notebook.community

Edit and run 





In [2]:



    


import numpy as np
import pandas as pd
import pickle
import hdbscan



    











In [3]:



    


hdbscan_model = pickle.load(open('hdbscan_cluster_jac_10.pickle', 'rb'))



    











In [6]:



    


col_names = ['S'+str(i) for i in range(52)]



    











In [10]:



    


bmatrix_date = np.load('bmatrix_train_date.npy')
path_df = pd.DataFrame(data=bmatrix_date, columns=col_names, dtype=int)
del bmatrix_date



    











In [11]:



    


path_df.head()



    


















    
Out[11]:











  
    

      
      S0
      S1
      S2
      S3
      S4
      S5
      S6
      S7
      S8
      S9
      ...
      S42
      S43
      S44
      S45
      S46
      S47
      S48
      S49
      S50
      S51
    

  
  
    

      0
      0
      0
      0
      0
      0
      0
      0
      0
      0
      0
      ...
      0
      0
      0
      0
      0
      0
      0
      0
      0
      0
    

    

      1
      0
      0
      0
      0
      0
      0
      0
      0
      0
      0
      ...
      0
      0
      0
      0
      0
      0
      0
      0
      0
      0
    

    

      2
      0
      0
      0
      0
      0
      0
      0
      0
      0
      0
      ...
      0
      0
      0
      0
      0
      0
      0
      0
      0
      0
    

    

      3
      1
      1
      1
      0
      1
      0
      0
      1
      1
      0
      ...
      0
      0
      0
      0
      0
      0
      0
      0
      0
      0
    

    

      4
      0
      0
      0
      0
      0
      0
      0
      0
      0
      0
      ...
      0
      0
      0
      0
      0
      0
      0
      0
      0
      0
    

  



5 rows × 52 columns



















In [12]:



    


path_df['idx'] = path_df.index + 1



    











In [15]:



    


path_df = path_df.set_index('idx')



    











In [16]:



    


path_df_train = path_df[path_df.sum(axis=1) > 0]



    











In [20]:



    


test_labels, strengths = hdbscan.approximate_predict(hdbscan_model, path_df_train.values)
test_labels



    


















    
Out[20]:








array([665, 375, 670, ..., 778, 839, 811])

















In [29]:



    


%%time

hdbscan_model_man_10 = pickle.load(open('hdbscan_cluster_man_10.pickle', 'rb'))



    


















    






Wall time: 23.9 s

















In [44]:



    


test_labels_man_10, strengths = hdbscan.approximate_predict(hdbscan_model_man_10, path_df_train.values)
test_labels_man_10



    


















    
Out[44]:








array([ 570,  714, 1150, ...,  871,  890,  864])

















In [49]:



    


test_labels_man_10.shape



    


















    
Out[49]:








(1183165L,)

















In [50]:



    


path_df_train.shape



    


















    
Out[50]:








(1183165, 53)

















In [51]:



    


path_df_train['product'] = test_labels_man_10



    


















    






d:\Anaconda\envs\Deep2\lib\site-packages\ipykernel_launcher.py:1: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy
  """Entry point for launching an IPython kernel.

















In [52]:



    


path_df_train.to_csv("product_man_10.csv")



    











In [ ]:

Content source: zakkum42/Bosch

Similar notebooks:

notebook.community | gallery | about