Mercurial > repos > bgruening > nn_classifier

comparison nn_classifier.xml @ 18:c64f57fe1b97 draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/tools/sklearn commit 5d71c93a3dd804b1469852240a86021ab9130364
author bgruening
date Mon, 09 Jul 2018 14:26:44 -0400
parents 25a68adb2ade
children fa36c40c2990
comparison
equal deleted inserted replaced
17:9e0d360e54ea 18:c64f57fe1b97
19 import numpy as np 19 import numpy as np
20 import sklearn.neighbors 20 import sklearn.neighbors
21 import pandas 21 import pandas
22 import pickle 22 import pickle
23 23
24 @COLUMNS_FUNCTION@
25 @GET_X_y_FUNCTION@
26
24 input_json_path = sys.argv[1] 27 input_json_path = sys.argv[1]
25 params = json.load(open(input_json_path, "r")) 28 params = json.load(open(input_json_path, "r"))
26 29
27 30
28 #if $selected_tasks.selected_task == "load": 31 #if $selected_tasks.selected_task == "load":
29 32
30 classifier_object = pickle.load(open("$infile_model", 'r')) 33 classifier_object = pickle.load(open("$infile_model", 'r'))
31 34
32 data = pandas.read_csv("$selected_tasks.infile_data", sep='\t', header=0, index_col=None, parse_dates=True, encoding=None, tupleize_cols=False ) 35 header = 'infer' if params["selected_tasks"]["header"] else None
36 data = pandas.read_csv("$selected_tasks.infile_data", sep='\t', header=header, index_col=None, parse_dates=True, encoding=None, tupleize_cols=False)
33 prediction = classifier_object.predict(data) 37 prediction = classifier_object.predict(data)
34 prediction_df = pandas.DataFrame(prediction) 38 prediction_df = pandas.DataFrame(prediction)
35 res = pandas.concat([data, prediction_df], axis=1) 39 res = pandas.concat([data, prediction_df], axis=1)
36 res.to_csv(path_or_buf = "$outfile_predict", sep="\t", index=False) 40 res.to_csv(path_or_buf = "$outfile_predict", sep="\t", index=False)
37 41
38 #else: 42 #else:
39 43
40 data_train = pandas.read_csv("$selected_tasks.infile_train", sep='\t', header=0, index_col=None, parse_dates=True, encoding=None, tupleize_cols=False ) 44 X, y = get_X_y(params, "$selected_tasks.selected_algorithms.input_options.infile1" ,"$selected_tasks.selected_algorithms.input_options.infile2")
41
42 data = data_train.ix[:,0:len(data_train.columns)-1]
43 labels = np.array(data_train[data_train.columns[len(data_train.columns)-1]])
44 45
45 selected_algorithm = params["selected_tasks"]["selected_algorithms"]["selected_algorithm"] 46 selected_algorithm = params["selected_tasks"]["selected_algorithms"]["selected_algorithm"]
46 47
47 if selected_algorithm == "nneighbors": 48 if selected_algorithm == "nneighbors":
48 classifier = params["selected_tasks"]["selected_algorithms"]["sampling_methods"]["sampling_method"] 49 classifier = params["selected_tasks"]["selected_algorithms"]["sampling_methods"]["sampling_method"]
53 options = params["selected_tasks"]["selected_algorithms"]["options"] 54 options = params["selected_tasks"]["selected_algorithms"]["options"]
54 classifier = "NearestCentroid" 55 classifier = "NearestCentroid"
55 56
56 my_class = getattr(sklearn.neighbors, classifier) 57 my_class = getattr(sklearn.neighbors, classifier)
57 classifier_object = my_class(**options) 58 classifier_object = my_class(**options)
58 classifier_object.fit(data,labels) 59 classifier_object.fit(X, y)
59 60
60 pickle.dump(classifier_object,open("$outfile_fit", 'w+')) 61 pickle.dump(classifier_object,open("$outfile_fit", 'w+'))
61 62
62 #end if 63 #end if
63 64
64 ]]> 65 ]]>
65 </configfile> 66 </configfile>
66 </configfiles> 67 </configfiles>
