I have a hypothesis that holidays can affect to energy consumption. For example in holidays Educational buildings should use less energy, water, but another situation should be in cafes and similar entertainment buildings. So there is a list of US holidays for 2016-2018 years:
\n\n| Date | Holiday name | Day |\n| --- | --- |\n| January 1, 2016 | New Year's Day | Friday |\n| January 18, 2016 | Martin Luther King Day | Monday |\n| February 15, 2016 | Presidents' Day | Monday |\n| May 30, 2016 | Memorial Day | Monday |\n| July 4, 2016 | Independence Day | Monday |\n| September 5, 2016 | Labor Day | Monday |\n| October 10, 2016 | Columbus Day | Monday |\n| November 11, 2016 | Veterans Day | Friday |\n| November 24, 2016 | Thanksgiving Day | Thursday |\n| December 25, 2016 | Christmas Day | Sunday |\n| December 26, 2016 | Christmas Day (observed) | Monday |\n| January 1, 2017 | New Year's Day | Sunday |\n| January 2, 2017 | New Year's Day (observed) | Monday |\n| January 16, 2017 | Martin Luther King Day | Monday |\n| February 20, 2017 | Presidents' Day | Monday |\n| May 29, 2017 | Memorial Day | Monday |\n| July 4, 2017 | Independence Day | Tuesday |\n| September 4, 2017 | Labor Day | Monday \n| October 9, 2017 | Columbus Day | Monday |\n|November 10, 2017| Veterans Day (observed) | Friday |\n| November 11, 2017 | Veterans Day | Saturday |\n| November 23, 2017 | Thanksgiving Day | Thursday |\n| December 25, 2017 | Christmas Day | Monday |\n| January 1, 2018 | New Year's Day | Monday |\n| January 15, 2018 | Martin Luther King Day | Monday |\n| February 19, 2018 | Presidents' Day | Monday |\n| May 28, 2018 | Memorial Day | Monday |\n| July 4, 2018 | Independence Day | Wednesday |\n| September 3, 2018 | Labor Day | Monday |\n| October 8, 2018 | Columbus Day | Monday |\n| November 11, 2018 | Veterans Day | Sunday |\n| November 12, 2018 | Veterans Day (observed) | Monday |\n| November 22, 2018 | Thanksgiving Day | Thursday |\n| December 25, 2018 | Christmas Day | Tuesday |
\n\nIn my solution feature based on this data bring improvements on validation = 0.02.
", "messages": [ { "id": 652035, "postDate": "2019-10-18T08:48:24.253Z", "content": "I have a hypothesis that holidays can affect to energy consumption. For example in holidays Educational buildings should use less energy, water, but another situation should be in cafes and similar entertainment buildings. So there is a list of US holidays for 2016-2018 years:
\n\n| Date | Holiday name | Day |\n| --- | --- |\n| January 1, 2016 | New Year's Day | Friday |\n| January 18, 2016 | Martin Luther King Day | Monday |\n| February 15, 2016 | Presidents' Day | Monday |\n| May 30, 2016 | Memorial Day | Monday |\n| July 4, 2016 | Independence Day | Monday |\n| September 5, 2016 | Labor Day | Monday |\n| October 10, 2016 | Columbus Day | Monday |\n| November 11, 2016 | Veterans Day | Friday |\n| November 24, 2016 | Thanksgiving Day | Thursday |\n| December 25, 2016 | Christmas Day | Sunday |\n| December 26, 2016 | Christmas Day (observed) | Monday |\n| January 1, 2017 | New Year's Day | Sunday |\n| January 2, 2017 | New Year's Day (observed) | Monday |\n| January 16, 2017 | Martin Luther King Day | Monday |\n| February 20, 2017 | Presidents' Day | Monday |\n| May 29, 2017 | Memorial Day | Monday |\n| July 4, 2017 | Independence Day | Tuesday |\n| September 4, 2017 | Labor Day | Monday \n| October 9, 2017 | Columbus Day | Monday |\n|November 10, 2017| Veterans Day (observed) | Friday |\n| November 11, 2017 | Veterans Day | Saturday |\n| November 23, 2017 | Thanksgiving Day | Thursday |\n| December 25, 2017 | Christmas Day | Monday |\n| January 1, 2018 | New Year's Day | Monday |\n| January 15, 2018 | Martin Luther King Day | Monday |\n| February 19, 2018 | Presidents' Day | Monday |\n| May 28, 2018 | Memorial Day | Monday |\n| July 4, 2018 | Independence Day | Wednesday |\n| September 3, 2018 | Labor Day | Monday |\n| October 8, 2018 | Columbus Day | Monday |\n| November 11, 2018 | Veterans Day | Sunday |\n| November 12, 2018 | Veterans Day (observed) | Monday |\n| November 22, 2018 | Thanksgiving Day | Thursday |\n| December 25, 2018 | Christmas Day | Tuesday |
\n\nIn my solution feature based on this data bring improvements on validation = 0.02.
