Hello @aknvictor, the GPU implementation is excellent. I was trying to call the bootstrap method, but I am always getting this error:
TypeError: resample() missing 1 required positional argument: 'y'
Which is caused:
----> 8 result = model.bootstrap(df_f, n_sampling=200)
313 total_effects = np.zeros([n_sampling, X.shape[1], X.shape[1]])
314 for i in range(n_sampling):
--> 315 self.fit(resample(X))
316 adjacency_matrices[i] = self._adjacency_matrix
317
I checked the code, and the bootstrap method only receives X(which is the whole dataset) and n_sampling. When I checked the code from the CPU lingam, it changed how the self-fit is called. I am unsure if I am sending the parameters wrong, or if there is an error in the bootstrap implementation of culingam. Thank you in advance for any help or advice with this issue I am having.
Hello @aknvictor, the GPU implementation is excellent. I was trying to call the bootstrap method, but I am always getting this error:
TypeError: resample() missing 1 required positional argument: 'y'
Which is caused:
----> 8 result = model.bootstrap(df_f, n_sampling=200)
313 total_effects = np.zeros([n_sampling, X.shape[1], X.shape[1]])
314 for i in range(n_sampling):
--> 315 self.fit(resample(X))
316 adjacency_matrices[i] = self._adjacency_matrix
317
I checked the code, and the bootstrap method only receives X(which is the whole dataset) and n_sampling. When I checked the code from the CPU lingam, it changed how the self-fit is called. I am unsure if I am sending the parameters wrong, or if there is an error in the bootstrap implementation of culingam. Thank you in advance for any help or advice with this issue I am having.