "gaOtTZvLbLsw" }, "source": [ "# Preprocessing" ] }, { "cell_type": "markdown", "metadata": { "id": "WgvQI7AGbO3Z" }, "source": [ "The full text of [Preprocessing graph](https://doc.retentioneering.com/stable/doc/user_guides/preprocessing.html) user guide is available on the retentioneering website." ] }, { "cell_type": "code", "execution_count": 2, "metadata": { "id": "01M2kmcXmBhk" }, "outputs": [], "source": [ "import pandas as pd\n", "from retentioneering import datasets\n", "\n", "stream = datasets.load_simple_shop()" ] }, { "cell_type": "markdown", "metadata": { "id": "nU5gm9OZyWFZ" }, "source": [ "## Preprocessing GUI" ] }, { "cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "
\n", "\n", "" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "pgraph = stream.preprocessing_graph()" ] }, { "cell_type": "code", "execution_count": 11, "metadata": { "id": "gXQmtFZdGj-8" }, "outputs": [], "source": [ "def group_products(df, schema):\n", " return df[schema.event_name].isin(['product1', 'product2'])" ] }, { "cell_type": "code", "execution_count": 12, "metadata": { "id": "HmEonwicHHTW" }, "outputs": [], "source": [ "def get_new_users(df, schema):\n", " new_users = df[df[schema.event_name] == 'new_user']\\\n", " [schema.user_id]\\\n", " .unique()\n", " return df[schema.user_id].isin(new_users)\n", "\n", "def get_existing_users(df, schema):\n", " existing_users = df[df[schema.event_name] == 'existing_user']\\\n", " [schema.user_id]\\\n", " .unique()\n", " return df[schema.user_id].isin(existing_users)" ] }, { "cell_type": "code", "execution_count": 13, "metadata": { "id": "PMEB_Py1HZ3I" }, "outputs": [], "source": [ "def remove_truncated_paths(df, schema):\n", " truncated_users = df[df[schema.event_name].isin(['cropped_left', 'cropped_right'])]\\\n", " [schema.user_id]\\\n", " .unique()\n", " return ~df[schema.user_id].isin(truncated_users)" ] }, { "cell_type": "code", "execution_count": 14, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 423 }, "executionInfo": { "elapsed": 233, "status": "ok", "timestamp": 1682610476218, "user": { "displayName": "Julia Ostanina", "userId": "13687663492290466770" }, "user_tz": -120 }, "id": "6yWA_SWmIUBZ", "outputId": "5a0a60ce-640b-4cbf-da3a-41a83a20e1a0" }, "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", "
event_idevent_typeevent_indexeventtimestampuser_idsession_id
0ca9cc3ed-5085-4a60-bb8e-038740cb8d6epath_start0path_start2019-11-01 17:59:13.273932219483890.0219483890.0_1
1f4b7bcee-65f1-4cbf-b658-179a2e37266eexisting_user0existing_user2019-11-01 17:59:13.273932219483890.0219483890.0_1
2da1231c6-e117-4e43-9e21-0af016cd1fbesession_start0session_start2019-11-01 17:59:13.273932219483890.0219483890.0_1
3bb732172-8b25-4204-94e7-af80418abca5raw0catalog2019-11-01 17:59:13.273932219483890.0219483890.0_1
4dd4b2d2c-816d-4a08-a555-11b6a7cb6f12group_alias1product2019-11-01 17:59:28.459271219483890.0219483890.0_1
........................
