diff --git a/docs/source/notebooks/clv/bg_nbd.ipynb b/docs/source/notebooks/clv/bg_nbd.ipynb deleted file mode 100644 index 27534b9fc..000000000 --- a/docs/source/notebooks/clv/bg_nbd.ipynb +++ /dev/null @@ -1,4292 +0,0 @@ -{ - "cells": [ - { - "cell_type": "markdown", - "id": "51e3591e", - "metadata": {}, - "source": [ - "# BG/NBD Model\n", - "\n", - "In this notebook we show how to fit a BG/NBD model in PyMC-Marketing. We compare the results with the [`lifetimes`](https://github.com/CamDavidsonPilon/lifetimes) package (no longer maintained). The model is presented in the paper: Fader, P. S., Hardie, B. G., & Lee, K. L. (2005). [“Counting your customers” the easy way: An alternative to the Pareto/NBD model. Marketing science, 24(2), 275-284.](http://www.brucehardie.com/papers/bgnbd_2004-04-20.pdf)" - ] - }, - { - "cell_type": "markdown", - "id": "68f7ba7e", - "metadata": {}, - "source": [ - "## Prepare Notebook" - ] - }, - { - "cell_type": "code", - "execution_count": 7, - "id": "81c950fb", - "metadata": {}, - "outputs": [], - "source": [ - "import arviz as az\n", - "import matplotlib.pyplot as plt\n", - "import pandas as pd\n", - "import xarray as xr\n", - "from fastprogress.fastprogress import progress_bar\n", - "from lifetimes import BetaGeoFitter\n", - "\n", - "from pymc_marketing import clv\n", - "\n", - "# Plotting configuration\n", - "az.style.use(\"arviz-darkgrid\")\n", - "plt.rcParams[\"figure.figsize\"] = [12, 7]\n", - "plt.rcParams[\"figure.dpi\"] = 100\n", - "plt.rcParams[\"figure.facecolor\"] = \"white\"\n", - "\n", - "%load_ext autoreload\n", - "%autoreload 2\n", - "%config InlineBackend.figure_format = \"retina\"" - ] - }, - { - "cell_type": "markdown", - "id": "6e4b3b25", - "metadata": {}, - "source": [ - "## Read Data\n", - "\n", - "We use the `CDNOW` dataset (see lifetimes [quick-start](https://lifetimes.readthedocs.io/en/latest/Quickstart.html))." - ] - }, - { - "cell_type": "code", - "execution_count": 8, - "id": "a99638b5", - "metadata": {}, - "outputs": [ - { - "data": { - "text/html": [ - "
\n", - " | frequency | \n", - "recency | \n", - "T | \n", - "monetary_value | \n", - "
---|---|---|---|---|
0 | \n", - "2 | \n", - "30.43 | \n", - "38.86 | \n", - "22.35 | \n", - "
1 | \n", - "1 | \n", - "1.71 | \n", - "38.86 | \n", - "11.77 | \n", - "
2 | \n", - "0 | \n", - "0.00 | \n", - "38.86 | \n", - "0.00 | \n", - "
3 | \n", - "0 | \n", - "0.00 | \n", - "38.86 | \n", - "0.00 | \n", - "
4 | \n", - "0 | \n", - "0.00 | \n", - "38.86 | \n", - "0.00 | \n", - "
<xarray.Dataset> Size: 264kB\n", - "Dimensions: (chain: 4, draw: 2000)\n", - "Coordinates:\n", - " * chain (chain) int32 16B 0 1 2 3\n", - " * draw (draw) int32 8kB 0 1 2 3 4 5 6 ... 1994 1995 1996 1997 1998 1999\n", - "Data variables:\n", - " a (chain, draw) float64 64kB 0.8818 1.206 1.286 ... 1.069 0.6795\n", - " b (chain, draw) float64 64kB 3.302 4.155 3.621 ... 6.614 3.475 2.11\n", - " alpha (chain, draw) float64 64kB 5.412 3.828 3.875 ... 5.134 4.236 4.289\n", - " r (chain, draw) float64 64kB 0.2699 0.222 0.2225 ... 0.2322 0.2335\n", - "Attributes:\n", - " created_at: 2024-10-27T20:00:52.300444+00:00\n", - " arviz_version: 0.20.0\n", - " inference_library: pymc\n", - " inference_library_version: 5.15.1\n", - " sampling_time: 285.2232115268707\n", - " tuning_steps: 1000
