From 18fcb3c78907249352d8100cc825057d9b3e83d7 Mon Sep 17 00:00:00 2001
From: prabathbr <60564552+prabathbr@users.noreply.github.com>
Date: Wed, 19 May 2021 16:18:28 +1000
Subject: [PATCH] Updated to be compatible with latest FBProphet
Updated to be compatible with latest FBProphet version
---
COVID19_Trend.ipynb | 4058 +------------------------------------------
1 file changed, 42 insertions(+), 4016 deletions(-)
diff --git a/COVID19_Trend.ipynb b/COVID19_Trend.ipynb
index 610ff0b..cb711de 100644
--- a/COVID19_Trend.ipynb
+++ b/COVID19_Trend.ipynb
@@ -9,7 +9,7 @@
},
{
"cell_type": "code",
- "execution_count": 1,
+ "execution_count": null,
"metadata": {},
"outputs": [],
"source": [
@@ -22,7 +22,7 @@
},
{
"cell_type": "code",
- "execution_count": 2,
+ "execution_count": null,
"metadata": {},
"outputs": [],
"source": [
@@ -36,19 +36,19 @@
},
{
"cell_type": "code",
- "execution_count": 3,
+ "execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"# Inital Parameter 2 :\n",
"#\n",
"# Select Country\n",
- "show_country = 'Spain'"
+ "show_country = 'Oceania'"
]
},
{
"cell_type": "code",
- "execution_count": 4,
+ "execution_count": null,
"metadata": {},
"outputs": [],
"source": [
@@ -60,111 +60,28 @@
},
{
"cell_type": "code",
- "execution_count": 5,
+ "execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"# Inital Parameter 4 :\n",
"#\n",
"# Set number of forcasting days\n",
- "days_forcasting = 90"
+ "days_forcasting = 10"
]
},
{
"cell_type": "code",
- "execution_count": 6,
+ "execution_count": null,
"metadata": {},
- "outputs": [
- {
- "data": {
- "text/html": [
- " \n",
- " "
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- }
- ],
+ "outputs": [],
"source": [
"# Import Pandas\n",
"import pandas as pd\n",
"# Import Facebook Prophet\n",
- "from fbprophet import Prophet\n",
+ "from prophet import Prophet\n",
"# Import Plotly\n",
- "from fbprophet.plot import plot_plotly\n",
+ "from prophet.plot import plot_plotly\n",
"import plotly.offline as py\n",
"# Initialise Plotly\n",
"py.init_notebook_mode()"
@@ -172,7 +89,7 @@
},
{
"cell_type": "code",
- "execution_count": 7,
+ "execution_count": null,
"metadata": {},
"outputs": [],
"source": [
@@ -182,17 +99,9 @@
},
{
"cell_type": "code",
- "execution_count": 8,
+ "execution_count": null,
"metadata": {},
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "{'Afghanistan', 'Central African Republic', 'Panama', 'Greece', 'Saint Kitts and Nevis', 'Anguilla', 'Thailand', 'United Kingdom', 'Cape Verde', 'Taiwan', 'Turkey', \"Cote d'Ivoire\", 'Sweden', 'Mauritius', 'Namibia', 'San Marino', 'United States Virgin Islands', 'Greenland', 'New Zealand', 'Dominican Republic', 'Colombia', 'Liechtenstein', 'Niger', 'Finland', 'South Korea', 'Lithuania', 'Angola', 'Madagascar', 'Turks and Caicos Islands', 'Iran', 'Bhutan', 'Belize', 'Faeroe Islands', 'Guam', 'Gibraltar', 'Philippines', 'Grenada', 'Kosovo', 'Liberia', 'Montserrat', 'Saudi Arabia', 'Guinea', 'Ukraine', 'China', 'Morocco', 'Algeria', 'Suriname', 'Libya', 'Poland', 'Chile', 'Cameroon', 