67 <inputs> 68 <inputs>
68 <expand macro="train_loadConditional" model="zip"><!--Todo: add sparse to targets--> 69 <expand macro="sl_Conditional" model="zip"><!--Todo: add sparse to targets-->
69 <param name="selected_algorithm" type="select" label="Classifier type"> 70 <param name="selected_algorithm" type="select" label="Classifier type">
70 <option value="nneighbors">Nearest Neighbors</option> 71 <option value="nneighbors">Nearest Neighbors</option>
71 <option value="ncentroid">Nearest Centroid</option> 72 <option value="ncentroid">Nearest Centroid</option>
72 </param> 73 </param>
73 <when value="nneighbors"> 74 <when value="nneighbors">
75 <expand macro="sl_mixed_input"/>
74 <conditional name="sampling_methods"> 76 <conditional name="sampling_methods">
75 <param name="sampling_method" type="select" label="Neighbor selection method"> 77 <param name="sampling_method" type="select" label="Neighbor selection method">
76 <option value="KNeighborsClassifier" selected="true">K-nearest neighbors</option> 78 <option value="KNeighborsClassifier" selected="true">K-nearest neighbors</option>
77 <option value="RadiusNeighborsClassifier">Radius-based</option> 79 <option value="RadiusNeighborsClassifier">Radius-based</option>
78 </param> 80 </param>
88 </expand> 90 </expand>
89 </when> 91 </when>
90 </conditional> 92 </conditional>
91 </when> 93 </when>
92 <when value="ncentroid"> 94 <when value="ncentroid">
95 <expand macro="sl_mixed_input"/>
93 <section name="options" title="Advanced Options" expanded="False"> 96 <section name="options" title="Advanced Options" expanded="False">
94 <param argument="metric" type="text" optional="true" value="euclidean" label="Metric" 97 <param argument="metric" type="text" optional="true" value="euclidean" label="Metric"
95 help="The metric to use when calculating distance between instances in a feature array."/> 98 help="The metric to use when calculating distance between instances in a feature array."/>
96 <param argument="shrink_threshold" type="float" optional="true" value="" label="Shrink threshold" 99 <param argument="shrink_threshold" type="float" optional="true" value="" label="Shrink threshold"
97 help="Floating point number for shrinking centroids to remove features."/> 100 help="Floating point number for shrinking centroids to remove features."/>
102 105
103 <expand macro="output"/> 106 <expand macro="output"/>
104 107
105 <tests> 108 <tests>
106 <test> 109 <test>
107 <param name="infile_train" value="train_set.tabular" ftype="tabular"/> 110 <param name="infile1" value="train_set.tabular" ftype="tabular"/>
111 <param name="infile2" value="train_set.tabular" ftype="tabular"/>
112 <param name="header1" value="True"/>
113 <param name="header2" value="True"/>
114 <param name="col1" value="1,2,3,4"/>
115 <param name="col2" value="5"/>
108 <param name="selected_task" value="train"/> 116 <param name="selected_task" value="train"/>
109 <param name="selected_algorithm" value="nneighbors"/> 117 <param name="selected_algorithm" value="nneighbors"/>
110 <param name="sampling_method" value="KNeighborsClassifier" /> 118 <param name="sampling_method" value="KNeighborsClassifier" />
111 <param name="algorithm" value="brute" /> 119 <param name="algorithm" value="brute" />
112 <output name="outfile_fit" file="nn_model01.txt"/> 120 <output name="outfile_fit" file="nn_model01.txt"/>
113 </test> 121 </test>
114 <test> 122 <test>
115 <param name="infile_train" value="train_set.tabular" ftype="tabular"/> 123 <param name="infile1" value="train_set.tabular" ftype="tabular"/>