", "rawMarkdown": "I have a hypothesis that holidays can affect to energy consumption. For example in holidays Educational buildings should use less energy, water, but another situation should be in cafes and similar entertainment buildings. So there is a list of US holidays for 2016-2018 years:\n\n| Date | Holiday name | Day |\n| --- | --- |\n| January 1, 2016 | New Year's Day | Friday |\n| January 18, 2016 | Martin Luther King Day | Monday |\n| February 15, 2016 | Presidents' Day | Monday |\n| May 30, 2016 | Memorial Day | Monday |\n| July 4, 2016 | Independence Day | Monday |\n| September 5, 2016 | Labor Day | Monday |\n| October 10, 2016 | Columbus Day | Monday |\n| November 11, 2016 | Veterans Day | Friday |\n| November 24, 2016 | Thanksgiving Day | Thursday |\n| December 25, 2016 | Christmas Day | Sunday |\n| December 26, 2016 | Christmas Day (observed) | Monday |\n| January 1, 2017 | New Year's Day | Sunday |\n| January 2, 2017 | New Year's Day (observed) | Monday |\n| January 16, 2017 | Martin Luther King Day | Monday |\n| February 20, 2017 | Presidents' Day | Monday |\n| May 29, 2017 | Memorial Day | Monday |\n| July 4, 2017 | Independence Day | Tuesday |\n| September 4, 2017 | Labor Day | Monday \n| October 9, 2017 | Columbus Day | Monday |\n|November 10, 2017| Veterans Day (observed) | Friday |\n| November 11, 2017 | Veterans Day | Saturday |\n| November 23, 2017 | Thanksgiving Day | Thursday |\n| December 25, 2017 | Christmas Day | Monday |\n| January 1, 2018 | New Year's Day | Monday |\n| January 15, 2018 | Martin Luther King Day | Monday |\n| February 19, 2018 | Presidents' Day | Monday |\n| May 28, 2018 | Memorial Day | Monday |\n| July 4, 2018 | Independence Day | Wednesday |\n| September 3, 2018 | Labor Day | Monday |\n| October 8, 2018 | Columbus Day | Monday |\n| November 11, 2018 | Veterans Day | Sunday |\n| November 12, 2018 | Veterans Day (observed) | Monday |\n| November 22, 2018 | Thanksgiving Day | Thursday |\n| December 25, 2018 | Christmas Day | Tuesday |\n\nIn my solution feature based on this data bring improvements on validation = 0.02.", "votes": 42 }, { "id": 652108, "postDate": "2019-10-18T11:16:04.360Z", "content": "```
\n\nfrom pandas.tseries.holiday import USFederalHolidayCalendar as calendar\ndates_range = pd.date_range(start='2015-12-31', end='2019-01-01')\nus_holidays = calendar().holidays(start=dates_range.min(), end=dates_range.max())
\n\ndf['is_holiday'] = (df['timestamp'].dt.date.astype('datetime64').isin(us_holidays)).astype(np.int8)
\n\n```
\n\nYou can use pandas datetime for it. The problem here is that we don't know site_id geo location (US or not)
", "rawMarkdown": "```\n\nfrom pandas.tseries.holiday import USFederalHolidayCalendar as calendar\ndates_range = pd.date_range(start='2015-12-31', end='2019-01-01')\nus_holidays = calendar().holidays(start=dates_range.min(), end=dates_range.max())\n\ndf['is_holiday'] = (df['timestamp'].dt.date.astype('datetime64').isin(us_holidays)).astype(np.int8)\n\n```\n\nYou can use pandas datetime for it. The problem here is that we don't know site_id geo location (US or not)", "votes": 27 }, { "id": 652465, "postDate": "2019-10-18T22:09:06.400Z", "content": "Nice job - I agree with your guess that the data is USA cities. While a world wide organization now, data matching this has been collected in USA for last decade.
\n\nWould recommend some changes however if your working on USA assumption you should go all the way!