5640218992bc9-a756-4457-b4d9-fba17840b99braw32280catalog2020-04-29 12:48:01.809577501098384.0501098384.0_6
5640371bcf87f-f69c-421c-92a2-0b24e4ed9268raw32281main2020-04-29 12:48:01.938488501098384.0501098384.0_6
564042805ba82-ba52-4cce-a23b-e86f9a8d9d86raw32282catalog2020-04-29 12:48:06.595390501098384.0501098384.0_6
56405403258be-3bb5-405d-bdc3-c2f69c25e56dsession_end32282session_end2020-04-29 12:48:06.595390501098384.0501098384.0_6
5640651fd4cdc-eeb4-481d-8d20-3770ebba03aapath_end32282path_end2020-04-29 12:48:06.595390501098384.0501098384.0_6
\n", "

56407 rows × 7 columns

\n", "
" ], "text/plain": [ " event_id event_type event_index \\\n", "0 ca9cc3ed-5085-4a60-bb8e-038740cb8d6e path_start 0 \n", "1 f4b7bcee-65f1-4cbf-b658-179a2e37266e existing_user 0 \n", "2 da1231c6-e117-4e43-9e21-0af016cd1fbe session_start 0 \n", "3 bb732172-8b25-4204-94e7-af80418abca5 raw 0 \n", "4 dd4b2d2c-816d-4a08-a555-11b6a7cb6f12 group_alias 1 \n", "... ... ... ... \n", "56402 18992bc9-a756-4457-b4d9-fba17840b99b raw 32280 \n", "56403 71bcf87f-f69c-421c-92a2-0b24e4ed9268 raw 32281 \n", "56404 2805ba82-ba52-4cce-a23b-e86f9a8d9d86 raw 32282 \n", "56405 403258be-3bb5-405d-bdc3-c2f69c25e56d session_end 32282 \n", "56406 51fd4cdc-eeb4-481d-8d20-3770ebba03aa path_end 32282 \n", "\n", " event timestamp user_id session_id \n", "0 path_start 2019-11-01 17:59:13.273932 219483890.0 219483890.0_1 \n", "1 existing_user 2019-11-01 17:59:13.273932 219483890.0 219483890.0_1 \n", "2 session_start 2019-11-01 17:59:13.273932 219483890.0 219483890.0_1 \n", "3 catalog 2019-11-01 17:59:13.273932 219483890.0 219483890.0_1 \n", "4 product 2019-11-01 17:59:28.459271 219483890.0 219483890.0_1 \n", "... ... ... ... ... \n", "56402 catalog 2020-04-29 12:48:01.809577 501098384.0 501098384.0_6 \n", "56403 main 2020-04-29 12:48:01.938488 501098384.0 501098384.0_6 \n", "56404 catalog 2020-04-29 12:48:06.595390 501098384.0 501098384.0_6 \n", "56405 session_end 2020-04-29 12:48:06.595390 501098384.0 501098384.0_6 \n", "56406 path_end 2020-04-29 12:48:06.595390 501098384.0 501098384.0_6 \n", "\n", "[56407 rows x 7 columns]" ] }, "execution_count": 14, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# press 'Combine'/'Save & Combine' button in GUI before running this cell\n", "pgraph.combine_result.to_dataframe()" ] }, { "cell_type": "markdown", "metadata": { "id": "tPnzp04Wtj1h" }, "source": [ "## Code-generated preprocessing graph\n" ] }, { "cell_type": "code", "execution_count": 12, "metadata": { "id": "ZFP_E3vBnsMA" }, "outputs": [], "source": [ "from retentioneering.preprocessing_graph import PreprocessingGraph\n", "\n", "pgraph = PreprocessingGraph(stream)" ] }, { "cell_type": "markdown", "metadata": { "id": "2pUaMVnGuLew" }, "source": [ "### Creating a single node" ] }, { "cell_type": "code", "execution_count": 13, "metadata": { "id": "zImapuA3bHA0" }, "outputs": [], "source": [ "from retentioneering.preprocessing_graph import EventsNode\n", "from retentioneering.data_processors_lib import GroupEvents, GroupEventsParams\n", "\n", "def group_products(df, schema):\n", " return df[schema.event_name].isin(['product1', 'product2'])\n", "\n", "group_events_params = {\n", " \"event_name\": \"product\",\n", " \"func\": group_products\n", "}\n", "\n", "data_processor_params = GroupEventsParams(**group_events_params)\n", "data_processor = GroupEvents(params=data_processor_params)\n", "node2 = EventsNode(data_processor)" ] }, { "cell_type": "code", "execution_count": 14, "metadata": { "id": "jtko6iebTvTM" }, "outputs": [], "source": [ "node2 = EventsNode(GroupEvents(params=GroupEventsParams(**group_events_params)))" ] }, { "cell_type": "code", "execution_count": 15, "metadata": { "id": "5YA97SSQXqZp" }, "outputs": [], "source": [ "from retentioneering.data_processors_lib import AddStartEndEvents, AddStartEndEventsParams\n", "\n", "node1 = EventsNode(AddStartEndEvents(params=AddStartEndEventsParams()))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Linking nodes" ] }, { "cell_type": "code", "execution_count": 16, "metadata": { "id": "i9xprcHxT8Aj" }, "outputs": [], "source": [ "pgraph.add_node(node=node1, parents=[pgraph.root])\n", "pgraph.add_node(node=node2, parents=[node1])" ] }, { "cell_type": "markdown", "metadata": { "id": "j5z5__VjZ2Bp" }, "source": [ "### Building the whole graph" ] }, { "cell_type": "code", "execution_count": 17, "metadata": { "id": "5GPmlBdaZyHf" }, "outputs": [], "source": [ "from retentioneering.data_processors_lib import LabelNewUsers, LabelNewUsersParams\n", "\n", "google_spreadsheet_id = '1iggpIT5CZcLILLZ94wCZPQv90tERwi1IB5Y1969C8zc'\n", "link = f'https://docs.google.com/spreadsheets/u/1/d/{google_spreadsheet_id}/export?format=csv&id={google_spreadsheet_id}'\n", "new_users = pd.read_csv(link, header=None)[0].tolist()\n", "node3 = EventsNode(LabelNewUsers(params=LabelNewUsersParams(new_users_list=new_users)))\n", "pgraph.add_node(node=node3, parents=[node2])" ] }, { "cell_type": "code", "execution_count": 18, "metadata": { "id": "h3q8inbaTurN" }, "outputs": [], "source": [ "from retentioneering.data_processors_lib import FilterEvents, FilterEventsParams\n", "\n", "def get_new_users(df, schema):\n", " new_users = df[df[schema.event_name] == 'new_user']\\\n", " [schema.user_id]\\\n", " .unique()\n", " return df[schema.user_id].isin(new_users)\n", "\n", "def get_existing_users(df, schema):\n", " existing_users = df[df[schema.event_name] == 'existing_user']\\\n", " [schema.user_id]\\\n", " .unique()\n", " return df[schema.user_id].isin(existing_users)\n", "\n", "\n", "node4 = EventsNode(FilterEvents(params=FilterEventsParams(func=get_new_users)))\n", "node5 = EventsNode(FilterEvents(params=FilterEventsParams(func=get_existing_users)))\n", "pgraph.add_node(node=node4, parents=[node3])\n", "pgraph.add_node(node=node5, parents=[node3])" ] }, { "cell_type": "code", "execution_count": 19, "metadata": { "id": "5CFDNCo3eUv3" }, "outputs": [], "source": [ "from retentioneering.data_processors_lib import LabelCroppedPaths, LabelCroppedPathsParams\n", "\n", "params = {\n", " \"left_cutoff\": (1, 'h'),\n", " \"right_cutoff\": (1, 'h'),\n", "}\n", "node6 = EventsNode(LabelCroppedPaths(params=LabelCroppedPathsParams(**params)))\n", "pgraph.add_node(node=node6, parents=[node5])" ] }, { "cell_type": "code", "execution_count": 20, "metadata": { "id": "ZAHfra7hfEKZ" }, "outputs": [], "source": [ "def remove_truncated_paths(df, schema):\n", " truncated_users = df[df[schema.event_name].isin(['cropped_left', 'cropped_right'])]\\\n", " [schema.user_id]\\\n", " .unique()\n", " return ~df[schema.user_id].isin(truncated_users)\n", "\n", "node7 = EventsNode(FilterEvents(params=FilterEventsParams(func=remove_truncated_paths)))\n", "pgraph.add_node(node=node7, parents=[node6])" ] }, { "cell_type": "code", "execution_count": 