'Seychelles', 'Costa Rica', 'Uganda', 'Pakistan', 'Tunisia', 'United States', 'Ethiopia', 'Curacao', 'Swaziland', 'Timor', 'Jordan', 'Palestine', 'Botswana', 'Saint Vincent and the Grenadines', 'Dominica', 'Slovakia', 'Lebanon', 'Gambia', 'Armenia', 'Croatia', 'Honduras', 'Argentina', 'Peru', 'New Caledonia', 'Gabon', 'Paraguay', 'Indonesia', 'Ghana', 'Falkland Islands', 'Nigeria', 'Burkina Faso', 'Mexico', 'Guernsey', 'Bonaire Sint Eustatius and Saba', 'Georgia', 'Mali', 'India', 'Brazil', 'Zimbabwe', 'Cyprus', 'Jersey', 'Moldova', 'Sint Maarten (Dutch part)', 'Tajikistan', 'Uruguay', 'Senegal', 'Brunei', 'Chad', 'Bahrain', 'Vatican', 'Guatemala', 'Bahamas', 'Bangladesh', 'Cayman Islands', 'Romania', 'Serbia', 'Uzbekistan', 'Slovenia', 'Israel', 'Venezuela', 'Northern Mariana Islands', 'Bosnia and Herzegovina', 'Italy', 'France', 'Vietnam', 'Australia', 'Tanzania', 'Japan', 'Canada', 'Aruba', 'South Africa', 'Syria', 'Zambia', 'Bolivia', 'Monaco', 'Guyana', 'Iraq', 'Qatar', 'Oman', 'Democratic Republic of Congo', 'Belarus', 'Hong Kong', 'Germany', 'Maldives', 'Malaysia', 'Spain', 'Guinea-Bissau', 'Papua New Guinea', 'Cuba', 'World', 'International', 'Mozambique', 'Burundi', 'Fiji', 'Belgium', 'Latvia', 'Russia', 'Rwanda', 'Macedonia', 'Azerbaijan', 'Austria', 'Ecuador', 'Ireland', 'Malta', 'Albania', 'Laos', 'Estonia', 'Kuwait', 'French Polynesia', 'Djibouti', 'Cambodia', 'Egypt', 'Nepal', 'Somalia', 'Trinidad and Tobago', 'British Virgin Islands', 'Malawi', 'Bermuda', 'Sierra Leone', 'Saint Lucia', 'Hungary', 'Sudan', 'Sao Tome and Principe', 'Kazakhstan', 'Switzerland', 'Singapore', 'Antigua and Barbuda', 'El Salvador', 'United Arab Emirates', 'Western Sahara', 'Benin', 'Myanmar', 'Kyrgyzstan', 'Haiti', 'Norway', 'Comoros', 'Nicaragua', 'Mongolia', 'Portugal', 'Mauritania', 'Congo', 'Puerto Rico', 'Andorra', 'Luxembourg', 'Czech Republic', 'Togo', 'Isle of Man', 'Barbados', 'Yemen', 'Netherlands', 'Lesotho', 'Denmark', 'Kenya', 'Montenegro', 'South Sudan', 'Jamaica', 'Iceland', 'Sri Lanka', 'Equatorial Guinea', 'Eritrea', 'Bulgaria'}\n"
- ]
- }
- ],
+ "outputs": [],
"source": [
"# Show a list of available countries which can be set at \"Inital Parameter 2\"\n",
"country_list = set(df_read.location)\n",
@@ -201,223 +110,9 @@
},
{
"cell_type": "code",
- "execution_count": 9,
+ "execution_count": null,
"metadata": {},
- "outputs": [
- {
- "data": {
- "text/html": [
- "
\n",
- "\n",
- "
\n",
- " \n",
- " \n",
- " | \n",
- " iso_code | \n",
- " continent | \n",
- " location | \n",
- " date | \n",
- " total_cases | \n",
- " new_cases | \n",
- " total_deaths | \n",
- " new_deaths | \n",
- " total_cases_per_million | \n",
- " new_cases_per_million | \n",
- " ... | \n",
- " aged_70_older | \n",
- " gdp_per_capita | \n",
- " extreme_poverty | \n",
- " cardiovasc_death_rate | \n",
- " diabetes_prevalence | \n",
- " female_smokers | \n",
- " male_smokers | \n",
- " handwashing_facilities | \n",
- " hospital_beds_per_thousand | \n",
- " life_expectancy | \n",
- "
\n",
- " \n",
- " \n",
- " \n",