<xarray.Dataset> Size: 984kB\n", - "Dimensions: (chain: 4, draw: 2000)\n", - "Coordinates:\n", - " * chain (chain) int32 16B 0 1 2 3\n", - " * draw (draw) int32 8kB 0 1 2 3 4 ... 1996 1997 1998 1999\n", - "Data variables: (12/17)\n", - " reached_max_treedepth (chain, draw) bool 8kB False False ... False False\n", - " acceptance_rate (chain, draw) float64 64kB 0.8957 1.0 ... 0.9759\n", - " tree_depth (chain, draw) int64 64kB 3 4 2 2 4 4 ... 5 5 3 5 4 4\n", - " max_energy_error (chain, draw) float64 64kB 0.2757 -0.2003 ... 0.04974\n", - " step_size (chain, draw) float64 64kB 0.2515 0.2515 ... 0.2934\n", - " step_size_bar (chain, draw) float64 64kB 0.2619 0.2619 ... 0.2359\n", - " ... ...\n", - " n_steps (chain, draw) float64 64kB 7.0 15.0 3.0 ... 15.0 15.0\n", - " smallest_eigval (chain, draw) float64 64kB nan nan nan ... nan nan\n", - " process_time_diff (chain, draw) float64 64kB 0.04688 0.0625 ... 0.04688\n", - " index_in_trajectory (chain, draw) int64 64kB 4 -14 -1 -2 ... -6 -14 12 12\n", - " energy_error (chain, draw) float64 64kB 0.1063 -0.1665 ... 0.04764\n", - " perf_counter_start (chain, draw) float64 64kB 1.632e+06 ... 1.633e+06\n", - "Attributes:\n", - " created_at: 2024-10-27T20:00:52.380448+00:00\n", - " arviz_version: 0.20.0\n", - " inference_library: pymc\n", - " inference_library_version: 5.15.1\n", - " sampling_time: 285.2232115268707\n", - " tuning_steps: 1000
<xarray.Dataset> Size: 94kB\n", - "Dimensions: (index: 2357)\n", - "Coordinates:\n", - " * index (index) int64 19kB 0 1 2 3 4 5 ... 2352 2353 2354 2355 2356\n", - "Data variables:\n", - " customer_id (index) int64 19kB 0 1 2 3 4 5 ... 2352 2353 2354 2355 2356\n", - " frequency (index) int64 19kB 2 1 0 0 0 7 1 0 2 0 ... 7 1 2 0 0 0 5 0 4 0\n", - " recency (index) float64 19kB 30.43 1.71 0.0 0.0 ... 24.29 0.0 26.57 0.0\n", - " T (index) float64 19kB 38.86 38.86 38.86 38.86 ... 27.0 27.0 27.0
\n", - " | mean | \n", - "sd | \n", - "hdi_3% | \n", - "hdi_97% | \n", - "mcse_mean | \n", - "mcse_sd | \n", - "ess_bulk | \n", - "ess_tail | \n", - "r_hat | \n", - "
---|---|---|---|---|---|---|---|---|---|
a | \n", - "0.978 | \n", - "0.288 | \n", - "0.536 | \n", - "1.498 | \n", - "0.006 | \n", - "0.004 | \n", - "2999.0 | \n", - "2254.0 | \n", - "1.0 | \n", - "
b | \n", - "3.215 | \n", - "1.211 | \n", - "1.465 | \n", - "5.335 | \n", - "0.024 | \n", - "0.018 | \n", - "3023.0 | \n", - "2487.0 | \n", - "1.0 | \n", - "
alpha | \n", - "4.489 | \n", - "0.388 | \n", - "3.769 | \n", - "5.215 | \n", - "0.006 | \n", - "0.004 | \n", - "3906.0 | \n", - "4270.0 | \n", - "1.0 | \n", - "
r | \n", - "0.244 | \n", - "0.013 | \n", - "0.221 | \n", - "0.270 | \n", - "0.000 | \n", - "0.000 | \n", - "4012.0 | \n", - "3793.0 | \n", - "1.0 | \n", - "
\n", - "\n" - ], - "text/plain": [ - "\n" - ] - }, - "metadata": {}, - "output_type": "display_data" - }, - { - "data": { - "text/html": [ - "\n" - ], - "text/plain": [] - }, - "metadata": {}, - "output_type": "display_data" - } - ], - "source": [ - "model_map = clv.BetaGeoModel(data=data)\n", - "idata_map = model_map.fit(fit_method=\"map\")" - ] - }, - { - "cell_type": "code", - "execution_count": 17, - "id": "c3d3eb19", - "metadata": {}, - "outputs": [ - { - "data": { - "text/html": [ - "\n", - "
<xarray.Dataset> Size: 40B\n", - "Dimensions: (chain: 1, draw: 1)\n", - "Coordinates:\n", - " * chain (chain) int32 4B 0\n", - " * draw (draw) int32 4B 0\n", - "Data variables:\n", - " a (chain, draw) float64 8B 0.793\n", - " b (chain, draw) float64 8B 2.426\n", - " alpha (chain, draw) float64 8B 4.414\n", - " r (chain, draw) float64 8B 0.2426\n", - "Attributes:\n", - " created_at: 2024-10-27T20:01:20.695750+00:00\n", - " arviz_version: 0.20.0\n", - " inference_library: pymc\n", - " inference_library_version: 5.15.1
