NeerajSarwan
question on load_files() function
looking at the code for the load_files() function, why did you append the values of the test file in the data list?
for index, row in test.iterrows(): data.append(row.values) target.append(list(row.values))
Given, the sequential nature of the problem, test set can be merged with the train set to create a bigger training dataset. Hence the append.
Thank you for the response.
Can the merging be optional? And I'm looking forward for the python implementation of the other algorithms (when can this be?). And one last thing, if you have any idea on time series that can be made into an example using CPT is a great help.
Again thank you
The merging is currently optional. Merging happens only when you supply a test file to the load_files. Currently modifying the structure. Keep an eye out.
Thank you.
I have a question, why do you add the list() command in test.append(list(row.values)), while you did not use it in 'train.append(row.values)' ?
Given, the sequential nature of the problem, test set can be merged with the train set to create a bigger training dataset. Hence the append. it is mistaken to add test data for training, they call it data leaking. can you test your code without such a bug?
@Sandy4321 I am well aware about what data leaking is. But this isn't the case of data leaking as far as I know. Also, there merging of train and test is not mandatory as I have mentioned above. Feel free to fork the repository and test out the code.
@jermaine1ronquillo, I have removed the inconsistency. The train and test append operation should be consistent now. Thanks for pointing it out.