notebook.community

Edit and run 





In [1]:



    


import fingerprint as fp



    


















    






/usr/local/lib/python2.7/site-packages/matplotlib/font_manager.py:273: UserWarning: Matplotlib is building the font cache using fc-list. This may take a moment.
  warnings.warn('Matplotlib is building the font cache using fc-list. This may take a moment.')

















In [2]:



    


s_all=fp.read_pickle("struct_all.pickle")



    











In [4]:



    


import pandas as pd
import tqdm



    











In [5]:



    


s250=s_all[0:250]
f_ones_250=[fp.get_phi(s250[i],obser="ones",rmax=12,delta=0.05) for i in tqdm.tqdm_notebook(range(len(s250)))]
f_Z_250=[fp.get_phi(s250[i],obser="Z",rmax=12,delta=0.05) for i in tqdm.tqdm_notebook(range(len(s250)))]
f_Chi_250=[fp.get_phi(s250[i],obser="Chi",rmax=12,delta=0.05) for i in tqdm.tqdm_notebook(range(len(s250)))]



    


















    

























In [6]:



    


import matplotlib.pyplot as plt
%matplotlib inline
import numpy as np



    











In [12]:



    


plt.figure(figsize=(10,10))
r=np.linspace(0.05,12,(12/0.05))
for i in range(2):
    lab=s250[i].composition
    plt.plot(r,f_Z_250[i],label=lab)
plt.legend()



    


















    
Out[12]:








<matplotlib.legend.Legend at 0x12830d1d0>










    
























In [14]:



    


df_ones=pd.DataFrame(f_ones_250)



    











In [15]:



    


metric_ones=np.array([np.dot(f_ones_250[i],f_ones_250[j]) for i in range(250) for j in range(250)]).reshape(250,250)



    











In [16]:



    


metric_Z=np.array([np.dot(f_ones_250[i],f_ones_250[j]) for i in range(250) for j in range(250)]).reshape(250,250)



    











In [17]:



    


dist=np.array([np.dot(f_ones_250[0],f_ones_250[i]) for i in range(250)])



    











In [18]:



    


df_ones=pd.DataFrame({"phi_ones":f_ones_250})



    











In [19]:



    


df_ones["dist"]=dist

dt_sort=df_ones.sort("dist").drop("dist",axis=1)



    


















    






/usr/local/lib/python2.7/site-packages/ipykernel/__main__.py:3: FutureWarning: sort(columns=....) is deprecated, use sort_values(by=.....)
  app.launch_new_instance()

















In [20]:



    


comps=[s250[i].composition for i in range(250)]



    











In [21]:



    


df_ones["Composition"]=comps



    











In [22]:



    


dt_sort=df_ones.sort("dist",ascending=False)



    


















    






/usr/local/lib/python2.7/site-packages/ipykernel/__main__.py:1: FutureWarning: sort(columns=....) is deprecated, use sort_values(by=.....)
  if __name__ == '__main__':

















In [23]:



    


dt_sort.set_index(np.arange(dt_sort.shape[0]),inplace=True)



    











In [24]:



    


for i in range(5):
    print dt_sort["Composition"][i]



    


















    






Nb1 Ag1 O3
Ni1 Ag1 F3
Mg1 Ag1 F3
Zn1 Ag1 F3
S2 Bi1 Ag1

















In [25]:



    


sorted_f_ones=dt_sort['phi_ones'].values



    











In [26]:



    


metric_ones=np.array([np.dot(sorted_f_ones[i],sorted_f_ones[j]) for i in range(250) for j in range(250)]).reshape(250,250)



    











In [27]:



    


#for i in range(250):    
#    metric_ones[i][i]=0
plt.figure(figsize=(15,15))
plt.imshow(metric_ones[0:250,0:250])
plt.colorbar()



    


















    
Out[27]:








<matplotlib.colorbar.Colorbar at 0x11e4e2950>










    
























In [73]:



    


import sklearn



    











In [74]:



    


from sklearn.metrics.pairwise import euclidean_distances



    











In [79]:



    


euclid=np.array([np.sqrt(np.dot(sorted_f_ones[i],sorted_f_ones[i])+np.dot(sorted_f_ones[j],sorted_f_ones[j])-2*np.dot(sorted_f_ones[i],sorted_f_ones[j])) for i in tqdm.tqdm_notebook(range(250)) for j in range(250)]).reshape(250,250)



    


















    






























In [82]:



    


plt.figure(figsize=(15,15))
plt.imshow(euclid[0:250,0:250])
plt.colorbar()



    


















    
Out[82]:








<matplotlib.colorbar.Colorbar at 0x11a183790>










    
































In [85]:



    


len(s_all)



    


















    
Out[85]:








17311

























In [95]:



    


dist2=np.array([np.dot(f_ones_250[0]-f_ones_250[i],f_ones_250[0]-f_ones_250[i]) for i in range(250)])
df_ones["dist2"]=dist2
dt_sort=df_ones.sort("dist2").drop("dist2",axis=1)
dt_sort.set_index(np.arange(dt_sort.shape[0]),inplace=True)
sorted_f_ones=dt_sort['phi_ones'].values
euclid=np.array([np.sqrt(np.dot(sorted_f_ones[i],sorted_f_ones[i])+np.dot(sorted_f_ones[j],sorted_f_ones[j])-2*np.dot(sorted_f_ones[i],sorted_f_ones[j])) for i in tqdm.tqdm_notebook(range(250)) for j in range(250)]).reshape(250,250)
plt.figure(figsize=(12,12))
plt.imshow(euclid[0:250,0:250])
plt.colorbar()



    


















    






/usr/local/lib/python2.7/site-packages/ipykernel/__main__.py:3: FutureWarning: sort(columns=....) is deprecated, use sort_values(by=.....)
  app.launch_new_instance()










    















    
Out[95]:








<matplotlib.colorbar.Colorbar at 0x1346657d0>










    
































In [99]:



    


plt.figure(figsize=(10,10))
plt.imshow(euclid[0:100,0:100],interpolation='nearest')
plt.colorbar()



    


















    
Out[99]:








<matplotlib.colorbar.Colorbar at 0x135c9d690>










    
































In [96]:



    


dt_sort["Composition"][0:20]



    


















    
Out[96]:








0      (O, Nb, Ag)
1      (F, Ni, Ag)
2      (F, Mg, Ag)
3      (F, Zn, Ag)
4      (O, Nb, Ag)
5       (O, W, Ag)
6      (O, Te, Ag)
7      (Hf, F, Ag)
8      (S, Na, Ag)
9     (Se, Ho, Ag)
10     (O, Te, Ag)
11    (Se, Er, Ag)
12     (O, Ru, Ag)
13     (O, Nb, Ag)
14     (F, Sn, Ag)
15    (As, Nd, Ag)
16      (O, W, Ag)
17     (Ge, O, Ag)
18     (S, Bi, Ag)
19     (F, Ru, Ag)
Name: Composition, dtype: object

























In [102]:



    


df_Chi=pd.DataFrame({"phi_Chi":f_Chi_250})
dist2=np.array([np.dot(f_Chi_250[0]-f_Chi_250[i],f_Chi_250[0]-f_Chi_250[i]) for i in range(250)])
df_Chi["dist2"]=dist2
dt_sort=df_Chi.sort("dist2").drop("dist2",axis=1)
dt_sort.set_index(np.arange(dt_sort.shape[0]),inplace=True)
sorted_f_ones=dt_sort['phi_Chi'].values
euclid=np.array([np.sqrt(np.dot(sorted_f_ones[i],sorted_f_ones[i])+np.dot(sorted_f_ones[j],sorted_f_ones[j])-2*np.dot(sorted_f_ones[i],sorted_f_ones[j])) for i in tqdm.tqdm_notebook(range(250)) for j in range(250)]).reshape(250,250)
plt.figure(figsize=(12,12))
plt.imshow(euclid[0:250,0:250])
plt.colorbar()



    


















    






/usr/local/lib/python2.7/site-packages/ipykernel/__main__.py:4: FutureWarning: sort(columns=....) is deprecated, use sort_values(by=.....)










    















    
Out[102]:








<matplotlib.colorbar.Colorbar at 0x193847510>










    
































In [103]:



    


from sklearn.cluster import KMeans



    



















In [104]:



    


s250=s_all[0:2000]
f_ones_2000=[fp.get_phi(s250[i],obser="ones",rmax=12,delta=0.05) for i in tqdm.tqdm_notebook(range(len(s250)))]



    


















    






























In [111]:



    


inertia=np.zeros(50,dtype=float)
for i in tqdm.tqdm_notebook(range(50)):
    Km=KMeans(n_clusters=5+i)
    Km.fit_predict(f_ones_2000)
    inertia[i]=Km.inertia_
plt.plot(inertia)



    


















    















    
Out[111]:








[<matplotlib.lines.Line2D at 0x134f6c610>]










    
































In [108]:



    


comps=[s250[i].composition for i in range(2000)]
df_ones=pd.DataFrame({"phi_ones":f_ones_2000})
df_ones["cluster"]=clust
df_ones["Composition"]=comps
df_sorted=df_ones.sort("cluster")
sorted_f_ones=df_sorted['phi_ones'].values
euclid=np.array([np.sqrt(np.dot(sorted_f_ones[i],sorted_f_ones[i])+np.dot(sorted_f_ones[j],sorted_f_ones[j])-2*np.dot(sorted_f_ones[i],sorted_f_ones[j])) for i in tqdm.tqdm_notebook(range(2000)) for j in range(2000)]).reshape(2000,2000)
plt.figure(figsize=(12,12))
plt.imshow(euclid)
plt.colorbar()



    


















    






/usr/local/lib/python2.7/site-packages/ipykernel/__main__.py:5: FutureWarning: sort(columns=....) is deprecated, use sort_values(by=.....)










    















    
Out[108]:








<matplotlib.colorbar.Colorbar at 0x136da7750>










    
































In [136]:



    


Km=KMeans(n_clusters=12)
clust=Km.fit_predict(f_ones_2000)
comps=[s250[i].composition.formula for i in range(2000)]
df_ones=pd.DataFrame({"phi_ones":f_ones_2000})
df_ones["cluster"]=clust
df_ones["Composition"]=comps
df_sorted=df_ones.sort("cluster")
df_sorted.set_index(np.arange(2000))
sorted_f_ones=df_sorted['phi_ones'].values
euclid=np.array([np.sqrt(np.dot(sorted_f_ones[i],sorted_f_ones[i])+np.dot(sorted_f_ones[j],sorted_f_ones[j])-2*np.dot(sorted_f_ones[i],sorted_f_ones[j])) for i in tqdm.tqdm_notebook(range(2000)) for j in range(2000)]).reshape(2000,2000)
plt.figure(figsize=(12,12))
plt.imshow(euclid)
plt.colorbar()



    


















    






/usr/local/lib/python2.7/site-packages/ipykernel/__main__.py:7: FutureWarning: sort(columns=....) is deprecated, use sort_values(by=.....)










    















    
Out[136]:








<matplotlib.colorbar.Colorbar at 0x118fab490>










    
































In [139]:



    


df_sorted=df_sorted.set_index(np.arange(2000))
df_sorted



    


















    
Out[139]:










  
    

      
      phi_ones
      cluster
      Composition
    

  
  
    

      0
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      0
      Ag4 Cl4 O8
    

    

      1
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      0
      K4 Na2 As2
    

    

      2
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      0
      Ba32 Ge8 As32
    

    

      3
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      0
      Rb8 Ag4 Br12
    

    

      4
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      0
      Ba1 Cu2 As2
    

    

      5
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      0
      Sr2 Ag2 O4
    

    

      6
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      0
      K20 Sn4 As12
    

    

      7
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      0
      K3 Ag3 As2
    

    

      8
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      0
      K1 Zn1 As1
    

    

      9
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      0
      Tl12 Ag4 Te8
    

    

      10
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      0
      K14 Nb2 As8
    

    

      11
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      0
      Ba32 Ti8 As32
    

    

      12
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      0
      K4 Zn1 As2
    

    

      13
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      0
      Rb14 Nb2 As8
    

    

      14
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      0
      K14 Ta2 As8
    

    

      15
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      0
      K16 Ga16 As24
    

    

      16
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      0
      K12 In8 As12
    

    

      17
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      0
      K16 Ga8 As16
    

    

      18
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      0
      Ba1 Ag2 Ge2
    

    

      19
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      0
      Tl2 Ag2 Cl6
    

    

      20
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      0
      K4 Ag8 Se6
    

    

      21
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      0
      Rb4 Ag8 S6
    

    

      22
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      0
      Ag3 C6 N9
    

    

      23
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      0
      Sr2 Ag2 As2
    

    

      24
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      0
      Tl4 Ag12 S8
    

    

      25
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      0
      Eu2 Ag2 Sb2
    

    

      26
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      0
      Tl1 Ag1 Cl3
    

    

      27
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      0
      Tl4 Ag4 Se4
    

    

      28
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      0
      K4 Ag2 P2
    

    

      29
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      0
      K18 Nb4 As12
    

    

      ...
      ...
      ...
      ...
    

