|
Apr
16 |
|
awarded | Notable Question |
|
Jan
23 |
|
awarded | Popular Question |
|
Dec
2 |
|
awarded | Popular Question |
|
Apr
21 |
|
revised |
R error: `contrasts<-`(`*tmp*`, value = contr.funs[1 + isOF[nn]]) deleted 12 characters in body |
|
Apr
21 |
|
revised |
R error: `contrasts<-`(`*tmp*`, value = contr.funs[1 + isOF[nn]]) added 1058 characters in body |
|
Apr
21 |
|
revised |
R error: `contrasts<-`(`*tmp*`, value = contr.funs[1 + isOF[nn]]) edited tags |
|
Apr
21 |
|
revised |
Missing data and Attributes selection edited tags |
|
Apr
21 |
|
comment |
R error: `contrasts<-`(`*tmp*`, value = contr.funs[1 + isOF[nn]]) I'm sorry I didn't get ur point. I'm new to R. can you explain ur point in some details? |
|
Apr
21 |
|
comment |
R error: `contrasts<-`(`*tmp*`, value = contr.funs[1 + isOF[nn]]) That is part of it. str(mydata) 'data.frame': 50000 obs. of 273 variables: $ CLICK_FLG : int 0 0 0 0 0 0 0 0 0 0 ... $ OPEN_FLG :Factor w/ 2 levels "N","Y": 1 1 1 1 1 1 ... $ ADDR_VER_CD : Factor w/ 4 levels "","M","N","Y": 1 1 1 1 1 1 1 1 1 1 ... $ AQI :Factor w/ 6 levels "","E","G","M",..: 2 2 2 2 2 2 2 2 2 2 ... $ BIRD_QTY : int 0 NA 0 NA NA 0 0 0 0 0 ... |
|
Apr
21 |
|
comment |
R error: `contrasts<-`(`*tmp*`, value = contr.funs[1 + isOF[nn]]) @MYaseen208 can you please describe more. my data has factors with 2 or more levels, int, num. some of the int columns has only 1 value. could that be the reason? |
|
Apr
21 |
|
asked | R error: `contrasts<-`(`*tmp*`, value = contr.funs[1 + isOF[nn]]) |
|
Apr
21 |
|
awarded | Editor |
|
Apr
21 |
|
revised |
Missing data and Attributes selection added 581 characters in body |
|
Apr
21 |
|
awarded | Supporter |
|
Apr
20 |
|
comment |
Missing data and feature selection I thought of eliminating bcz the value is missing for ALL records except very few (like 20 records of 1,785,000 records). Also the validation data (9500 records) has totally missing data for that variable. It's loke I have this column with no information at all so I just can't understand how its presence will be valuable for training process! For Q2, [1 refers to music preference, 3 to reading preference, 5 to activity preference]. The goal is to predict if the user will response to email by click(Y) or notClick(N). No missing data for DV. Only nearly 1% in training set response by Y. |
|
Apr
20 |
|
asked | taking data sample in R |
|
Apr
20 |
|
asked | Missing data and feature selection |
|
Apr
20 |
|
comment |
Regression Model for categorical data thank you for your answer. My problem with weka that it gives the memory error message when I open the csv file, i.e it even can't read the data mentioned above, I divided it to smaller csv files and the maximum records it was able to read only 100,000 even though I fixed the heap size to 2G (i have 4G RAM). any ideas about this? |
|
Apr
20 |
|
asked | Missing data and Attributes selection |
|
Apr
18 |
|
comment |
Logistic regression on categorical data and sorry another beginner Q: in the previous code i put only one independent variable. how will be the code if I want the model to include all the IV in mydata? i know it will be like mydata.logr <- glm(mydata ~ new_id + mailing_id + Age, family=binomial("logit")) Is there another way to write it to include all the fields in the data? |
