Quillot Mathias / Clustering

Download as
Browse Code »

Commit ca0fcf2c3f1b1faa7cdef49e229d135c891176c6

Authored by Mathias Quillot
1 parent 0b30718782
Exists in master

Adapting the script to my library. Need to work again

Showing 1 changed file with 17 additions and 12 deletions Side-by-side Diff

bin/plot_clusters.py
Diff comments   View file @ ca0fcf2
... ... @@ -55,8 +55,8 @@
55 55  
56 56  
57 57  
  58 +from data import read_file, index_by_id
58 59  
59   -
60 60 # Defining argparse
61 61 parser = argparse.ArgumentParser(prog='Plotter', description='Plot a file of 2d ou 3d dimension')
62 62 parser.add_argument('clusterfile', type=str,
... ... @@ -66,8 +66,6 @@
66 66 parser.add_argument('-o-', '--output', type=str,
67 67 default='plot.pdf',
68 68 help='the path of the ploted file')
69   -parser.add_argument('-t', '--toy', action='store_true',
70   - help='test the script on a toy example. Do not test all the file content')
71 69  
72 70 args = parser.parse_args()
73 71  
74 72  
75 73  
... ... @@ -75,12 +73,24 @@
75 73 CLUSTERFILE_PATH=args.clusterfile
76 74 VECTORFILE_PATH=args.vectorfile
77 75 OUTFILE_PATH = args.output
78   -TOY_VERSION = args.toy
79 76  
80   -if check_files(VECTORFILE_PATH, CLUSTERFILE_PATH) == False:
81   - print("Les fichiers ne sont pas dans le meme ordre. Dans une version futur, cela générera une exception. On stop le processus.")
82   - exit(1)
  77 +data_vector = read_file(VECTORFILE_PATH)
  78 +features = np.asarray([x[1] for x in data_vector])
  79 +features_T = np.transpose(features)
83 80  
  81 +data_cluster = read_file(CLUSTERFILE_PATH)
  82 +data_cluster_ind = index_by_id(data_cluster)
  83 +clusters = [data_cluster_ind[x[0][0]][x[0][3]][0][1] for x in data_vector]
  84 +
  85 +
  86 +# TODO: compute tsne file
  87 +# TODO: adapt the script for the new library
  88 +df = pd.DataFrame(dict(
  89 + x=features_T[0],
  90 + y=features_T[1],
  91 + cluster=np.transpose(clusters)[0]
  92 + ))
  93 +exit(1)
84 94 # Get Vectors
85 95 metas, vectors = read_vector_file(VECTORFILE_PATH, toy_version = TOY_VERSION)
86 96 vectors_T = np.transpose(vectors)
... ... @@ -90,11 +100,6 @@
90 100  
91 101 #print(np.transpose(clusters)[0])
92 102 #print(np.transpose(metas)[0])
93   -df = pd.DataFrame(dict(
94   - x=vectors_T[0],
95   - y=vectors_T[1],
96   - cluster=np.transpose(clusters)[0]
97   - ))
98 103  
99 104 groups = df.groupby('cluster')
100 105