How to increase the performance of Sklearn GMM ()?

I am using Sklearn to estimate a Gaussian Mixture Model (GMM) for some data.

After evaluation, I have many query points. I would like to get their probabilities of belonging to each of the estimated Gaussian ones.

Below is the code. However, the part is gmm_sk.predict_proba(query_points)

very slow as I need to run it multiple times on 100000 sample sets where each sample contains 1000 points.

I assume this is because it is consistent. Is there a way to make it parallel? Or any other way to make it faster? Maybe on a GPU using TensorFlow?

I saw that TensorFlow has its own GMM algorithm , but it was very difficult to implement.

Here is the code I wrote:

import numpy as np
from sklearn.mixture import GaussianMixture
import time


n_gaussians = 1000
covariance_type = 'diag'
points = np.array(np.random.rand(10000, 3), dtype=np.float32)
query_points = np.array(np.random.rand(1000, 3), dtype=np.float32)
start = time.time()

#GMM with sklearn
gmm_sk = GaussianMixture(n_components = n_gaussians, covariance_type=covariance_type)
gmm_sk.fit(points)
mid_t = time.time()
elapsed = time.time() - start
print("learning took "+ str(elapsed))

temp = []
for i in range(2000):
    temp.append(gmm_sk.predict_proba(query_points))

end_t = time.time() - mid_t
print("predictions took " + str(end_t))    

      

        
        
        
      

    

I solved it! using multiprocessing

. just replaced

temp = []
for i in range(2000):
    temp.append(gmm_sk.predict_proba(query_points))

      

        
        
        
      

    

from

import multiprocessing as mp
    query_points = query_points.tolist()
    parallel = mp.Pool()
    fv = parallel.map(par_gmm, query_points)
    parallel.close()
    parallel.join()