\n\nDays with high energy use possible.\nDay after Thanksgiving.\nSuperbowl Sunday.
\n\nOther Days with Low.\nChristmas eve.\nAny four day long weekend connected with holidays.
\n\nDays where lots of folks still work and might not get much of an energy drop.\nColumbus Day
\n\nHere in Pennsylvania over one million folks out in woods hunting deer on the Monday after Thanksgiving.
", "rawMarkdown": "Nice job - I agree with your guess that the data is USA cities. While a world wide organization now, data matching this has been collected in USA for last decade.\n\nWould recommend some changes however if your working on USA assumption you should go all the way!\n\nDays with high energy use possible.\nDay after Thanksgiving.\nSuperbowl Sunday.\n\nOther Days with Low.\nChristmas eve.\nAny four day long weekend connected with holidays.\n\nDays where lots of folks still work and might not get much of an energy drop.\nColumbus Day\n\nHere in Pennsylvania over one million folks out in woods hunting deer on the Monday after Thanksgiving. ", "votes": 1, "replies": [ { "id": 652468, "postDate": "2019-10-18T22:16:57.560Z", "content": "Hi @pcjimmmy , \nVery useful remarks)\nWill try it is in my code.\nI am not an expert in US holidays just took days that I found. Thanks for help!
", "rawMarkdown": "Hi @pcjimmmy , \nVery useful remarks)\nWill try it is in my code.\nI am not an expert in US holidays just took days that I found. Thanks for help!" }, { "id": 652488, "postDate": "2019-10-18T23:14:02.303Z", "content": "Another low energy - day before new years.
", "rawMarkdown": "Another low energy - day before new years. " }, { "id": 654648, "postDate": "2019-10-22T06:21:07.573Z", "content": "I plotted the average air temperature yearly groupby month.\nA lot of the yearly temperature can be found in US.
\n\ncheck this out -> https://www.kaggle.com/c/ashrae-energy-prediction/discussion/113772#latest-654627
", "rawMarkdown": "I plotted the average air temperature yearly groupby month.\nA lot of the yearly temperature can be found in US.\n\ncheck this out -> https://www.kaggle.com/c/ashrae-energy-prediction/discussion/113772#latest-654627" } ] }, { "id": 652064, "postDate": "2019-10-18T09:27:16.923Z", "content": "Cheers @isaienkov \nI couldn't find where the data is coming from. Are you sure it's US [only]?
", "rawMarkdown": "Cheers @isaienkov \nI couldn't find where the data is coming from. Are you sure it's US [only]?", "votes": 2, "replies": [ { "id": 652071, "postDate": "2019-10-18T09:35:51.413Z", "content": "@hmendonca , I took data from here http://www.calendarpedia.com/holidays/federal-holidays-2016.html.\nBut you are completely correct, I am not sure that it is only US buildings.
", "rawMarkdown": "@hmendonca , I took data from here http://www.calendarpedia.com/holidays/federal-holidays-2016.html.\nBut you are completely correct, I am not sure that it is only US buildings. ", "votes": 1 }, { "id": 652078, "postDate": "2019-10-18T09:52:00.917Z", "content": "It is probably US only since the ASHRAE stands for The American Society of Heating, Refrigerating and Air-Conditioning Engineers
@gunesevitan , \nThat's why I did my hypothesis.
", "rawMarkdown": "@gunesevitan , \nThat's why I did my hypothesis. " }, { "id": 652625, "postDate": "2019-10-19T06:08:52.683Z", "content": "Its probably USA because studies with this exact data breakdown have occurred multiple times in last decade. I can find summary reports for those years, but yet to find any actual data to supplement modeling.
