add CMTL least L21, doesnot work yet, set up test already

Author: DaPraxis

Vampyr_MTL/functions/tests/test_C_Least_L21.py

@@ -0,0 +1,67 @@
+from .test_data import get_data
+from ..MTL_Cluster_Least_L21 import MTL_Cluster_Least_L21
+from ...evaluations.utils import MTL_data_extract, MTL_data_split, opts
+import numpy as np
+import math
+
+# iris data
+X_train, X_test, Y_train, Y_test, df = get_data()
+opts = opts(1500,2)
+opts.tol = 10**(-6)
+
+
+# customized data
+clus_var = 900
+task_var = 16
+nois_var = 150
+
+clus_num = 2
+clus_task_num = 10
+task_num = clus_num * clus_task_num
+sample_size = 100
+dimension = 20
+comm_dim = 2
+clus_dim = math.floor((dimension - comm_dim)/2)
+
+# generate cluster model
+cluster_weight = np.random.randn(dimension, clus_num)* clus_var
+for i in range(clus_num):
+    bll = np.random.permutation(range(dimension-clus_num))<=clus_dim
+    blc = np.array([False]*clus_num)
+    bll = np.hstack((bll, blc))
+    cluster_weight[:,i][bll]=0
+cluster_weight[-1-comm_dim:, :]=0
+W = np.tile(cluster_weight, (1, clus_task_num))
+cluster_index = np.tile(range(clus_num), (1, clus_task_num)).T
+
+# generate task and intra-cluster variance
+W_it = np.random.randn(dimension, task_num) * task_var
+for i in range(task_num):
+    bll = np.hstack(((W[:-1-comm_dim+1,i]==0).reshape(1,-1), np.zeros((1,comm_dim))==1))
+    W_it[:,i][bll.flatten()]=0
+W = W+W_it
+
+W = W + np.random.randn(dimension, task_num)*nois_var
+
+X = [0]*task_num
+Y = [0]*task_num
+for i in range(task_num):
+    X[i] = np.random.randn(sample_size, dimension)
+    xw = X[i] @ W[:,i]
+    s= xw.shape
+    xw = xw + np.random.randn(s[0]) * nois_var
+    Y[i] = np.sign(xw)
+
+class Test_CMTL_Least_classification(object):
+    def test_basic_mat(self):
+        clf = MTL_Cluster_Least_L21(opts, 3)
+        clf.fit(X, Y)
+        corr = clf.analyse()
+        print(corr)
+        
+    
+    def test_iris_accuracy(self):
+        clf = MTL_Cluster_Least_L21(opts, 3)
+        clf.fit(X_train, Y_train)
+        corr = clf.analyse()
+        print(corr)

Vampyr_MTL/functions/tests/test_data.py

@@ -0,0 +1,22 @@
+import numpy as np
+import pandas as pd
+from sklearn import datasets
+from ...evaluations.utils import MTL_data_extract, MTL_data_split, opts
+
+
+def get_data():
+    iris = datasets.load_iris()
+    df = pd.DataFrame(data= np.c_[iris['data'], iris['target']],
+                        columns= iris['feature_names'] + ['target'])
+    df['cat1']=0
+    df['cat2']=0
+    df['target'] = df['target'].astype(int)
+    df.loc[df['petal width (cm)']<=0.8, 'cat1'] = 0
+    df.loc[(df['petal width (cm)']>0.8) & (df['petal width (cm)']<=1.6), 'cat1'] = 1
+    df.loc[(df['petal width (cm)']>1.6) & (df['petal width (cm)']<=2.4), 'cat1'] = 2
+    df.loc[df['petal length (cm)']<=2.3, 'cat2'] = 0
+    df.loc[(df['petal length (cm)']>2.3) & (df['petal length (cm)']<=4.6), 'cat2'] = 1
+    df.loc[(df['petal length (cm)']>4.6) & (df['petal length (cm)']<=6.9), 'cat2'] = 2
+    X_i, Y_i = MTL_data_extract(df, 'cat2', 'target')
+    X_train, X_test, Y_train, Y_test = MTL_data_split(X_i, Y_i, test_size=0.4)
+    return X_train, X_test, Y_train, Y_test, df

Vampyr_MTL/functions/tests/test_softmax_L21_hinge.py

@@ -3,6 +3,7 @@
 import pandas as pd
 from sklearn import datasets
 from ...evaluations.utils import MTL_data_extract, MTL_data_split, opts
+from .test_data import get_data
 
 # class opts:
 # 	def __init__(self, maxIter, init):
@@ -12,21 +13,7 @@
 
 opts = opts(1000,2)
 
-## iris data
-iris = datasets.load_iris()
-df = pd.DataFrame(data= np.c_[iris['data'], iris['target']],
-                     columns= iris['feature_names'] + ['target'])
-df['cat1']=0
-df['cat2']=0
-df['target'] = df['target'].astype(int)
-df.loc[df['petal width (cm)']<=0.8, 'cat1'] = 0
-df.loc[(df['petal width (cm)']>0.8) & (df['petal width (cm)']<=1.6), 'cat1'] = 1
-df.loc[(df['petal width (cm)']>1.6) & (df['petal width (cm)']<=2.4), 'cat1'] = 2
-df.loc[df['petal length (cm)']<=2.3, 'cat2'] = 0
-df.loc[(df['petal length (cm)']>2.3) & (df['petal length (cm)']<=4.6), 'cat2'] = 1
-df.loc[(df['petal length (cm)']>4.6) & (df['petal length (cm)']<=6.9), 'cat2'] = 2
-X_i, Y_i = MTL_data_extract(df, 'cat2', 'target')
-X_train, X_test, Y_train, Y_test = MTL_data_split(X_i, Y_i, test_size=0.4)
+X_train, X_test, Y_train, Y_test, df = get_data()
 
 df2 = df.copy()
 df2.loc[df['target']==0, 'target'] = 'flower1'