21, "metadata": { "id": "8IPCIwn6fNwa" }, "outputs": [], "source": [ "from retentioneering.preprocessing_graph import MergeNode\n", "\n", "node8 = MergeNode()\n", "pgraph.add_node(node=node8, parents=[node4, node7])" ] }, { "cell_type": "code", "execution_count": 22, "metadata": { "id": "3gp2ClQZfvB6" }, "outputs": [], "source": [ "from retentioneering.data_processors_lib import SplitSessions, SplitSessionsParams\n", "\n", "node9 = EventsNode(SplitSessions(params=SplitSessionsParams(timeout=(30, 'm'))))\n", "pgraph.add_node(node=node9, parents=[node8])" ] }, { "cell_type": "markdown", "metadata": { "id": "NnuX-N8VgI70" }, "source": [ "### Running the calculation\n" ] }, { "cell_type": "code", "execution_count": 23, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 608 }, "executionInfo": { "elapsed": 12697, "status": "ok", "timestamp": 1682611683371, "user": { "displayName": "Julia Ostanina", "userId": "13687663492290466770" }, "user_tz": -120 }, "id": "y508oKhMgHmQ", "outputId": "049fb227-981a-40b2-b4b2-842fa9a4b4f1", "scrolled": true }, "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", "
event_idevent_typeevent_indexeventtimestampuser_idsession_id
05f3c4b0d-36d5-47b5-8f2d-332aa6e206d8path_start0path_start2019-11-01 17:59:13.273932219483890.0219483890.0_1
1814840bf-27a3-45b4-bd18-10358a48fa1fexisting_user0existing_user2019-11-01 17:59:13.273932219483890.0219483890.0_1
25190b07d-238b-4907-866b-ee7811757f7dsession_start0session_start2019-11-01 17:59:13.273932219483890.0219483890.0_1
3c0f910d4-c2c9-4baf-8300-19a74f7a2b0draw0catalog2019-11-01 17:59:13.273932219483890.0219483890.0_1
47933f7d8-fca4-4ed8-b7f3-5436b13e3ca1group_alias1product2019-11-01 17:59:28.459271219483890.0219483890.0_1
\n", "
" ], "text/plain": [ " event_id event_type event_index \\\n", "0 5f3c4b0d-36d5-47b5-8f2d-332aa6e206d8 path_start 0 \n", "1 814840bf-27a3-45b4-bd18-10358a48fa1f existing_user 0 \n", "2 5190b07d-238b-4907-866b-ee7811757f7d session_start 0 \n", "3 c0f910d4-c2c9-4baf-8300-19a74f7a2b0d raw 0 \n", "4 7933f7d8-fca4-4ed8-b7f3-5436b13e3ca1 group_alias 1 \n", "\n", " event timestamp user_id session_id \n", "0 path_start 2019-11-01 17:59:13.273932 219483890.0 219483890.0_1 \n", "1 existing_user 2019-11-01 17:59:13.273932 219483890.0 219483890.0_1 \n", "2 session_start 2019-11-01 17:59:13.273932 219483890.0 219483890.0_1 \n", "3 catalog 2019-11-01 17:59:13.273932 219483890.0 219483890.0_1 \n", "4 product 2019-11-01 17:59:28.459271 219483890.0 219483890.0_1 " ] }, "execution_count": 23, "metadata": {}, "output_type": "execute_result" } ], "source": [ "processed_stream = pgraph.combine(node=node9)\n", "processed_stream.to_dataframe().head()" ] }, { "cell_type": "markdown", "metadata": { "id": "ULU7yZc3hsDV" }, "source": [ "### Summary" ] }, { "cell_type": "markdown", "metadata": { "id": "SEV_C9YJh3j0" }, "source": [ "Here we just provide the same code combined in a single chunk so you could simply copy and paste it and see the results." ] }, { "cell_type": "code", "execution_count": 24, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 608 }, "executionInfo": { "elapsed": 14967, "status": "ok", "timestamp": 1682611824536, "user": { "displayName": "Julia Ostanina", "userId": "13687663492290466770" }, "user_tz": -120 }, "id": "CS9LQhcLgzQx", "outputId": "df26c4b9-ed31-4fca-ec92-51f344f0e14b" }, "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", "