", "rawMarkdown": "Its probably USA because studies with this exact data breakdown have occurred multiple times in last decade. I can find summary reports for those years, but yet to find any actual data to supplement modeling." }, { "id": 652944, "postDate": "2019-10-19T17:17:49.690Z", "content": "@gunreddy @isaienkov @pcjimmmy \nFair point. But a quick google search will review that ASHRAE has a strong global presence.\n- https://www.ashraeindia.org/\n- https://midlands.ashrae.uk/\n- https://ashrae.org.sg/\n- http://www.ashrae.org.hk/
\n\nIn fact, many building authorities follow ASHRAE guidelines with regards to HVAC.\n- https://www.researchgate.net/profile/Sam_C_M_Hui/publication/2410830_Review_of_building_energy_standards_and_implications_for_Hong_Kong/links/543fb7d30cf2be1758cf51cb/Review-of-building-energy-standards-and-implications-for-Hong-Kong.pdf\n- https://www.jstage.jst.go.jp/article/shasetaikai/2014.6/0/2014.6_17/_article/-char/ja/\n- http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.549.6457&rep=rep1&type=pdf
", "rawMarkdown": "@gunreddy @isaienkov @pcjimmmy \nFair point. But a quick google search will review that ASHRAE has a strong global presence.\n- https://www.ashraeindia.org/\n- https://midlands.ashrae.uk/\n- https://ashrae.org.sg/\n- http://www.ashrae.org.hk/\n\n In fact, many building authorities follow ASHRAE guidelines with regards to HVAC.\n- https://www.researchgate.net/profile/Sam_C_M_Hui/publication/2410830_Review_of_building_energy_standards_and_implications_for_Hong_Kong/links/543fb7d30cf2be1758cf51cb/Review-of-building-energy-standards-and-implications-for-Hong-Kong.pdf\n- https://www.jstage.jst.go.jp/article/shasetaikai/2014.6/0/2014.6_17/_article/-char/ja/\n- http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.549.6457&rep=rep1&type=pdf\n\n", "votes": 2 }, { "id": 653028, "postDate": "2019-10-19T20:21:14.890Z", "content": "Did not check your links yet - but do any include annual summary reports that contain the ASHRAE buildings, etc ?
\n\nI was very modestly involved in data collection several years ago for a building in DOE survey. It's a decent investment to collect the data, standardize things, etc for the building owner and data collector. Someone has to pay for that survey effort - I know it's been done in USA by DOE. You can find reports on USA for summary of the years where surveys made.
\n\nNext point - plot the temperatures over the 3 years. They are certainly north hemisphere values. For example, not going to match Hong Kong weather patterns.
\n\nLast point - it's your model. Do it your way :)
", "rawMarkdown": "Did not check your links yet - but do any include annual summary reports that contain the ASHRAE buildings, etc ? \n\nI was very modestly involved in data collection several years ago for a building in DOE survey. It's a decent investment to collect the data, standardize things, etc for the building owner and data collector. Someone has to pay for that survey effort - I know it's been done in USA by DOE. You can find reports on USA for summary of the years where surveys made.\n\nNext point - plot the temperatures over the 3 years. They are certainly north hemisphere values. For example, not going to match Hong Kong weather patterns.\n\nLast point - it's your model. Do it your way :)" }, { "id": 653812, "postDate": "2019-10-21T01:37:25.537Z", "content": "The data description mentions: \"The dataset includes three years of hourly meter readings from over one thousand buildings at several different sites around the world.\" \nThe data might be from different parts of the world as well? Thoughts?
", "rawMarkdown": "The data description mentions: \"The dataset includes three years of hourly meter readings from over one thousand buildings at several different sites around the world.\" \nThe data might be from different parts of the world as well? Thoughts?", "votes": 1 } ] }, { "id": 652192, "postDate": "2019-10-18T13:51:24.443Z", "content": "Which one are considered long weekends? :)))
", "rawMarkdown": "Which one are considered long weekends? :)))" }, { "id": 666230, "postDate": "2019-11-05T22:51:27.957Z", "rawMarkdown": "", "isDeleted": true } ], "comments": [ { "id": 652108, "author_name": "Konstantin Yakovlev", "author_url": "", "post_date": "2019-10-18T11:16:04.360000", "content": "```
\n\nfrom pandas.tseries.holiday import USFederalHolidayCalendar as calendar\ndates_range = pd.date_range(start='2015-12-31', end='2019-01-01')\nus_holidays = calendar().holidays(start=dates_range.min(), end=dates_range.max())
\n\ndf['is_holiday'] = (df['timestamp'].dt.date.astype('datetime64').isin(us_holidays)).astype(np.int8)
\n\n```
\n\nYou can use pandas datetime for it. The problem here is that we don't know site_id geo location (US or not)
", "votes": 27, "replies": [] }, { "id": 652465, "author_name": "PC Jimmmy", "author_url": "", "post_date": "2019-10-18T22:09:06.400000", "content": "Nice job - I agree with your guess that the data is USA cities. While a world wide organization now, data matching this has been collected in USA for last decade.
\n\nWould recommend some changes however if your working on USA assumption you should go all the way!
\n\nDays with high energy use possible.\nDay after Thanksgiving.\nSuperbowl Sunday.