<xarray.Dataset> Size: 94kB\n", - "Dimensions: (index: 2357)\n", - "Coordinates:\n", - " * index (index) int64 19kB 0 1 2 3 4 5 ... 2352 2353 2354 2355 2356\n", - "Data variables:\n", - " customer_id (index) int64 19kB 0 1 2 3 4 5 ... 2352 2353 2354 2355 2356\n", - " frequency (index) int64 19kB 2 1 0 0 0 7 1 0 2 0 ... 7 1 2 0 0 0 5 0 4 0\n", - " recency (index) float64 19kB 30.43 1.71 0.0 0.0 ... 24.29 0.0 26.57 0.0\n", - " T (index) float64 19kB 38.86 38.86 38.86 38.86 ... 27.0 27.0 27.0
\n", - " | coef | \n", - "se(coef) | \n", - "lower 95% bound | \n", - "upper 95% bound | \n", - "
---|---|---|---|---|
r | \n", - "0.242593 | \n", - "0.012557 | \n", - "0.217981 | \n", - "0.267205 | \n", - "
alpha | \n", - "4.413532 | \n", - "0.378221 | \n", - "3.672218 | \n", - "5.154846 | \n", - "
a | \n", - "0.792886 | \n", - "0.185719 | \n", - "0.428877 | \n", - "1.156895 | \n", - "
b | \n", - "2.425752 | \n", - "0.705345 | \n", - "1.043276 | \n", - "3.808229 | \n", - "
\n", - " | customer_id | \n", - "frequency | \n", - "recency | \n", - "T | \n", - "
---|---|---|---|---|
1 | \n", - "1 | \n", - "1 | \n", - "1.71 | \n", - "38.86 | \n", - "
6 | \n", - "6 | \n", - "1 | \n", - "5.00 | \n", - "38.86 | \n", - "
10 | \n", - "10 | \n", - "5 | \n", - "24.43 | \n", - "38.86 | \n", - "
18 | \n", - "18 | \n", - "3 | \n", - "28.29 | \n", - "38.71 | \n", - "
45 | \n", - "45 | \n", - "12 | \n", - "34.43 | \n", - "38.57 | \n", - "
1412 | \n", - "1412 | \n", - "14 | \n", - "30.29 | \n", - "31.57 | \n", - "
\n", + " | Date | \n", + "TikTok | \n", + "Google Ads | \n", + "Sales | \n", + "|
---|---|---|---|---|---|
count | \n", + "200 | \n", + "200.000000 | \n", + "200.000000 | \n", + "200.000000 | \n", + "200.000000 | \n", + "
mean | \n", + "2019-12-04 12:00:00 | \n", + "2946.207650 | \n", + "2213.585050 | \n", + "1520.722550 | \n", + "10668.141500 | \n", + "
min | \n", + "2018-01-07 00:00:00 | \n", + "0.000000 | \n", + "0.000000 | \n", + "0.000000 | \n", + "4532.330000 | \n", + "
25% | \n", + "2018-12-21 06:00:00 | \n", + "0.000000 | \n", + "0.000000 | \n", + "1657.195000 | \n", + "8396.942500 | \n", + "
50% | \n", + "2019-12-04 12:00:00 | \n", + "0.000000 | \n", + "0.000000 | \n", + "1918.990000 | \n", + "10853.105000 | \n", + "
75% | \n", + "2020-11-16 18:00:00 | \n", + "7938.527500 | \n", + "4624.027500 | \n", + "2069.767500 | \n", + "12566.995000 | \n", + "
max | \n", + "2021-10-31 00:00:00 | \n", + "13901.550000 | \n", + "7696.220000 | \n", + "2518.880000 | \n", + "17668.340000 | \n", + "
std | \n", + "NaN | \n", + "4749.646908 | \n", + "2505.967886 | \n", + "870.764354 | \n", + "2700.706683 | \n", + "
\n", + " | TikTok | \n", + "Google Ads | \n", + "|
---|---|---|---|
TikTok | \n", + "1.000000 | \n", + "0.084128 | \n", + "0.021735 | \n", + "
0.084128 | \n", + "1.000000 | \n", + "-0.100946 | \n", + "|
Google Ads | \n", + "0.021735 | \n", + "-0.100946 | \n", + "1.000000 | \n", + "
<xarray.DataArray ()>\n", - "array(23.5)
<xarray.DataArray (chain: 1, draw: 1000, customer_id: 500)>\n", - "array([[[1.77178601e-04, 6.39336399e-05, 9.89599883e-05, ...,\n", - " 1.01606742e-05, 3.87580543e-07, 1.62659379e-05],\n", - " [1.57042685e-04, 5.66366587e-05, 8.97643466e-05, ...,\n", - " 8.76911701e-06, 3.48630257e-07, 1.54153940e-05],\n", - " [1.62031718e-04, 6.08710433e-05, 8.96099220e-05, ...,\n", - " 9.72377742e-06, 3.64083561e-07, 1.68103170e-05],\n", - " ...,\n", - " [4.35264438e-05, 2.07515429e-05, 2.78013366e-05, ...,\n", - " 3.00041766e-06, 1.35433826e-07, 1.23333238e-05],\n", - " [2.24439083e-04, 7.48667922e-05, 1.21583909e-04, ...,\n", - " 1.16158223e-05, 4.39413252e-07, 1.68733060e-05],\n", - " [2.15888835e-04, 7.73232871e-05, 1.15060188e-04, ...,\n", - " 1.27732285e-05, 4.52420617e-07, 1.74784424e-05]]])\n", - "Coordinates:\n", - " * customer_id (customer_id) int64 0 1 2 3 4 5 6 ... 494 495 496 497 498 499\n", - " * chain (chain) int64 0\n", - " * draw (draw) int64 0 1 2 3 4 5 6 7 ... 993 994 995 996 997 998 999