event_idevent_typeevent_indexeventtimestampuser_idsession_id
0ccef9a9f-112e-4231-a1ed-109d9ca79c4apath_start0path_start2019-11-01 17:59:13.273932219483890.0219483890.0_1
1d491bc30-e638-452f-9d34-0644c310e860existing_user0existing_user2019-11-01 17:59:13.273932219483890.0219483890.0_1
26037d2f2-4526-432f-9335-5b1a553dbb2esession_start0session_start2019-11-01 17:59:13.273932219483890.0219483890.0_1
3bb1f68c8-dd15-4446-b940-f9ec9688d943raw0catalog2019-11-01 17:59:13.273932219483890.0219483890.0_1
4e19e9d4b-029e-4e8b-b077-16b77c25578cgroup_alias1product2019-11-01 17:59:28.459271219483890.0219483890.0_1
\n", "
" ], "text/plain": [ " event_id event_type event_index \\\n", "0 ccef9a9f-112e-4231-a1ed-109d9ca79c4a path_start 0 \n", "1 d491bc30-e638-452f-9d34-0644c310e860 existing_user 0 \n", "2 6037d2f2-4526-432f-9335-5b1a553dbb2e session_start 0 \n", "3 bb1f68c8-dd15-4446-b940-f9ec9688d943 raw 0 \n", "4 e19e9d4b-029e-4e8b-b077-16b77c25578c group_alias 1 \n", "\n", " event timestamp user_id session_id \n", "0 path_start 2019-11-01 17:59:13.273932 219483890.0 219483890.0_1 \n", "1 existing_user 2019-11-01 17:59:13.273932 219483890.0 219483890.0_1 \n", "2 session_start 2019-11-01 17:59:13.273932 219483890.0 219483890.0_1 \n", "3 catalog 2019-11-01 17:59:13.273932 219483890.0 219483890.0_1 \n", "4 product 2019-11-01 17:59:28.459271 219483890.0 219483890.0_1 " ] }, "execution_count": 24, "metadata": {}, "output_type": "execute_result" } ], "source": [ "import pandas as pd\n", "from retentioneering import datasets\n", "from retentioneering.data_processors_lib import AddStartEndEvents, AddStartEndEventsParams\n", "from retentioneering.data_processors_lib import GroupEvents, GroupEventsParams\n", "from retentioneering.data_processors_lib import LabelNewUsers, LabelNewUsersParams\n", "from retentioneering.data_processors_lib import FilterEvents, FilterEventsParams\n", "from retentioneering.data_processors_lib import LabelCroppedPaths, LabelCroppedPathsParams\n", "from retentioneering.data_processors_lib import SplitSessions, SplitSessionsParams\n", "from retentioneering.preprocessing_graph import PreprocessingGraph, EventsNode, MergeNode\n", "\n", "stream = datasets.load_simple_shop()\n", "\n", "# node1\n", "node1 = EventsNode(AddStartEndEvents(params=AddStartEndEventsParams()))\n", "\n", "# node2\n", "def group_products(df, schema):\n", " return df[schema.event_name].isin(['product1', 'product2'])\n", "\n", "group_events_params={\n", " \"event_name\": \"product\",\n", " \"func\": group_products\n", "}\n", "node2 = EventsNode(GroupEvents(params=GroupEventsParams(**group_events_params)))\n", "\n", "# node3\n", "google_spreadsheet_id = '1iggpIT5CZcLILLZ94wCZPQv90tERwi1IB5Y1969C8zc'\n", "link = f'https://docs.google.com/spreadsheets/u/1/d/{google_spreadsheet_id}/export?format=csv&id={google_spreadsheet_id}'\n", "new_users = pd.read_csv(link, header=None)[0].tolist()\n", "node3 = EventsNode(LabelNewUsers(params=LabelNewUsersParams(new_users_list=new_users)))\n", "\n", "# node4, node5\n", "def get_new_users(df, schema):\n", " new_users = df[df[schema.event_name] == 'new_user']\\\n", " [schema.user_id]\\\n", " .unique()\n", " return df[schema.user_id].isin(new_users)\n", "\n", "def get_existing_users(df, schema):\n", " existing_users = df[df[schema.event_name] == 'existing_user']\\\n", " [schema.user_id]\\\n", " .unique()\n", " return df[schema.user_id].isin(existing_users)\n", "\n", "node4 = EventsNode(FilterEvents(params=FilterEventsParams(func=get_new_users)))\n", "node5 = EventsNode(FilterEvents(params=FilterEventsParams(func=get_existing_users)))\n", "\n", "# node6\n", "params = {\n", " \"left_cutoff\": (1, 'h'),\n", " \"right_cutoff\": (1, 'h'),\n", "}\n", "node6 = EventsNode(LabelCroppedPaths(params=LabelCroppedPathsParams(**params)))\n", "\n", "# node7, node8, node9\n", "def remove_truncated_paths(df, schema):\n", " truncated_users = df[df[schema.event_name].isin(['cropped_left', 'cropped_right'])]\\\n", " [schema.user_id]\\\n", " .unique()\n", " return ~df[schema.user_id].isin(truncated_users)\n", "\n", "node7 = EventsNode(FilterEvents(params=FilterEventsParams(func=remove_truncated_paths)))\n", "node8 = MergeNode()\n", "node9 = EventsNode(SplitSessions(params=SplitSessionsParams(timeout=(30, 'm'))))\n", "\n", "# linking the nodes to get the graph\n", "pgraph = PreprocessingGraph(stream)\n", "pgraph.add_node(node=node1, parents=[pgraph.root])\n", "pgraph.add_node(node=node2, parents=[node1])\n", "pgraph.add_node(node=node3, parents=[node2])\n", "pgraph.add_node(node=node4, parents=[node3])\n", "pgraph.add_node(node=node5, parents=[node3])\n", "pgraph.add_node(node=node6, parents=[node5])\n", "pgraph.add_node(node=node7, parents=[node6])\n", "pgraph.add_node(node=node8, parents=[node4, node7])\n", "pgraph.add_node(node=node9, parents=[node8])\n", "\n", "# getting the calculation results\n", "processed_stream = pgraph.combine(node=node9)\n", "processed_stream.to_dataframe().head()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Saving preprocessing config" ] }, { "cell_type": "code", "execution_count": 17, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "
\n", "\n", "" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# save preprocessing config to file\n", "pgraph.export_to_file('/path/to/pgraph_config.json')\n", "\n", "# load data and create new PreprocessingGraph instance\n", "new_stream = datasets.load_simple_shop()\n", "new_pgraph = new_stream.preprocessing_graph()\n", "\n", "# restore the preserved preprocessing configuration\n", "new_pgraph.import_from_file('/path/to/pgraph_config.json')" ] }, { "cell_type": "markdown", "metadata": { "id": "hzekk6SBorGQ" }, "source": [ "Method chaining preprocessing graph\n", "-----------------------------------" ] }, { "cell_type": "code", "execution_count": 25, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 942 }, "executionInfo": { "elapsed": 5, "status": "ok", "timestamp": 1682611824536, "user": { "displayName": "Julia Ostanina", "userId": "13687663492290466770" }, "user_tz": -120 }, "id": "3229EogKow_k", "outputId": "77d5bce6-6a9d-42e2-ff8c-f937709c2fdc" }, "outputs": [ { "data": { "text/html": [ "\n", "\n", "\n", "" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/plain": [ "" ] }, "execution_count": 25, "metadata": {}, "output_type": "execute_result" } ], "source": [ "stream.transition_graph(edges_norm_type='node')" ] }, { "cell_type": "code", "execution_count": 26, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "executionInfo": { "elapsed": 5052, "status": "ok", "timestamp": 1682612027013, "user": { "displayName": "Julia Ostanina", "userId": "13687663492290466770" }, "user_tz": -120 }, "id": "AWB57NQdsBLB", "outputId": "93350811-f505-4ee4-f00e-97d16080b3cc" }, "outputs": [], "source": [ "def group_browsing(df, schema):\n", " return df[schema.event_name].isin(['catalog', 