\n\nOther Days with Low.\nChristmas eve.\nAny four day long weekend connected with holidays.
\n\nDays where lots of folks still work and might not get much of an energy drop.\nColumbus Day
\n\nHere in Pennsylvania over one million folks out in woods hunting deer on the Monday after Thanksgiving.
", "votes": 1, "replies": [ { "id": 652468, "author_name": "Kostiantyn Isaienkov", "author_url": "", "post_date": "2019-10-18T22:16:57.560000", "content": "Hi @pcjimmmy , \nVery useful remarks)\nWill try it is in my code.\nI am not an expert in US holidays just took days that I found. Thanks for help!
", "votes": 0, "replies": [] }, { "id": 652488, "author_name": "PC Jimmmy", "author_url": "", "post_date": "2019-10-18T23:14:02.303000", "content": "Another low energy - day before new years.
", "votes": 0, "replies": [] }, { "id": 654648, "author_name": "CC Joshua ", "author_url": "", "post_date": "2019-10-22T06:21:07.573000", "content": "I plotted the average air temperature yearly groupby month.\nA lot of the yearly temperature can be found in US.
\n\ncheck this out -> https://www.kaggle.com/c/ashrae-energy-prediction/discussion/113772#latest-654627
", "votes": 0, "replies": [] } ] }, { "id": 652064, "author_name": "Henrique Mendonça", "author_url": "", "post_date": "2019-10-18T09:27:16.923000", "content": "Cheers @isaienkov \nI couldn't find where the data is coming from. Are you sure it's US [only]?
", "votes": 2, "replies": [ { "id": 652071, "author_name": "Kostiantyn Isaienkov", "author_url": "", "post_date": "2019-10-18T09:35:51.413000", "content": "@hmendonca , I took data from here http://www.calendarpedia.com/holidays/federal-holidays-2016.html.\nBut you are completely correct, I am not sure that it is only US buildings.
", "votes": 1, "replies": [] }, { "id": 652078, "author_name": "Gunes Evitan", "author_url": "", "post_date": "2019-10-18T09:52:00.917000", "content": "It is probably US only since the ASHRAE stands for The American Society of Heating, Refrigerating and Air-Conditioning Engineers
@gunesevitan , \nThat's why I did my hypothesis.
", "votes": 0, "replies": [] }, { "id": 652625, "author_name": "PC Jimmmy", "author_url": "", "post_date": "2019-10-19T06:08:52.683000", "content": "Its probably USA because studies with this exact data breakdown have occurred multiple times in last decade. I can find summary reports for those years, but yet to find any actual data to supplement modeling.
", "votes": 0, "replies": [] }, { "id": 652944, "author_name": "Daniel", "author_url": "", "post_date": "2019-10-19T17:17:49.690000", "content": "@gunreddy @isaienkov @pcjimmmy \nFair point. But a quick google search will review that ASHRAE has a strong global presence.\n- https://www.ashraeindia.org/\n- https://midlands.ashrae.uk/\n- https://ashrae.org.sg/\n- http://www.ashrae.org.hk/
\n\nIn fact, many building authorities follow ASHRAE guidelines with regards to HVAC.\n- https://www.researchgate.net/profile/Sam_C_M_Hui/publication/2410830_Review_of_building_energy_standards_and_implications_for_Hong_Kong/links/543fb7d30cf2be1758cf51cb/Review-of-building-energy-standards-and-implications-for-Hong-Kong.pdf\n- https://www.jstage.jst.go.jp/article/shasetaikai/2014.6/0/2014.6_17/_article/-char/ja/\n- http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.549.6457&rep=rep1&type=pdf
", "votes": 2, "replies": [] }, { "id": 653028, "author_name": "PC Jimmmy", "author_url": "", "post_date": "2019-10-19T20:21:14.890000", "content": "Did not check your links yet - but do any include annual summary reports that contain the ASHRAE buildings, etc ?
\n\nI was very modestly involved in data collection several years ago for a building in DOE survey. It's a decent investment to collect the data, standardize things, etc for the building owner and data collector. Someone has to pay for that survey effort - I know it's been done in USA by DOE. You can find reports on USA for summary of the years where surveys made.
\n\nNext point - plot the temperatures over the 3 years. They are certainly north hemisphere values. For example, not going to match Hong Kong weather patterns.
