Pfthb

I am trying to write a function which will split a dataframe "df" into smaller dataframes df1, df2...dfn such that all records with the same ID (from column "UNIT-ID") are grouped together and stored in those smaller dataframes, in the folder respectively. I want them named as df1,df2..dfn since I plan to iterate over those n dataframes individually and use them to train my neural network model. I am also looking to write the neural network model as a function and apply the function over these split dataframes.

I have checked this and this but it talks about groupby and iterating over them. I want the extraction of the groups to be done within the same function without having to reference it or extract it with another line of code outside the function, as described in the mentioned links

Below is an example of how my data looks and how I want it to be split.

Input:

UNIT-ID Q1 Q2 Q3
110-P1 37 487 0
110-P1 31 140 1
110-P1 46 214 1
110-P2 29 287 1
110-P2 45 131 1
110-P2 39 260 0
110-P5 13 413 1
110-P5 13 259 1
110-P5 16 226 1
110-P5 49 456 1
110-P5 39 275 1
110-P10 41 189 0
110-P10 32 269 0
110-P10 31 123 1
110-P10 32 487 0
110-P10 24 398 0
110-15 23 346 0
110-15 31 419 1
110-15 37 287 0
110-15 36 228 1
110-15 48 309 1

Output:
df1: 
 UNIT-ID Q1 Q2 Q3
 110-P1 37 487 0
 110-P1 31 140 1
 110-P1 46 214 1

df2: 
 UNIT-ID Q1 Q2 Q3
 110-P2 29 287 1
 110-P2 45 131 1
 110-P2 39 260 0

df3: 
 UNIT-ID Q1 Q2 Q3
 110-P5 13 413 1
 110-P5 13 259 1
 110-P5 16 226 1
 110-P5 49 456 1
 110-P5 39 275 1

df4: 
 UNIT-ID Q1 Q2 Q3
 110-P10 41 189 0
 110-P10 32 269 0
 110-P10 31 123 1
 110-P10 32 487 0
 110-P10 24 398 0

df5: 
 UNIT-ID Q1 Q2 Q3
 110-15 23 346 0
 110-15 31 419 1
 110-15 37 287 0
 110-15 36 228 1
 110-15 48 309 1
 110-15 34 390 1
 110-15 67 238 1

Any help would be appreciated!