\n", - " | id | \n", - "date | \n", - "first | \n", - "
---|---|---|---|
0 | \n", - "1 | \n", - "2015-01-01 | \n", - "True | \n", - "
1 | \n", - "1 | \n", - "2015-02-06 | \n", - "False | \n", - "
2 | \n", - "2 | \n", - "2015-01-01 | \n", - "True | \n", - "
3 | \n", - "3 | \n", - "2015-01-01 | \n", - "True | \n", - "
4 | \n", - "3 | \n", - "2015-01-02 | \n", - "False | \n", - "
5 | \n", - "3 | \n", - "2015-01-05 | \n", - "False | \n", - "
6 | \n", - "4 | \n", - "2015-01-16 | \n", - "True | \n", - "
7 | \n", - "4 | \n", - "2015-02-02 | \n", - "False | \n", - "
8 | \n", - "4 | \n", - "2015-02-05 | \n", - "False | \n", - "
10 | \n", - "5 | \n", - "2015-01-16 | \n", - "True | \n", - "
11 | \n", - "5 | \n", - "2015-01-17 | \n", - "False | \n", - "
12 | \n", - "5 | \n", - "2015-01-18 | \n", - "False | \n", - "
13 | \n", - "6 | \n", - "2015-02-02 | \n", - "True | \n", - "
\n", - " | customer_id | \n", - "frequency | \n", - "recency | \n", - "monetary_value | \n", - "
---|---|---|---|---|
0 | \n", - "1 | \n", - "2.0 | \n", - "0.0 | \n", - "1.5 | \n", - "
1 | \n", - "2 | \n", - "1.0 | \n", - "5.0 | \n", - "2.0 | \n", - "
2 | \n", - "3 | \n", - "2.0 | \n", - "4.0 | \n", - "4.5 | \n", - "
3 | \n", - "4 | \n", - "2.0 | \n", - "0.0 | \n", - "7.0 | \n", - "
4 | \n", - "5 | \n", - "1.0 | \n", - "3.0 | \n", - "12.0 | \n", - "
\n", - " | customer_id | \n", - "frequency | \n", - "recency | \n", - "T | \n", - "monetary_value | \n", - "test_frequency | \n", - "test_monetary_value | \n", - "test_T | \n", - "
---|---|---|---|---|---|---|---|---|
0 | \n", - "1 | \n", - "0.0 | \n", - "0.0 | \n", - "31.0 | \n", - "0.0 | \n", - "1.0 | \n", - "2.0 | \n", - "5.0 | \n", - "
1 | \n", - "2 | \n", - "0.0 | \n", - "0.0 | \n", - "31.0 | \n", - "0.0 | \n", - "0.0 | \n", - "0.0 | \n", - "5.0 | \n", - "
2 | \n", - "3 | \n", - "2.0 | \n", - "4.0 | \n", - "31.0 | \n", - "3.0 | \n", - "0.0 | \n", - "0.0 | \n", - "5.0 | \n", - "
3 | \n", - "4 | \n", - "0.0 | \n", - "0.0 | \n", - "16.0 | \n", - "0.0 | \n", - "2.0 | \n", - "4.0 | \n", - "5.0 | \n", - "
4 | \n", - "5 | \n", - "2.0 | \n", - "2.0 | \n", - "16.0 | \n", - "4.5 | \n", - "0.0 | \n", - "0.0 | \n", - "5.0 | \n", - "
\n", - "\n" - ], - "text/plain": [ - "\n" - ] - }, - "metadata": {}, - "output_type": "display_data" - }, - { - "data": { - "text/html": [ - "\n" - ], - "text/plain": [] - }, - "metadata": {}, - "output_type": "display_data" - }, - { - "data": { - "text/html": [ - "
\n", - "\n" - ], - "text/plain": [ - "\n" - ] - }, - "metadata": {}, - "output_type": "display_data" - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "
\n", - " | actual | \n", - "predicted | \n", - "
---|---|---|
1996-12-30/1997-01-05 | \n", - "0 | \n", - "4.232350 | \n", - "
1997-01-06/1997-01-12 | \n", - "3 | \n", - "15.415502 | \n", - "
1997-01-13/1997-01-19 | \n", - "17 | \n", - "33.593211 | \n", - "
1997-01-20/1997-01-26 | \n", - "44 | \n", - "59.287376 | \n", - "
1997-01-27/1997-02-02 | \n", - "67 | \n", - "92.716297 | \n", - "
... | \n", - "... | \n", - "... | \n", - "
2000-04-03/2000-04-09 | \n", - "4004 | \n", - "6563.077204 | \n", - "
2000-04-10/2000-04-16 | \n", - "4004 | \n", - "6586.673815 | \n", - "
2000-04-17/2000-04-23 | \n", - "4004 | \n", - "6610.195558 | \n", - "
2000-04-24/2000-04-30 | \n", - "4004 | \n", - "6633.643085 | \n", - "
2000-05-01/2000-05-07 | \n", - "4004 | \n", - "6657.017036 | \n", - "
175 rows × 2 columns
\n", - "\n", + " | id | \n", + "date | \n", + "first | \n", + "
---|---|---|---|
0 | \n", + "1 | \n", + "2015-01-01 | \n", + "True | \n", + "
1 | \n", + "1 | \n", + "2015-02-06 | \n", + "False | \n", + "
