replacementAI comments

Results 21 comments of


                                            replacementAI

Automatic closing of the positions at the end of the simulation

Still able to see open trades that are closed because they are the last ones.

Unsupervised K-Means Clustering Algorithm

@MrinalTyagi Do you think this would require a seperate estimator? Or could it be implemented in an existing one by specifying ```n_clusters='auto'```?

Examples Notebook

I am extremely confused as to how this library relates to sklearn in any way whatsoever, no where do the cross validation classes work in the sklearn API as simply...

Unsupervised K-Means Clustering Algorithm

> Would love to pick this up if no one is working on this. @MrinalTyagi I converted the authors MATLAB code to Python using Copilot ``` # load dataset iris...

Unsupervised K-Means Clustering Algorithm

> Would love to pick this up if no one is working on this. @MrinalTyagi I tried converting the above code into a class. Have 0 domain knowledge of any...

Unsupervised K-Means Clustering Algorithm

I'm not sure if we need to: 1. make a new MixIn or Base class 2. create an option for n_neighbors== 'auto' 3. create a new estimator like https://scikit-learn.org/stable/modules/generated/sklearn.neighbors.NearestNeighbors.html#sklearn.neighbors.NearestNeighbors

Unsupervised K-Means Clustering Algorithm

in class KNeighborsMixin: ``` def kneighbors(self, X=None, n_neighbors=None, return_distance=True): if n_neighbors == 'unsupervised': n_neighbors = self.n_neighbors clust, c_history = Unsupervised_K_Means_Clustering(thres=0.0001, beta=2, gamma=0.0001, rate=0, t_max=1000).fit(X, y=None) n_neighbors = int(np.median(clust)) return super().kneighbors(X,...