- " 0 | \n",
- " AFG | \n",
- " Asia | \n",
- " Afghanistan | \n",
- " 2019-12-31 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " ... | \n",
- " 1.337 | \n",
- " 1803.987 | \n",
- " NaN | \n",
- " 597.029 | \n",
- " 9.59 | \n",
- " NaN | \n",
- " NaN | \n",
- " 37.746 | \n",
- " 0.5 | \n",
- " 64.83 | \n",
- "
\n",
- " \n",
- " 1 | \n",
- " AFG | \n",
- " Asia | \n",
- " Afghanistan | \n",
- " 2020-01-01 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " ... | \n",
- " 1.337 | \n",
- " 1803.987 | \n",
- " NaN | \n",
- " 597.029 | \n",
- " 9.59 | \n",
- " NaN | \n",
- " NaN | \n",
- " 37.746 | \n",
- " 0.5 | \n",
- " 64.83 | \n",
- "
\n",
- " \n",
- " 2 | \n",
- " AFG | \n",
- " Asia | \n",
- " Afghanistan | \n",
- " 2020-01-02 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " ... | \n",
- " 1.337 | \n",
- " 1803.987 | \n",
- " NaN | \n",
- " 597.029 | \n",
- " 9.59 | \n",
- " NaN | \n",
- " NaN | \n",
- " 37.746 | \n",
- " 0.5 | \n",
- " 64.83 | \n",
- "
\n",
- " \n",
- " 3 | \n",
- " AFG | \n",
- " Asia | \n",
- " Afghanistan | \n",
- " 2020-01-03 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " ... | \n",
- " 1.337 | \n",
- " 1803.987 | \n",
- " NaN | \n",
- " 597.029 | \n",
- " 9.59 | \n",
- " NaN | \n",
- " NaN | \n",
- " 37.746 | \n",
- " 0.5 | \n",
- " 64.83 | \n",
- "
\n",
- " \n",
- " 4 | \n",
- " AFG | \n",
- " Asia | \n",
- " Afghanistan | \n",
- " 2020-01-04 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " ... | \n",
- " 1.337 | \n",
- " 1803.987 | \n",
- " NaN | \n",
- " 597.029 | \n",
- " 9.59 | \n",
- " NaN | \n",
- " NaN | \n",
- " 37.746 | \n",
- " 0.5 | \n",
- " 64.83 | \n",
- "
\n",
- " \n",
- "
\n",
- "
5 rows × 34 columns
\n",
- "
"
- ],
- "text/plain": [
- " iso_code continent location date total_cases new_cases \\\n",
- "0 AFG Asia Afghanistan 2019-12-31 0.0 0.0 \n",
- "1 AFG Asia Afghanistan 2020-01-01 0.0 0.0 \n",
- "2 AFG Asia Afghanistan 2020-01-02 0.0 0.0 \n",
- "3 AFG Asia Afghanistan 2020-01-03 0.0 0.0 \n",
- "4 AFG Asia Afghanistan 2020-01-04 0.0 0.0 \n",
- "\n",
- " total_deaths new_deaths total_cases_per_million new_cases_per_million \\\n",
- "0 0.0 0.0 0.0 0.0 \n",
- "1 0.0 0.0 0.0 0.0 \n",
- "2 0.0 0.0 0.0 0.0 \n",
- "3 0.0 0.0 0.0 0.0 \n",
- "4 0.0 0.0 0.0 0.0 \n",
- "\n",
- " ... aged_70_older gdp_per_capita extreme_poverty cardiovasc_death_rate \\\n",
- "0 ... 1.337 1803.987 NaN 597.029 \n",
- "1 ... 1.337 1803.987 NaN 597.029 \n",
- "2 ... 1.337 1803.987 NaN 597.029 \n",
- "3 ... 1.337 1803.987 NaN 597.029 \n",
- "4 ... 1.337 1803.987 NaN 597.029 \n",
- "\n",
- " diabetes_prevalence female_smokers male_smokers handwashing_facilities \\\n",
- "0 9.59 NaN NaN 37.746 \n",
- "1 9.59 NaN NaN 37.746 \n",
- "2 9.59 NaN NaN 37.746 \n",
- "3 9.59 NaN NaN 37.746 \n",
- "4 9.59 NaN NaN 37.746 \n",
- "\n",
- " hospital_beds_per_thousand life_expectancy \n",
- "0 0.5 64.83 \n",
- "1 0.5 64.83 \n",
- "2 0.5 64.83 \n",
- "3 0.5 64.83 \n",
- "4 0.5 64.83 \n",
- "\n",
- "[5 rows x 34 columns]"
- ]
- },
- "execution_count": 9,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