2 | \n", + "2 | \n", + "2015-01-01 | \n", + "True | \n", + "
3 | \n", + "3 | \n", + "2015-01-01 | \n", + "True | \n", + "
4 | \n", + "3 | \n", + "2015-01-02 | \n", + "False | \n", + "
5 | \n", + "3 | \n", + "2015-01-05 | \n", + "False | \n", + "
6 | \n", + "4 | \n", + "2015-01-16 | \n", + "True | \n", + "
7 | \n", + "4 | \n", + "2015-02-02 | \n", + "False | \n", + "
8 | \n", + "4 | \n", + "2015-02-05 | \n", + "False | \n", + "
10 | \n", + "5 | \n", + "2015-01-16 | \n", + "True | \n", + "
11 | \n", + "5 | \n", + "2015-01-17 | \n", + "False | \n", + "
12 | \n", + "5 | \n", + "2015-01-18 | \n", + "False | \n", + "
13 | \n", + "6 | \n", + "2015-02-02 | \n", + "True | \n", + "
\n", + " | customer_id | \n", + "frequency | \n", + "recency | \n", + "monetary_value | \n", + "
---|---|---|---|---|
0 | \n", + "1 | \n", + "2.0 | \n", + "0.0 | \n", + "1.5 | \n", + "
1 | \n", + "2 | \n", + "1.0 | \n", + "5.0 | \n", + "2.0 | \n", + "
2 | \n", + "3 | \n", + "2.0 | \n", + "4.0 | \n", + "4.5 | \n", + "
3 | \n", + "4 | \n", + "2.0 | \n", + "0.0 | \n", + "7.0 | \n", + "
4 | \n", + "5 | \n", + "1.0 | \n", + "3.0 | \n", + "12.0 | \n", + "
\n", + " | customer_id | \n", + "frequency | \n", + "recency | \n", + "T | \n", + "monetary_value | \n", + "test_frequency | \n", + "test_monetary_value | \n", + "test_T | \n", + "
---|---|---|---|---|---|---|---|---|
0 | \n", + "1 | \n", + "0.0 | \n", + "0.0 | \n", + "31.0 | \n", + "0.0 | \n", + "1.0 | \n", + "2.0 | \n", + "5.0 | \n", + "
1 | \n", + "2 | \n", + "0.0 | \n", + "0.0 | \n", + "31.0 | \n", + "0.0 | \n", + "0.0 | \n", + "0.0 | \n", + "5.0 | \n", + "
2 | \n", + "3 | \n", + "2.0 | \n", + "4.0 | \n", + "31.0 | \n", + "3.0 | \n", + "0.0 | \n", + "0.0 | \n", + "5.0 | \n", + "
3 | \n", + "4 | \n", + "0.0 | \n", + "0.0 | \n", + "16.0 | \n", + "0.0 | \n", + "2.0 | \n", + "4.0 | \n", + "5.0 | \n", + "
4 | \n", + "5 | \n", + "2.0 | \n", + "2.0 | \n", + "16.0 | \n", + "4.5 | \n", + "0.0 | \n", + "0.0 | \n", + "5.0 | \n", + "
\n", + "\n" + ], + "text/plain": [ + "\n" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "\n" + ], + "text/plain": [] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
\n", + "\n" + ], + "text/plain": [ + "\n" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "\n", + "
<xarray.Dataset> Size: 48B\n", + "Dimensions: (chain: 1, draw: 1)\n", + "Coordinates:\n", + " * chain (chain) int64 8B 0\n", + " * draw (draw) int64 8B 0\n", + "Data variables:\n", + " alpha (chain, draw) float64 8B 14.46\n", + " beta (chain, draw) float64 8B 10.48\n", + " r (chain, draw) float64 8B 0.6338\n", + " s (chain, draw) float64 8B 0.4882\n", + "Attributes:\n", + " created_at: 2024-11-23T22:38:58.227947+00:00\n", + " arviz_version: 0.18.0\n", + " inference_library: pymc\n", + " inference_library_version: 5.15.1
<xarray.Dataset> Size: 57kB\n", + "Dimensions: (customer_id: 2357, obs_var: 2)\n", + "Coordinates:\n", + " * customer_id (customer_id) int64 19kB 1 2 3 4 ... 2354 2355 2356 2357\n", + " * obs_var (obs_var) <U9 72B 'recency' 'frequency'\n", + "Data variables:\n", + " recency_frequency (customer_id, obs_var) float64 38kB 30.0 2.0 ... 0.0 0.0\n", + "Attributes:\n", + " created_at: 2024-11-23T22:38:58.230403+00:00\n", + " arviz_version: 0.18.0\n", + " inference_library: pymc\n", + " inference_library_version: 5.15.1
<xarray.Dataset> Size: 94kB\n", + "Dimensions: (index: 2357)\n", + "Coordinates:\n", + " * index (index) int64 19kB 0 1 2 3 4 5 ... 2352 2353 2354 2355 2356\n", + "Data variables:\n", + " customer_id (index) int64 19kB 1 2 3 4 5 6 ... 2353 2354 2355 2356 2357\n", + " frequency (index) float64 19kB 2.0 1.0 0.0 0.0 0.0 ... 5.0 0.0 4.0 0.0\n", + " recency (index) float64 19kB 30.0 2.0 0.0 0.0 0.0 ... 24.0 0.0 26.0 0.0\n", + " T (index) float64 19kB 39.0 39.0 39.0 39.0 ... 27.0 27.0 27.0
