Knn accuracy and code doubt

[rohit_1906 (Rohit Dhonukshe)]

https://drive.google.com/file/d/1kqXPRDOWT7uha7esHGvihvu-Dt7xsJzh/view?usp=sharing

is 73 percent accuracy okay for knn or something can be improved?

it is showing me a warning in the code

:11: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify ‘dtype=object’ when creating the ndarray
v = np.array(v)

how can i fix this

[prashant_ml (Prashant Arora)]

@rohit_1906

is 73 percent accuracy okay for knn or something can be improved?

yeah its absolutly good to have such accuracy with KNN. if you try to do more feature engineering techniques , then i guess you can improve it more to around 78-80 %.

@rohit_1906

how can i fix this

For this i guess you are converting your a list of tuples or something to a numpy array. Hence it is giving such warning. You can ignore it or specify dtype="object" like v= np.array(v,dtype="object")

I hope this helps you .