'main'])\n", "\n", "def group_products(df, schema):\n", " return df[schema.event_name].isin(['product1', 'product2'])\n", "\n", "def group_delivery(df, schema):\n", " return df[schema.event_name].isin(['delivery_choice', 'delivery_courier', 'delivery_pickup'])\n", "\n", "def group_payment(df, schema):\n", " return df[schema.event_name].isin(['payment_choice', 'payment_done', 'payment_card', 'payment_cash'])\n", "\n", "stream_7_nodes = stream\\\n", " .drop_paths(min_steps=6)\\\n", " .add_start_end_events()\\\n", " .split_sessions(timeout=(30, 'm'))\\\n", " .group_events(event_name='browsing', func=group_browsing)\\\n", " .group_events(event_name='delivery', func=group_delivery)\\\n", " .group_events(event_name='payment', func=group_payment)" ] }, { "cell_type": "code", "execution_count": 27, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 997 }, "executionInfo": { "elapsed": 3319, "status": "ok", "timestamp": 1682612030331, "user": { "displayName": "Julia Ostanina", "userId": "13687663492290466770" }, "user_tz": -120 }, "id": "Ppv7B5PGo-pD", "outputId": "d197fb0c-403f-4e89-d4ca-5e0fa1d175e7" }, "outputs": [ { "data": { "text/html": [ "\n", "\n", "\n", "" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/plain": [ "" ] }, "execution_count": 27, "metadata": {}, "output_type": "execute_result" } ], "source": [ "stream_out = stream_7_nodes.collapse_loops()\n", "stream_out.transition_graph(edges_norm_type='node')" ] }, { "cell_type": "code", "execution_count": 28, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 997 }, "executionInfo": { "elapsed": 6740, "status": "ok", "timestamp": 1682612083442, "user": { "displayName": "Julia Ostanina", "userId": "13687663492290466770" }, "user_tz": -120 }, "id": "K8WXtrzAsZEF", "outputId": "233bb24f-1706-439e-b0a2-cdebfb80328c" }, "outputs": [ { "data": { "text/html": [ "\n", "\n", "\n", "" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/plain": [ "" ] }, "execution_count": 28, "metadata": {}, "output_type": "execute_result" } ], "source": [ "stream_out = stream_7_nodes.collapse_loops(suffix='loop')\n", "stream_out.transition_graph(edges_norm_type='node')" ] }, { "cell_type": "code", "execution_count": 29, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 980 }, "executionInfo": { "elapsed": 3476, "status": "ok", "timestamp": 1682612103347, "user": { "displayName": "Julia Ostanina", "userId": "13687663492290466770" }, "user_tz": -120 }, "id": "THGKp7-Ot5m3", "outputId": "2466920f-18d9-4e2d-bb36-3a3c0a3adbc9" }, "outputs": [ { "data": { "text/html": [ "\n", "\n", "\n", "" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "stream_out = stream_7_nodes.collapse_loops(suffix='count')\n", "stream_out.transition_graph();" ] } ], "metadata": { "colab": { "provenance": [], "toc_visible": true }, "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": "3.8.15" }, "varInspector": { "cols": { "lenName": 16, "lenType": 16, "lenVar": 40 }, "kernels_config": { "python": { "delete_cmd_postfix": "", "delete_cmd_prefix": "del ", "library": "var_list.py", "varRefreshCmd": "print(var_dic_list())" }, "r": { "delete_cmd_postfix": ") ", "delete_cmd_prefix": "rm(", "library": "var_list.r", "varRefreshCmd": "cat(var_dic_list()) " } }, "types_to_exclude": [ "module", "function", "builtin_function_or_method", "instance", "_Feature" ], "window_display": false } }, "nbformat": 4, "nbformat_minor": 4 }