\n", + " | actual | \n", + "predicted | \n", + "
---|---|---|
1996-12-30/1997-01-05 | \n", + "0 | \n", + "4.286469 | \n", + "
1997-01-06/1997-01-12 | \n", + "3 | \n", + "15.595394 | \n", + "
1997-01-13/1997-01-19 | \n", + "17 | \n", + "33.947389 | \n", + "
1997-01-20/1997-01-26 | \n", + "44 | \n", + "59.853467 | \n", + "
1997-01-27/1997-02-02 | \n", + "67 | \n", + "93.519782 | \n", + "
\n", + "\n" + ], + "text/plain": [ + "\n" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
\n", + "\n" + ], + "text/plain": [ + "\n" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "Sampling 4 chains for 2_500 tune and 3_000 draw iterations (10_000 + 12_000 draws total) took 6 seconds.\n", + "/Users/coltallen/miniconda3/envs/pymc-marketing-dev/lib/python3.10/site-packages/pytensor/tensor/rewriting/elemwise.py:702: UserWarning: Optimization Warning: The Op hyp2f1 does not provide a C implementation. As well as being potentially slow, this also disables loop fusion.\n", + " warn(\n", + "The rhat statistic is larger than 1.01 for some parameters. This indicates problems during sampling. See https://arxiv.org/abs/1903.08008 for details\n" + ] + } + ], + "source": [ + "pnbd = ParetoNBDModel(data=rfm_data)\n", + "\n", + "pnbd.build_model()\n", + "with pnbd.model:\n", + " pnbd.idata = pm.sample(\n", + " step=pm.DEMetropolisZ(),\n", + " tune=2500,\n", + " draws=3000,\n", + " idata_kwargs={\"log_likelihood\": True},\n", + " )" + ] + }, + { + "cell_type": "code", + "execution_count": 14, + "id": "309b4678-ecce-45af-8bf8-69e3d419bb8c", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + " | mean | \n", + "sd | \n", + "hdi_3% | \n", + "hdi_97% | \n", + "mcse_mean | \n", + "mcse_sd | \n", + "ess_bulk | \n", + "ess_tail | \n", + "r_hat | \n", + "
---|---|---|---|---|---|---|---|---|---|
alpha | \n", + "14.596 | \n", + "1.114 | \n", + "12.445 | \n", + "16.640 | \n", + "0.042 | \n", + "0.030 | \n", + "712.0 | \n", + "1280.0 | \n", + "1.01 | \n", + "
beta | \n", + "12.043 | \n", + "3.800 | \n", + "5.029 | \n", + "18.742 | \n", + "0.127 | \n", + "0.090 | \n", + "862.0 | \n", + "1294.0 | \n", + "1.00 | \n", + "
r | \n", + "0.640 | \n", + "0.054 | \n", + "0.545 | \n", + "0.742 | \n", + "0.002 | \n", + "0.001 | \n", + "894.0 | \n", + "1132.0 | \n", + "1.01 | \n", + "
s | \n", + "0.529 | \n", + "0.105 | \n", + "0.340 | \n", + "0.729 | \n", + "0.004 | \n", + "0.003 | \n", + "842.0 | \n", + "1553.0 | \n", + "1.01 | \n", + "
<xarray.Dataset> Size: 136kB\n", + "Dimensions: (chain: 4, draw: 1000)\n", + "Coordinates:\n", + " * chain (chain) int64 32B 0 1 2 3\n", + " * draw (draw) int64 8kB 0 3 6 9 12 15 18 ... 2982 2985 2988 2991 2994 2997\n", + "Data variables:\n", + " alpha (chain, draw) float64 32kB 14.46 12.74 12.74 ... 13.8 13.8 13.8\n", + " beta (chain, draw) float64 32kB 10.87 14.3 14.3 ... 13.58 13.58 13.58\n", + " r (chain, draw) float64 32kB 0.6681 0.56 0.56 ... 0.5876 0.5876\n", + " s (chain, draw) float64 32kB 0.5164 0.5783 0.5783 ... 0.5571 0.5571\n", + "Attributes:\n", + " created_at: 2024-11-23T22:39:12.899029+00:00\n", + " arviz_version: 0.18.0\n", + " inference_library: pymc\n", + " inference_library_version: 5.15.1\n", + " sampling_time: 6.328435897827148\n", + " tuning_steps: 2500
<xarray.Dataset> Size: 75MB\n", + "Dimensions: (chain: 4, draw: 1000, customer_id: 2357)\n", + "Coordinates:\n", + " * chain (chain) int64 32B 0 1 2 3\n", + " * draw (draw) int64 8kB 0 3 6 9 12 ... 2985 2988 2991 2994 2997\n", + " * customer_id (customer_id) int64 19kB 1 2 3 4 ... 2354 2355 2356 2357\n", + "Data variables:\n", + " recency_frequency (chain, draw, customer_id) float64 75MB -9.378 ... -0....\n", + "Attributes:\n", + " created_at: 2024-11-23T22:39:18.592759+00:00\n", + " arviz_version: 0.18.0\n", + " inference_library: pymc\n", + " inference_library_version: 5.15.1
