Gaussian Mixed Model(GMMs)

A GMMs module written in Python

Features

Two typpes of initialization:

• Randomly initializing
• K-means clustering initializing

Basic usage

  gmms = GMMs(data,k) # data [n_sample,n_var], k gaussian component number
  model.EM(is_k_mean=False,N_iter_max=2000,
           is_plot=False,fig_fpath=path1,
           is_gif,gif_fpath=fpath2)

Example

  from sklearn.datasets.samples_generator import make_blobs
  X,Y = make_blobs(cluster_std=2,random_state=np.random.randint(100),
                   n_samples=500,centers=3)

  gmms = GMMs(x=X,k=3,max_iter=100,lh_theta=1e-30)
  gmms.EM(is_k_mean=False,
          is_plot=True,fig_fpath='images/example.png',
          is_gif=True,gif_fpath='images/example.gif')

Result:

---

• Work (5)

• github (2)