\n\nLast point - it's your model. Do it your way :)
", "votes": 0, "replies": [] }, { "id": 653812, "author_name": "Pratik Gandhi", "author_url": "", "post_date": "2019-10-21T01:37:25.537000", "content": "The data description mentions: \"The dataset includes three years of hourly meter readings from over one thousand buildings at several different sites around the world.\" \nThe data might be from different parts of the world as well? Thoughts?
", "votes": 1, "replies": [] } ] }, { "id": 652192, "author_name": "Adrian Zinovei", "author_url": "", "post_date": "2019-10-18T13:51:24.443000", "content": "Which one are considered long weekends? :)))
", "votes": 0, "replies": [] }, { "id": 666230, "author_name": "", "author_url": "", "post_date": "2019-11-05T22:51:27.957000", "content": "", "votes": 0, "replies": [] } ], "raw_markdown_by_id": { "652035": "I have a hypothesis that holidays can affect to energy consumption. For example in holidays Educational buildings should use less energy, water, but another situation should be in cafes and similar entertainment buildings. So there is a list of US holidays for 2016-2018 years:\n\n| Date | Holiday name | Day |\n| --- | --- |\n| January 1, 2016 | New Year's Day | Friday |\n| January 18, 2016 | Martin Luther King Day | Monday |\n| February 15, 2016 | Presidents' Day | Monday |\n| May 30, 2016 | Memorial Day | Monday |\n| July 4, 2016 | Independence Day | Monday |\n| September 5, 2016 | Labor Day | Monday |\n| October 10, 2016 | Columbus Day | Monday |\n| November 11, 2016 | Veterans Day | Friday |\n| November 24, 2016 | Thanksgiving Day | Thursday |\n| December 25, 2016 | Christmas Day | Sunday |\n| December 26, 2016 | Christmas Day (observed) | Monday |\n| January 1, 2017 | New Year's Day | Sunday |\n| January 2, 2017 | New Year's Day (observed) | Monday |\n| January 16, 2017 | Martin Luther King Day | Monday |\n| February 20, 2017 | Presidents' Day | Monday |\n| May 29, 2017 | Memorial Day | Monday |\n| July 4, 2017 | Independence Day | Tuesday |\n| September 4, 2017 | Labor Day | Monday \n| October 9, 2017 | Columbus Day | Monday |\n|November 10, 2017| Veterans Day (observed) | Friday |\n| November 11, 2017 | Veterans Day | Saturday |\n| November 23, 2017 | Thanksgiving Day | Thursday |\n| December 25, 2017 | Christmas Day | Monday |\n| January 1, 2018 | New Year's Day | Monday |\n| January 15, 2018 | Martin Luther King Day | Monday |\n| February 19, 2018 | Presidents' Day | Monday |\n| May 28, 2018 | Memorial Day | Monday |\n| July 4, 2018 | Independence Day | Wednesday |\n| September 3, 2018 | Labor Day | Monday |\n| October 8, 2018 | Columbus Day | Monday |\n| November 11, 2018 | Veterans Day | Sunday |\n| November 12, 2018 | Veterans Day (observed) | Monday |\n| November 22, 2018 | Thanksgiving Day | Thursday |\n| December 25, 2018 | Christmas Day | Tuesday |\n\nIn my solution feature based on this data bring improvements on validation = 0.02.", "652108": "```\n\nfrom pandas.tseries.holiday import USFederalHolidayCalendar as calendar\ndates_range = pd.date_range(start='2015-12-31', end='2019-01-01')\nus_holidays = calendar().holidays(start=dates_range.min(), end=dates_range.max())\n\ndf['is_holiday'] = (df['timestamp'].dt.date.astype('datetime64').isin(us_holidays)).astype(np.int8)\n\n```\n\nYou can use pandas datetime for it. The problem here is that we don't know site_id geo location (US or not)", "652465": "Nice job - I agree with your guess that the data is USA cities. While a world wide organization now, data matching this has been collected in USA for last decade.\n\nWould recommend some changes however if your working on USA assumption you should go all the way!\n\nDays with high energy use possible.\nDay after Thanksgiving.\nSuperbowl Sunday.\n\nOther Days with Low.\nChristmas eve.\nAny four day long weekend connected with holidays.\n\nDays where lots of folks still work and might not get much of an energy drop.\nColumbus Day\n\nHere in Pennsylvania over one million folks out in woods hunting deer on the Monday after Thanksgiving. ", "652064": "Cheers @isaienkov \nI couldn't find where the data is coming from. Are you sure it's US [only]?", "652192": "Which one are considered long weekends? :)))", "666230": "" } }