<xarray.Dataset> Size: 108kB\n", + "Dimensions: (chain: 4, draw: 1000)\n", + "Coordinates:\n", + " * chain (chain) int64 32B 0 1 2 3\n", + " * draw (draw) int64 8kB 0 3 6 9 12 15 ... 2982 2985 2988 2991 2994 2997\n", + "Data variables:\n", + " accept (chain, draw) float64 32kB 8.341 0.01704 ... 0.03161 0.627\n", + " accepted (chain, draw) bool 4kB True False False ... False False False\n", + " lambda (chain, draw) float64 32kB 0.8415 0.8415 0.8415 ... 0.8415 0.8415\n", + " scaling (chain, draw) float64 32kB 0.0002288 0.0002288 ... 0.001 0.001\n", + "Attributes:\n", + " created_at: 2024-11-23T22:39:12.901456+00:00\n", + " arviz_version: 0.18.0\n", + " inference_library: pymc\n", + " inference_library_version: 5.15.1\n", + " sampling_time: 6.328435897827148\n", + " tuning_steps: 2500
<xarray.Dataset> Size: 57kB\n", + "Dimensions: (customer_id: 2357, obs_var: 2)\n", + "Coordinates:\n", + " * customer_id (customer_id) int64 19kB 1 2 3 4 ... 2354 2355 2356 2357\n", + " * obs_var (obs_var) <U9 72B 'recency' 'frequency'\n", + "Data variables:\n", + " recency_frequency (customer_id, obs_var) float64 38kB 30.0 2.0 ... 0.0 0.0\n", + "Attributes:\n", + " created_at: 2024-11-23T22:39:12.902959+00:00\n", + " arviz_version: 0.18.0\n", + " inference_library: pymc\n", + " inference_library_version: 5.15.1
\n", + "\n" + ], + "text/plain": [ + "\n" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
\n", + "\n" + ], + "text/plain": [ + "\n" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
<xarray.DataArray 'recency_frequency' (chain: 4, draw: 1000, customer_id: 2357)> Size: 75MB\n", + "array([[[ 2., 0., 0., ..., 7., 1., 1.],\n", + " [ 0., 0., 0., ..., 0., 0., 0.],\n", + " [ 0., 2., 0., ..., 0., 4., 0.],\n", + " ...,\n", + " [ 8., 2., 10., ..., 7., 0., 1.],\n", + " [ 1., 0., 0., ..., 0., 0., 0.],\n", + " [ 0., 0., 1., ..., 2., 6., 0.]],\n", + "\n", + " [[ 1., 3., 0., ..., 1., 2., 0.],\n", + " [ 0., 0., 0., ..., 0., 1., 0.],\n", + " [ 0., 3., 0., ..., 0., 2., 4.],\n", + " ...,\n", + " [ 2., 0., 0., ..., 0., 2., 0.],\n", + " [ 0., 0., 4., ..., 0., 1., 3.],\n", + " [ 0., 0., 0., ..., 0., 0., 2.]],\n", + "\n", + " [[ 1., 0., 1., ..., 3., 0., 0.],\n", + " [ 1., 1., 0., ..., 3., 0., 0.],\n", + " [ 0., 1., 0., ..., 0., 2., 0.],\n", + " ...,\n", + " [ 0., 0., 0., ..., 0., 6., 0.],\n", + " [ 1., 0., 0., ..., 0., 1., 0.],\n", + " [ 0., 0., 0., ..., 0., 0., 1.]],\n", + "\n", + " [[ 1., 0., 0., ..., 0., 0., 1.],\n", + " [ 3., 0., 0., ..., 2., 0., 0.],\n", + " [ 4., 1., 0., ..., 2., 5., 0.],\n", + " ...,\n", + " [ 1., 0., 3., ..., 2., 3., 0.],\n", + " [ 0., 0., 1., ..., 0., 0., 0.],\n", + " [ 0., 0., 0., ..., 2., 0., 0.]]])\n", + "Coordinates:\n", + " * chain (chain) int64 32B 0 1 2 3\n", + " * draw (draw) int64 8kB 0 3 6 9 12 15 ... 2985 2988 2991 2994 2997\n", + " * customer_id (customer_id) int64 19kB 1 2 3 4 5 ... 2353 2354 2355 2356 2357\n", + " obs_var <U9 36B 'frequency'
<xarray.DataArray (x: 6)> Size: 48B\n", + "array([1, 2, 2, 3, 3, 3])\n", + "Dimensions without coordinates: x
<xarray.DataArray (customer_id: 2357)> Size: 19kB\n", + "array([6., 5., 9., ..., 6., 6., 6.])\n", + "Coordinates:\n", + " * customer_id (customer_id) int64 19kB 1 2 3 4 5 ... 2353 2354 2355 2356 2357\n", + " obs_var <U9 36B 'frequency'\n", + " hdi <U6 24B 'higher'\n", + " variable <U17 68B 'recency_frequency'