+ "outputs": [],
"source": [
"# Show all available data fields\n",
"df_read.head()"
@@ -425,7 +120,7 @@
},
{
"cell_type": "code",
- "execution_count": 10,
+ "execution_count": null,
"metadata": {},
"outputs": [],
"source": [
@@ -435,230 +130,9 @@
},
{
"cell_type": "code",
- "execution_count": 11,
+ "execution_count": null,
"metadata": {},
- "outputs": [
- {
- "data": {
- "text/html": [
- "\n",
- "\n",
- "
\n",
- " \n",
- " \n",
- " | \n",
- " iso_code | \n",
- " continent | \n",
- " location | \n",
- " date | \n",
- " total_cases | \n",
- " new_cases | \n",
- " total_deaths | \n",
- " new_deaths | \n",
- " total_cases_per_million | \n",
- " new_cases_per_million | \n",
- " ... | \n",
- " aged_70_older | \n",
- " gdp_per_capita | \n",
- " extreme_poverty | \n",
- " cardiovasc_death_rate | \n",
- " diabetes_prevalence | \n",
- " female_smokers | \n",
- " male_smokers | \n",
- " handwashing_facilities | \n",
- " hospital_beds_per_thousand | \n",
- " life_expectancy | \n",
- "
\n",
- " \n",
- " \n",
- " \n",
- " 28034 | \n",
- " ESP | \n",
- " Europe | \n",
- " Spain | \n",
- " 2019-12-31 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " ... | \n",
- " 13.799 | \n",
- " 34272.36 | \n",
- " 1.0 | \n",
- " 99.403 | \n",
- " 7.17 | \n",
- " 27.4 | \n",
- " 31.4 | \n",
- " NaN | \n",
- " 2.97 | \n",
- " 83.56 | \n",
- "
\n",
- " \n",
- " 28035 | \n",
- " ESP | \n",
- " Europe | \n",
- " Spain | \n",
- " 2020-01-01 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " ... | \n",
- " 13.799 | \n",
- " 34272.36 | \n",
- " 1.0 | \n",
- " 99.403 | \n",
- " 7.17 | \n",
- " 27.4 | \n",
- " 31.4 | \n",
- " NaN | \n",
- " 2.97 | \n",
- " 83.56 | \n",
- "
\n",
- " \n",
- " 28036 | \n",
- " ESP | \n",
- " Europe | \n",
- " Spain | \n",
- " 2020-01-02 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " ... | \n",
- " 13.799 | \n",
- " 34272.36 | \n",
- " 1.0 | \n",
- " 99.403 | \n",
- " 7.17 | \n",
- " 27.4 | \n",
- " 31.4 | \n",
- " NaN | \n",
- " 2.97 | \n",
- " 83.56 | \n",
- "
\n",
- " \n",
- " 28037 | \n",
- " ESP | \n",
- " Europe | \n",
- " Spain | \n",
- " 2020-01-03 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " ... | \n",
- " 13.799 | \n",
- " 34272.36 | \n",
- " 1.0 | \n",
- " 99.403 | \n",
- " 7.17 | \n",
- " 27.4 | \n",
- " 31.4 | \n",
- " NaN | \n",
- " 2.97 | \n",
- " 83.56 | \n",
- "
\n",
- " \n",
- " 28038 | \n",
- " ESP | \n",
- " Europe | \n",
- " Spain | \n",
- " 2020-01-04 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " 0.0 | \n",
- " ... | \n",
- " 13.799 | \n",
- " 34272.36 | \n",
- " 1.0 | \n",
- " 99.403 | \n",
- " 7.17 | \n",
- " 27.4 | \n",
- " 31.4 | \n",
- " NaN | \n",
- " 2.97 | \n",
- " 83.56 | \n",
- "
\n",
- " \n",
- "
\n",
- "
5 rows × 34 columns
\n",
- "
"
- ],
- "text/plain": [
- " iso_code continent location date total_cases new_cases \\\n",
- "28034 ESP Europe Spain 2019-12-31 0.0 0.0 \n",
- "28035 ESP Europe Spain 2020-01-01 0.0 0.0 \n",
- "28036 ESP Europe Spain 2020-01-02 0.0 0.0 \n",
