algonell
Error on xgboost step
Hey,
They changed the data (~ a month ago, added new features and explicit validation set), this notebook is for the old data sets.
Your are welcome to update it and create a pull request.
Kindly regards,
On Wed, May 17, 2017 at 6:25 AM, AIAdventures [email protected] wrote:
Hi awesome notebook, trying to run through the examples. Best, Andrew
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/algonell/PyData/issues/1#issuecomment-301974963, or mute the thread https://github.com/notifications/unsubscribe-auth/AEc-SbkUmnLWMB2_IGc5UtCgl5LJ4aDEks5r6miogaJpZM4NdVMC .
--
Andrew Kreimer [email protected][email protected] +972-542114277
Algonell - Scientific FX Trading http://www.algonell.com
[image: Twitter] https://twitter.com/algonell [image: Facebook] https://www.facebook.com/algonell [image: Google Plus] https://www.google.com/+Algonell [image: Linkedin] https://il.linkedin.com/in/andrewkreimer https://sites.google.com/site/algonellfx/home
so i have tried to update the notebook a couple of times. dropping specific columns which i think is the only difference with the current data. however i keep getting the above error and i am not sure where i am going wrong. Many thanks for your help. Best, Andrew
hi,
it's hard to see what's going on, please share the code.
maybe the class variable is the first feature instead of the last one?
kindly regards,
On Mon, May 29, 2017 at 12:28 PM, AIAdventures [email protected] wrote:
so i have tried to update the notebook a couple of times. dropping specific columns which i think is the only difference with the current data. however i keep getting the above error and i am not sure where i am going wrong. Many thanks for your help. Best, Andrew
— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/algonell/PyData/issues/1#issuecomment-304616835, or mute the thread https://github.com/notifications/unsubscribe-auth/AEc-STNzXJM8P1IgQFr49Td8FMx0lGpFks5r-o-sgaJpZM4NdVMC .
--
Andrew Kreimer [email protected][email protected] +972-542114277
Algonell - Scientific FX Trading http://www.algonell.com
[image: Twitter] https://twitter.com/algonell [image: Facebook] https://www.facebook.com/algonell [image: Google Plus] https://www.google.com/+Algonell [image: Linkedin] https://il.linkedin.com/in/andrewkreimer https://sites.google.com/site/algonellfx/home
Almost certain it's the extra variables that are causing the problem but I can't seem the get the entire notebook the run. Many thanks, Best, Andrew
hi,
it's related to:
target = train_data['target'] train_data = pd.DataFrame(poly.transform(train_data.values[:,:-1])) test_data = pd.DataFrame(poly.transform(test_data.values)) train_data['target'] = target.values
be sure to have a copy of 'target' column before and after Poly features fit and transform.
make sure you run the cells in the right order, it's exactly like my ongoing code.
kindly regards,
On Mon, May 29, 2017 at 12:57 PM, AIAdventures [email protected] wrote:
Almost certain it's the extra variables that are causing the problem but I can't seem the get the entire notebook the run. Many thanks, Best, Andrew
— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/algonell/PyData/issues/1#issuecomment-304622891, or mute the thread https://github.com/notifications/unsubscribe-auth/AEc-SesyzP1Mgwrq3zVapCOL92oUEUIdks5r-pZ6gaJpZM4NdVMC .
--
Andrew Kreimer [email protected][email protected] +972-542114277
Algonell - Scientific FX Trading http://www.algonell.com
[image: Twitter] https://twitter.com/algonell [image: Facebook] https://www.facebook.com/algonell [image: Google Plus] https://www.google.com/+Algonell [image: Linkedin] https://il.linkedin.com/in/andrewkreimer https://sites.google.com/site/algonellfx/home
try running without the Poly features, check if the entire flow works.
On May 29, 2017 13:51, "AIAdventures" [email protected] wrote:
in my data set come NA values.
— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/algonell/PyData/issues/1#issuecomment-304632869, or mute the thread https://github.com/notifications/unsubscribe-auth/AEc-SbUcc6YITVzy1gfbKUs-wmDG6yxiks5r-qMjgaJpZM4NdVMC .
Ah ok will try and see what happens. Best, Andrew
On 29 May 2017 9:29 pm, "algonell" [email protected] wrote:
try running without the Poly features, check if the entire flow works.
On May 29, 2017 13:51, "AIAdventures" [email protected] wrote:
in my data set come NA values.
— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/algonell/PyData/issues/1#issuecomment-304632869, or mute the thread <https://github.com/notifications/unsubscribe- auth/AEc-SbUcc6YITVzy1gfbKUs-wmDG6yxiks5r-qMjgaJpZM4NdVMC> .
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/algonell/PyData/issues/1#issuecomment-304639125, or mute the thread https://github.com/notifications/unsubscribe-auth/AY5K57EXJj7hygnQ-qS66bfX-OdwOPfXks5r-qwlgaJpZM4NdVMC .