124 <param name="infile2" value="train_set.tabular" ftype="tabular"/>
125 <param name="header1" value="True"/>
126 <param name="header2" value="True"/>
127 <param name="col1" value="1,2,3,4"/>
128 <param name="col2" value="5"/>
116 <param name="selected_task" value="train"/> 129 <param name="selected_task" value="train"/>
117 <param name="selected_algorithm" value=""/> 130 <param name="selected_algorithm" value=""/>
118 <param name="selected_algorithm" value="nneighbors"/> 131 <param name="selected_algorithm" value="nneighbors"/>
119 <param name="sampling_method" value="RadiusNeighborsClassifier" /> 132 <param name="sampling_method" value="RadiusNeighborsClassifier" />
120 <output name="outfile_fit" file="nn_model02.txt"/> 133 <output name="outfile_fit" file="nn_model02.txt"/>
121 </test> 134 </test>
122 <test> 135 <test>
123 <param name="infile_train" value="train_set.tabular" ftype="tabular"/> 136 <param name="infile1" value="train_set.tabular" ftype="tabular"/>
137 <param name="infile2" value="train_set.tabular" ftype="tabular"/>
138 <param name="header1" value="True"/>
139 <param name="header2" value="True"/>
140 <param name="col1" value="1,2,3,4"/>
141 <param name="col2" value="5"/>
124 <param name="selected_task" value="train"/> 142 <param name="selected_task" value="train"/>
125 <param name="selected_algorithm" value="ncentroid"/> 143 <param name="selected_algorithm" value="ncentroid"/>
126 <output name="outfile_fit" file="nn_model03.txt"/> 144 <output name="outfile_fit" file="nn_model03.txt"/>
127 </test> 145 </test>
128 <test> 146 <test>
129 <param name="infile_model" value="nn_model01.txt" ftype="txt"/> 147 <param name="infile_model" value="nn_model01.txt" ftype="txt"/>
130 <param name="infile_data" value="test_set.tabular" ftype="tabular"/> 148 <param name="infile_data" value="test_set.tabular" ftype="tabular"/>
149 <param name="header" value="True"/>
131 <param name="selected_task" value="load"/> 150 <param name="selected_task" value="load"/>
132 <output name="outfile_predict" file="nn_prediction_result01.tabular"/> 151 <output name="outfile_predict" file="nn_prediction_result01.tabular"/>
133 </test> 152 </test>
134 <test> 153 <test>
135 <param name="infile_model" value="nn_model02.txt" ftype="txt"/> 154 <param name="infile_model" value="nn_model02.txt" ftype="txt"/>
136 <param name="infile_data" value="test_set.tabular" ftype="tabular"/> 155 <param name="infile_data" value="test_set.tabular" ftype="tabular"/>
156 <param name="header" value="True"/>
137 <param name="selected_task" value="load"/> 157 <param name="selected_task" value="load"/>
138 <output name="outfile_predict" file="nn_prediction_result02.tabular"/> 158 <output name="outfile_predict" file="nn_prediction_result02.tabular"/>
139 </test> 159 </test>
140 <test> 160 <test>
141 <param name="infile_model" value="nn_model03.txt" ftype="txt"/> 161 <param name="infile_model" value="nn_model03.txt" ftype="txt"/>
142 <param name="infile_data" value="test_set.tabular" ftype="tabular"/> 162 <param name="infile_data" value="test_set.tabular" ftype="tabular"/>
163 <param name="header" value="True"/>
143 <param name="selected_task" value="load"/> 164 <param name="selected_task" value="load"/>
144 <output name="outfile_predict" file="nn_prediction_result03.tabular"/> 165 <output name="outfile_predict" file="nn_prediction_result03.tabular"/>
145 </test> 166 </test>
146 </tests> 167 </tests>
147 <help><![CDATA[ 168 <help><![CDATA[