- "28037 ESP Europe Spain 2020-01-03 0.0 0.0 \n",
- "28038 ESP Europe Spain 2020-01-04 0.0 0.0 \n",
- "\n",
- " total_deaths new_deaths total_cases_per_million \\\n",
- "28034 0.0 0.0 0.0 \n",
- "28035 0.0 0.0 0.0 \n",
- "28036 0.0 0.0 0.0 \n",
- "28037 0.0 0.0 0.0 \n",
- "28038 0.0 0.0 0.0 \n",
- "\n",
- " new_cases_per_million ... aged_70_older gdp_per_capita \\\n",
- "28034 0.0 ... 13.799 34272.36 \n",
- "28035 0.0 ... 13.799 34272.36 \n",
- "28036 0.0 ... 13.799 34272.36 \n",
- "28037 0.0 ... 13.799 34272.36 \n",
- "28038 0.0 ... 13.799 34272.36 \n",
- "\n",
- " extreme_poverty cardiovasc_death_rate diabetes_prevalence \\\n",
- "28034 1.0 99.403 7.17 \n",
- "28035 1.0 99.403 7.17 \n",
- "28036 1.0 99.403 7.17 \n",
- "28037 1.0 99.403 7.17 \n",
- "28038 1.0 99.403 7.17 \n",
- "\n",
- " female_smokers male_smokers handwashing_facilities \\\n",
- "28034 27.4 31.4 NaN \n",
- "28035 27.4 31.4 NaN \n",
- "28036 27.4 31.4 NaN \n",
- "28037 27.4 31.4 NaN \n",
- "28038 27.4 31.4 NaN \n",
- "\n",
- " hospital_beds_per_thousand life_expectancy \n",
- "28034 2.97 83.56 \n",
- "28035 2.97 83.56 \n",
- "28036 2.97 83.56 \n",
- "28037 2.97 83.56 \n",
- "28038 2.97 83.56 \n",
- "\n",
- "[5 rows x 34 columns]"
- ]
- },
- "execution_count": 11,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
+ "outputs": [],
"source": [
"# Show a sample of selected data\n",
"df_filter.head()"
@@ -666,7 +140,7 @@
},
{
"cell_type": "code",
- "execution_count": 12,
+ "execution_count": null,
"metadata": {},
"outputs": [],
"source": [
@@ -676,7 +150,7 @@
},
{
"cell_type": "code",
- "execution_count": 13,
+ "execution_count": null,
"metadata": {},
"outputs": [],
"source": [
@@ -686,78 +160,9 @@
},
{
"cell_type": "code",
- "execution_count": 14,
+ "execution_count": null,
"metadata": {},
- "outputs": [
- {
- "data": {
- "text/html": [
- "\n",
- "\n",
- "
\n",
- " \n",
- " \n",
- " | \n",
- " ds | \n",
- " y | \n",
- "
\n",
- " \n",
- " \n",
- " \n",
- " 28034 | \n",
- " 2019-12-31 | \n",
- " 0.0 | \n",
- "
\n",
- " \n",
- " 28035 | \n",
- " 2020-01-01 | \n",
- " 0.0 | \n",
- "
\n",
- " \n",
- " 28036 | \n",
- " 2020-01-02 | \n",
- " 0.0 | \n",
- "
\n",
- " \n",
- " 28037 | \n",
- " 2020-01-03 | \n",
- " 0.0 | \n",
- "
\n",
- " \n",
- " 28038 | \n",
- " 2020-01-04 | \n",
- " 0.0 | \n",
- "
\n",
- " \n",
- "
\n",
- "
"
- ],
- "text/plain": [
- " ds y\n",
- "28034 2019-12-31 0.0\n",
- "28035 2020-01-01 0.0\n",
- "28036 2020-01-02 0.0\n",
- "28037 2020-01-03 0.0\n",
- "28038 2020-01-04 0.0"
- ]
- },
- "execution_count": 14,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
+ "outputs": [],
"source": [
"# Show a sample of converted selected data (first 5)\n",
"df_extracted.head()"
@@ -765,78 +170,9 @@
},
{
"cell_type": "code",
- "execution_count": 15,
+ "execution_count": null,
"metadata": {},
- "outputs": [
- {
- "data": {
- "text/html": [
- "\n",
- "\n",
- "
\n",
- " \n",
- " \n",
- " | \n",
- " ds | \n",
- " y | \n",
- "
\n",
- " \n",
- " \n",
- " \n",
- " 28240 | \n",
- " 2020-07-24 | \n",
- " 2255.0 | \n",
- "
\n",
- " \n",
- " 28241 | \n",
