1+ {
2+ "worksheets" : [
3+ {
4+ "cells" : [
5+ {
6+ "cell_type" : " markdown" ,
7+ "source" : " This downloads a classification task from OpenML, uses sklearn to build a RandomForest classifier, and uploads the results to the server.\n "
8+ },
9+ {
10+ "cell_type" : " markdown" ,
11+ "source" : " Initialization and login. This assumes you have a .openml dir in your homedir with a subdir for caches and a file with your API key. You find your API key in your account settings on openml.org."
12+ },
13+ {
14+ "cell_type" : " code" ,
15+ "input" : " from sklearn import preprocessing, ensemble\n import openml\n import os\n\n home_dir = os.path.expanduser(\" ~\" )\n openml_dir = os.path.join(home_dir, \" .openml\" )\n cache_dir = os.path.join(openml_dir, \" cache\" )\n with open(os.path.join(openml_dir, \" apikey.txt\" ), 'r') as fh:\n\t key = fh.readline().rstrip('\\ n')\n\n openml = APIConnector(cache_directory=cache_dir, apikey=key)" ,
16+ "outputs" : [],
17+ "language" : " python" ,
18+ "metadata" : {
19+ "cellView" : null ,
20+ "executionInfo" : {
21+ "content" : {
22+ "status" : " ok" ,
23+ "execution_count" : 28 ,
24+ "payload" : [],
25+ "user_expressions" : {},
26+ "user_variables" : {}
27+ },
28+ "timestamp" : 1455279958823 ,
29+ "user_tz" : -60 ,
30+ "user" : {
31+ "sessionId" : " 284b6c114ba378fe" ,
32+ "userId" : " 109221074076178034989" ,
33+ "permissionId" : " 09574386662628411592" ,
34+ "displayName" : " Joaquin Vanschoren" ,
35+ "color" : " #1FA15D" ,
36+ "isMe" : true ,
37+ "isAnonymous" : false ,
38+ "photoUrl" : " //lh6.googleusercontent.com/-9g2BD5NUkx4/AAAAAAAAAAI/AAAAAAAACC4/sACE-8zBzX4/s50-c-k-no/photo.jpg"
39+ }
40+ }
41+ }
42+ },
43+ {
44+ "cell_type" : " markdown" ,
45+ "source" : " Download a single OpenML task (id=10), create a scikit-learn classifier (RandomForest), and run it on the task"
46+ },
47+ {
48+ "cell_type" : " code" ,
49+ "input" : " task = openml.download_task(10)\n print task\n clf = ensemble.RandomForestClassifier()\n X, y = task.get_X_and_Y()\n clf.fit(X, y)\n print clf\n prediction_path, description_path = openml_run(task, clf)\n\n #import json\n #print(json.dumps(xmltodict.parse(open(os.path.abspath(description_path), \" r\" ).read()), indent=4))" ,
50+ "outputs" : [
51+ {
52+ "output_type" : " stream" ,
53+ "text" : " OpenMLTask instance.\n Task ID: 10\n Task type: Supervised Classification\n Dataset id: 10\n RandomForestClassifier(bootstrap=True, class_weight=None, criterion='gini',\n max_depth=None, max_features='auto', max_leaf_nodes=None,\n min_samples_leaf=1, min_samples_split=2,\n min_weight_fraction_leaf=0.0, n_estimators=10, n_jobs=1,\n oob_score=False, random_state=None, verbose=0,\n warm_start=False)\n 2638\n " ,
54+ "stream" : " stdout"
55+ }
56+ ],
57+ "language" : " python" ,
58+ "metadata" : {
59+ "cellView" : null ,
60+ "executionInfo" : {
61+ "content" : {
62+ "status" : " ok" ,
63+ "execution_count" : 35 ,
64+ "payload" : [],
65+ "user_expressions" : {},
66+ "user_variables" : {}
67+ },
68+ "timestamp" : 1455280601582 ,
69+ "user_tz" : -60 ,
70+ "user" : {
71+ "sessionId" : " 284b6c114ba378fe" ,
72+ "userId" : " 109221074076178034989" ,
73+ "permissionId" : " 09574386662628411592" ,
74+ "displayName" : " Joaquin Vanschoren" ,
75+ "color" : " #1FA15D" ,
76+ "isMe" : true ,
77+ "isAnonymous" : false ,
78+ "photoUrl" : " //lh6.googleusercontent.com/-9g2BD5NUkx4/AAAAAAAAAAI/AAAAAAAACC4/sACE-8zBzX4/s50-c-k-no/photo.jpg"
79+ }
80+ }
81+ }
82+ },
83+ {
84+ "cell_type" : " markdown" ,
85+ "source" : " Upload the run to the OpenML server"
86+ },
87+ {
88+ "cell_type" : " code" ,
89+ "input" : " import xmltodict\n\n prediction_abspath = os.path.abspath(prediction_path)\n description_abspath = os.path.abspath(description_path)\n\n return_code, response = openml.upload_run(prediction_abspath, description_abspath)\n\n if(return_code == 200):\n\t response_dict = xmltodict.parse(response.content)\n\t run_id = response_dict['oml:upload_run']['oml:run_id']\n\t print(\" Uploaded run with id %s\" % (run_id))\n\t print(\" Check it at www.openml.org/r/%s\" % (run_id))\n " ,
90+ "outputs" : [
91+ {
92+ "output_type" : " stream" ,
93+ "text" : " Uploaded run with id 524013\n Check it at www.openml.org/r/524013\n " ,
94+ "stream" : " stdout"
95+ }
96+ ],
97+ "language" : " python" ,
98+ "metadata" : {
99+ "cellView" : null ,
100+ "executionInfo" : {
101+ "content" : {
102+ "status" : " ok" ,
103+ "execution_count" : 31 ,
104+ "payload" : [],
105+ "user_expressions" : {},
106+ "user_variables" : {}
107+ },
108+ "timestamp" : 1455279991565 ,
109+ "user_tz" : -60 ,
110+ "user" : {
111+ "sessionId" : " 284b6c114ba378fe" ,
112+ "userId" : " 109221074076178034989" ,
113+ "permissionId" : " 09574386662628411592" ,
114+ "displayName" : " Joaquin Vanschoren" ,
115+ "color" : " #1FA15D" ,
116+ "isMe" : true ,
117+ "isAnonymous" : false ,
118+ "photoUrl" : " //lh6.googleusercontent.com/-9g2BD5NUkx4/AAAAAAAAAAI/AAAAAAAACC4/sACE-8zBzX4/s50-c-k-no/photo.jpg"
119+ }
120+ }
121+ }
122+ }
123+ ]
124+ }
125+ ],
126+ "metadata" : {
127+ "name" : " OpenMLDemo.ipynb" ,
128+ "colabVersion" : " 0.1"
129+ },
130+ "nbformat" : 3 ,
131+ "nbformat_minor" : 0
132+ }
0 commit comments