    

      1970
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      11
      Sr1 Zn2 As2
    

    

      1971
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      11
      Ba1 Al2 Si2
    

    

      1972
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      11
      Pr1 Al2 Si2
    

    

      1973
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      11
      Ba1 Zn2 As2
    

    

      1974
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      11
      In2 Ag2 S4
    

    

      1975
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      11
      Ti1 Co1 As1
    

    

      1976
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      11
      Al2 Ag2 Se4
    

    

      1977
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      11
      Mn1 Ni1 As1
    

    

      1978
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      11
      Cu2 As2 Se4
    

    

      1979
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      11
      Ca1 Cu4 As2
    

    

      1980
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      11
      Li22 Mn2 As12
    

    

      1981
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      11
      Li10 Si2 As6
    

    

      1982
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      11
      Li14 V2 As8
    

    

      1983
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      11
      Cd2 Ge2 As4
    

    

      1984
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      11
      Ca1 Zn2 As2
    

    

      1985
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      11
      Zr1 Ag1 B1
    

    

      1986
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      11
      Ta1 Mn1 As1
    

    

      1987
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      11
      Li2 Mn2 As2
    

    

      1988
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      11
      Eu1 Cu4 As2
    

    

      1989
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      11
      Gd1 Al2 Si2
    

    

      1990
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      11
      Li1 Al1 Si1
    

    

      1991
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      11
      Mn1 Nb1 As1
    

    

      1992
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      11
      Ce1 Al2 Ge2
    

    

      1993
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      11
      Li1 Zn1 As1
    

    

      1994
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      11
      Na2 Al2 Si2
    

    

      1995
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      11
      In4 Ag4 S8
    

    

      1996
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      11
      In5 Ag1 Se8
    

    

      1997
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      11
      In6 As2 Se6
    

    

      1998
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      11
      Ca3 As1 Br3
    

    

      1999
      [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1....
      11
      Ti1 As1 Rh1
    

  



2000 rows × 3 columns



























In [144]:



    


df_sorted[df_sorted["cluster"]==3]["Composition"]



    


















    
Out[144]:








739      La20 B20 C20
740        Sc1 Co3 B2
741         Fe2 B2 O6
742        U4 As2 Se2
743        Ho1 Co3 B2
744       Mg2 Cu2 As2
745       Yb2 Cu2 As4
746       Rb2 Ca2 As2
747        Sm4 As4 S4
748         Sm2 B4 C4
749       Mg8 Cu8 As8
750       Er2 Cu2 As4
751       Li6 Mn2 As4
752       Tm4 Ni2 As4
753        Al1 Mo1 O3
754        U2 As4 Pd2
755       Tm2 Cu2 As4
756        Eu1 As1 O3
757        Ce4 As4 S4
758       Ho4 Ni2 As4
759        Y2 Cu2 As4
760       Ge1 Te4 As2
761        Tm1 Co3 B2
762        Er1 Co3 B2
763         Cr2 B2 O6
764       Dy2 Cu2 As4
765        U2 Cu4 As5
766         Ca2 B4 C4
767        U2 Co2 As4
768        Yb1 Co3 B2
            ...      
839        Ce4 Ag4 P8
840     Tl4 Ag32 Te22
841        Tb5 Ag1 S8
842       Cs4 Ag8 I12
843        Sm2 Ag2 O5
844        Sr1 Ag1 O3
845       Tb2 Ag2 Se4
846      In6 Ag6 Te12
847       Ho2 Ag2 Se4
848        Pr5 Ag1 S8
849        Ag2 Bi2 S4
850       Ag2 Bi2 Se4
851       Ag9 Te4 Br3
852        Cu1 Ag1 F3
853       Ag3 Bi3 Se6
854        Al1 Ag1 O3
855         Al2 V8 C6
856       Nd2 Ag2 As4
857         Y1 Ag1 O3
858       Mg1 Al2 Si2
859        Y2 Ag2 Se4
860        Zn1 Ag1 F3
861        Tl2 Ag2 I4
862       Ce4 Ag4 As8
863       Nb8 Ag8 O24
864       Ag1 As1 Se2
865        Ag1 Bi1 S2
866    Ag40 Te16 Br12
867     Cs4 Ag20 Se12
868        Na2 Ag2 F8
Name: Composition, dtype: object

























In [ ]:



    






    
















    


    

        

            

            

                Content source: 3juholee/materialproject_ml
            

            
Similar notebooks:

            

            
                notebook.community |
                gallery |
                about