the data shape problem during adaboost

[colasgael]

the data shape problem during adaboost

[adelekuzmiakova]

Hi joonv2,

Thanks so much for trying out our code. Sincere apologies for the delay in addressing your question; I didn't check Gael's repository up until now. With that in mind, I don't recall we applied the AdaBoost classifier to our dataset. Could you please specify what script are your running and where the error occurs?

To try to address your question, I can't see shapes being the problem. If we were to apply the AdaBoost we would select all the relevant features, such as humidity, hour, could cover, month, etc that could help predict the response (solar output). In this case, the shape of the training data would be equal to the number of features. You can see a similar example here:

https://www.datacamp.com/community/tutorials/adaboost-classifier-python

where x_train has 4 features. So the number of features is entirely up to you and it doesn't have to be bounded by 2. Hope this helps :) and please let me know if you are still stuck on something

Best,
-Adele

[AIwalker1]

Hi, thanks for the above information and sharing the code.

But I still have the same question with joonv2. The question is when I run this 'rnn.py' file
at this website:
https://github.com/ColasGael/Machine-Learning-for-Solar-Energy-Prediction/tree/master/Recurrent%20Neural%20Network

with necessary 'csv' data copied from ‘https://github.com/ColasGael/Machine-Learning-for-Solar-Energy-Prediction/tree/master/Weighted%20Linear%20Regression’.

The error information is :
File "rnn.py", line 242, in
main()
File "rnn.py", line 179, in main
adaboost.fit(train_data, train_labels)
File "C:\Users\yanyu\Anaconda3\envs\tensorflow\lib\site-packages\sklearn\ensemble\weight_boosting.py", line 412, in fit
return super(AdaBoostClassifier, self).fit(X, y, sample_weight)
File "C:\Users\yanyu\Anaconda3\envs\tensorflow\lib\site-packages\sklearn\ensemble\weight_boosting.py", line 128, in fit
self._validate_estimator()
File "C:\Users\yanyu\Anaconda3\envs\tensorflow\lib\site-packages\sklearn\ensemble\weight_boosting.py", line 428, in validate_estimator
if not has_fit_parameter(self.base_estimator, "sample_weight"):
File "C:\Users\yanyu\Anaconda3\envs\tensorflow\lib\site-packages\sklearn\utils\validation.py", line 845, in has_fit_parameter
return parameter in signature(estimator.fit).parameters
AttributeError: 'function' object has no attribute 'fit'

Could you explain this? thanks.