\n", + "\n" + ], + "text/plain": [ + "\n" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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", "text/plain": [ "
<xarray.Dataset> Size: 216kB\n", - "Dimensions: (chain: 2, draw: 3000)\n", + "<xarray.Dataset> Size: 408kB\n", + "Dimensions: (chain: 4, draw: 3000)\n", "Coordinates:\n", - " * chain (chain) int64 16B 0 1\n", + " * chain (chain) int64 32B 0 1 2 3\n", " * draw (draw) int64 24kB 0 1 2 3 4 5 6 ... 2994 2995 2996 2997 2998 2999\n", "Data variables:\n", - " alpha (chain, draw) float64 48kB 15.07 14.17 14.17 ... 16.07 16.07 16.07\n", - " beta (chain, draw) float64 48kB 10.22 13.48 13.48 ... 11.98 11.98 11.98\n", - " r (chain, draw) float64 48kB 0.6065 0.5683 0.5683 ... 0.6557 0.6557\n", - " s (chain, draw) float64 48kB 0.3881 0.4211 0.4211 ... 0.4103 0.4103\n", + " alpha (chain, draw) float64 96kB 16.12 16.12 16.12 ... 16.76 16.76 16.76\n", + " beta (chain, draw) float64 96kB 16.98 16.98 16.98 ... 8.639 8.639 8.639\n", + " r (chain, draw) float64 96kB 0.635 0.635 0.635 ... 0.6551 0.6551\n", + " s (chain, draw) float64 96kB 0.5287 0.5287 0.5287 ... 0.3744 0.3744\n", "Attributes:\n", - " created_at: 2024-07-01T15:47:46.613682\n", - " arviz_version: 0.17.1\n", + " created_at: 2024-11-23T22:32:50.642976+00:00\n", + " arviz_version: 0.18.0\n", " inference_library: pymc\n", - " inference_library_version: 5.14.0\n", - " sampling_time: 90.08395767211914\n", - " tuning_steps: 2500
<xarray.Dataset> Size: 113MB\n", - "Dimensions: (chain: 2, draw: 3000, customer_id: 2349)\n", + "<xarray.Dataset> Size: 226MB\n", + "Dimensions: (chain: 4, draw: 3000, customer_id: 2349)\n", "Coordinates:\n", - " * chain (chain) int64 16B 0 1\n", + " * chain (chain) int64 32B 0 1 2 3\n", " * draw (draw) int64 24kB 0 1 2 3 4 ... 2995 2996 2997 2998 2999\n", " * customer_id (customer_id) int64 19kB 1 2 3 4 ... 2354 2355 2356 2357\n", "Data variables:\n", - " recency_frequency (chain, draw, customer_id) float64 113MB -14.3 ... -0....\n", + " recency_frequency (chain, draw, customer_id) float64 226MB -14.28 ... -0...\n", "Attributes:\n", - " created_at: 2024-07-01T15:48:11.016568\n", - " arviz_version: 0.17.1\n", + " created_at: 2024-11-23T22:32:56.529120+00:00\n", + " arviz_version: 0.18.0\n", " inference_library: pymc\n", - " inference_library_version: 5.14.0
<xarray.Dataset> Size: 174kB\n", - "Dimensions: (chain: 2, draw: 3000)\n", + "<xarray.Dataset> Size: 324kB\n", + "Dimensions: (chain: 4, draw: 3000)\n", "Coordinates:\n", - " * chain (chain) int64 16B 0 1\n", + " * chain (chain) int64 32B 0 1 2 3\n", " * draw (draw) int64 24kB 0 1 2 3 4 5 6 ... 2994 2995 2996 2997 2998 2999\n", "Data variables:\n", - " scaling (chain, draw) float64 48kB 0.0002059 0.0002059 ... 0.0003874\n", - " lambda (chain, draw) float64 48kB 0.8415 0.8415 0.8415 ... 0.8415 0.8415\n", - " accept (chain, draw) float64 48kB 0.01836 0.5286 ... 0.05464 0.02259\n", - " accepted (chain, draw) bool 6kB False True False ... False False False\n", + " accept (chain, draw) float64 96kB 0.4747 0.0004368 ... 0.007007 0.4167\n", + " accepted (chain, draw) bool 12kB False False False ... False False False\n", + " lambda (chain, draw) float64 96kB 0.8415 0.8415 0.8415 ... 0.8415 0.8415\n", + " scaling (chain, draw) float64 96kB 0.0002542 0.0002542 ... 0.0002288\n", "Attributes:\n", - " created_at: 2024-07-01T15:47:46.741295\n", - " arviz_version: 0.17.1\n", + " created_at: 2024-11-23T22:32:50.645501+00:00\n", + " arviz_version: 0.18.0\n", " inference_library: pymc\n", - " inference_library_version: 5.14.0\n", - " sampling_time: 90.08395767211914\n", - " tuning_steps: 2500
PandasIndex(Index(['recency', 'frequency'], dtype='object', name='obs_var'))