- " 2020-07-25 | \n",
- " 0.0 | \n",
- "
\n",
- " \n",
- " 28242 | \n",
- " 2020-07-26 | \n",
- " 0.0 | \n",
- "
\n",
- " \n",
- " 28243 | \n",
- " 2020-07-27 | \n",
- " 6361.0 | \n",
- "
\n",
- " \n",
- " 28244 | \n",
- " 2020-07-28 | \n",
- " 1828.0 | \n",
- "
\n",
- " \n",
- "
\n",
- "
"
- ],
- "text/plain": [
- " ds y\n",
- "28240 2020-07-24 2255.0\n",
- "28241 2020-07-25 0.0\n",
- "28242 2020-07-26 0.0\n",
- "28243 2020-07-27 6361.0\n",
- "28244 2020-07-28 1828.0"
- ]
- },
- "execution_count": 15,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
+ "outputs": [],
"source": [
"# Show last 5 available dates\n",
"df_extracted.tail()"
@@ -844,17 +180,9 @@
},
{
"cell_type": "code",
- "execution_count": 16,
+ "execution_count": null,
"metadata": {},
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Total number of available dates: 211\n"
- ]
- }
- ],
+ "outputs": [],
"source": [
"# Show number of available dates\n",
"total_dates = df_extracted.shape[0]\n",
@@ -863,28 +191,9 @@
},
{
"cell_type": "code",
- "execution_count": 17,
+ "execution_count": null,
"metadata": {},
- "outputs": [
- {
- "name": "stderr",
- "output_type": "stream",
- "text": [
- "INFO:fbprophet:Disabling yearly seasonality. Run prophet with yearly_seasonality=True to override this.\n",
- "INFO:fbprophet:Disabling daily seasonality. Run prophet with daily_seasonality=True to override this.\n"
- ]
- },
- {
- "data": {
- "text/plain": [
- ""
- ]
- },
- "execution_count": 17,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
+ "outputs": [],
"source": [
"# Initialise Facebook Prophet\n",
"m = Prophet(changepoint_prior_scale=trend_flexibility) \n",
@@ -894,7 +203,7 @@
},
{
"cell_type": "code",
- "execution_count": 18,
+ "execution_count": null,
"metadata": {},
"outputs": [],
"source": [
@@ -904,72 +213,9 @@
},
{
"cell_type": "code",
- "execution_count": 19,
+ "execution_count": null,
"metadata": {},
- "outputs": [
- {
- "data": {
- "text/html": [
- "\n",
- "\n",
- "
\n",
- " \n",
- " \n",
- " | \n",
- " ds | \n",
- "
\n",
- " \n",
- " \n",
- " \n",
- " 296 | \n",
- " 2020-10-22 | \n",
- "
\n",
- " \n",
- " 297 | \n",
- " 2020-10-23 | \n",
- "
\n",
- " \n",
- " 298 | \n",
- " 2020-10-24 | \n",
- "
\n",
- " \n",
- " 299 | \n",
- " 2020-10-25 | \n",
- "
\n",
- " \n",
- " 300 | \n",
- " 2020-10-26 | \n",
- "
\n",
- " \n",
- "
\n",
- "
"
- ],
- "text/plain": [
- " ds\n",
- "296 2020-10-22\n",
- "297 2020-10-23\n",
- "298 2020-10-24\n",
- "299 2020-10-25\n",
- "300 2020-10-26"
- ]
- },
- "execution_count": 19,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
+ "outputs": [],
"source": [
"# Show last 5 forcasting dates\n",
"future.tail()"
@@ -977,7 +223,7 @@
},
{
"cell_type": "code",
- "execution_count": 20,
+ "execution_count": null,
"metadata": {},
"outputs": [],
"source": [
@@ -988,20 +234,9 @@
},
{
"cell_type": "code",
- "execution_count": 21,
+ "execution_count": null,
"metadata": {},
- "outputs": [
- {
- "data": {
- "image/png": "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\n",
- "text/plain": [
- "