tag:blogger.com,1999:blog-1693014329567144872.post6371326443452591003..comments2023-08-27T06:49:20.658+01:00Comments on The Glowing Python: Forecasting beer consumption with sklearnJustGlowinghttp://www.blogger.com/profile/17212021288715206641noreply@blogger.comBlogger14125tag:blogger.com,1999:blog-1693014329567144872.post-51639778431539239492017-04-21T16:42:00.162+01:002017-04-21T16:42:00.162+01:00how can one add the analysis future prediction, ex...how can one add the analysis future prediction, ex 2013Q1Anonymoushttps://www.blogger.com/profile/06995910778550231533noreply@blogger.comtag:blogger.com,1999:blog-1693014329567144872.post-77654072389680226652016-10-15T13:02:07.281+01:002016-10-15T13:02:07.281+01:00Hi! make sure you have imported the required funct...Hi! make sure you have imported the required functions from matplotlib.pyplot.JustGlowinghttps://www.blogger.com/profile/17212021288715206641noreply@blogger.comtag:blogger.com,1999:blog-1693014329567144872.post-65235445836113123612016-10-15T12:54:40.206+01:002016-10-15T12:54:40.206+01:00Thanks for this interesting program. The following...Thanks for this interesting program. The following line does not work. Can you please help?<br />figure(figsize=(8,6))<br />Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-1693014329567144872.post-34750553952389284772016-08-03T04:24:38.694+01:002016-08-03T04:24:38.694+01:00hello, i am wondering when you predict 2012-q4, do...hello, i am wondering when you predict 2012-q4, do you mean predict('2012-q4')? to predict one by one? could you give an example? thank you.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-1693014329567144872.post-34716438032932913382016-04-26T21:20:16.778+01:002016-04-26T21:20:16.778+01:00That worked, thank you!That worked, thank you!Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-1693014329567144872.post-85522811659391318312016-04-22T20:49:22.014+01:002016-04-22T20:49:22.014+01:00hi, try to remove the "/" and merge the ...hi, try to remove the "/" and merge the line below:<br /><br />shape = to_forecast.shape[:-1] + (to_forecast.shape[-1] - window + 1, window)JustGlowinghttps://www.blogger.com/profile/17212021288715206641noreply@blogger.comtag:blogger.com,1999:blog-1693014329567144872.post-5885843119618997972016-04-22T20:31:46.693+01:002016-04-22T20:31:46.693+01:00i know this is an old post, but i was trying to re...i know this is an old post, but i was trying to replicate what you posted and i receive an error:<br /><br />on this line of code, it doesn't like the "+ /"; i get an invalid syntax error. <br /><br /> shape = to_forecast.shape[:-1] + /<br /><br />any thoughts on what could cause that?Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-1693014329567144872.post-807706512100157952015-08-05T08:52:09.108+01:002015-08-05T08:52:09.108+01:00If you consider the code in this post, you just ne...If you consider the code in this post, you just need to set h = 1 and use the predict method.JustGlowinghttps://www.blogger.com/profile/17212021288715206641noreply@blogger.comtag:blogger.com,1999:blog-1693014329567144872.post-24405485385949720462015-08-05T01:22:32.416+01:002015-08-05T01:22:32.416+01:00I am also interested in learning how to predict th...I am also interested in learning how to predict the value of a future point (say the next step). How would we implement the prediction on the next point in the time series?Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-1693014329567144872.post-34103992709357073042015-05-25T18:32:13.834+01:002015-05-25T18:32:13.834+01:00In first case, will accumulate some error, but les...In first case, will accumulate some error, but less or equal to second, right?<br /><br />I'm trying to do some example, but I'm kind a newbie on python and IA.<br /><br />Aah, by the way, it was me the Anonymous up there :).Cleyton Bonamigohttps://www.blogger.com/profile/11636677243692403732noreply@blogger.comtag:blogger.com,1999:blog-1693014329567144872.post-63813592966687122972015-05-24T09:13:17.229+01:002015-05-24T09:13:17.229+01:00Yes, you have two options. You can train the model...Yes, you have two options. You can train the model with the specific time horizon that you need or train the model with time horizon equal to 1 and predict the values one by one till you reach the point that you need. In the second case you have to reuse the predictions in between as known values to produce the prediction that follow, beware that in this way you will accumulate some error.JustGlowinghttps://www.blogger.com/profile/17212021288715206641noreply@blogger.comtag:blogger.com,1999:blog-1693014329567144872.post-57427786012583060712015-05-24T01:01:37.197+01:002015-05-24T01:01:37.197+01:00What I meant was: Can we antecipate values, like p...What I meant was: Can we antecipate values, like predict the values of 2012Q4 or more future?Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-1693014329567144872.post-59200475986598245102015-05-23T08:01:15.708+01:002015-05-23T08:01:15.708+01:00I don't really understand what you mean. If yo...I don't really understand what you mean. If you want an exampe of forecast you can refer to the picture. The red dashed line is the forecast and the light blue line is the actual consumption.JustGlowinghttps://www.blogger.com/profile/17212021288715206641noreply@blogger.comtag:blogger.com,1999:blog-1693014329567144872.post-46640890915792201052015-05-22T22:29:48.300+01:002015-05-22T22:29:48.300+01:00Do you have any example of how antecipate values?Do you have any example of how antecipate values?Anonymousnoreply@blogger.com