diff options
Diffstat (limited to 'src/convex_optimization.py')
| -rw-r--r-- | src/convex_optimization.py | 21 |
1 files changed, 11 insertions, 10 deletions
diff --git a/src/convex_optimization.py b/src/convex_optimization.py index 6a43705..1f34508 100644 --- a/src/convex_optimization.py +++ b/src/convex_optimization.py @@ -6,7 +6,7 @@ import timeout import cvxopt -@timeout.timeout(10) +@timeout.timeout(20) def sparse_recovery(M_val, w_val, lbda): """ Solves: @@ -24,7 +24,7 @@ def sparse_recovery(M_val, w_val, lbda): if M_val.dtype == bool: M_val = M_val.astype('float32') - if type(lbda) == int: + if type(lbda) == int or type(lbda) == float: lbda = np.array(lbda) m, n = M_val.shape @@ -37,8 +37,8 @@ def sparse_recovery(M_val, w_val, lbda): lbda = theano.shared(lbda.astype(theano.config.floatX)) y = lbda * theta_.norm(1) - 1./m*( - tensor.dot(1-w, tensor.log(1-tensor.exp(M.dot(theta_)))) - + tensor.dot(w, tensor.dot(M, theta_)) + (w).dot(tensor.log(1-tensor.exp(M.dot(theta_)))) + + (1-w).dot(tensor.dot(M, theta_)) ) return diff_and_opt(theta, theta_, M, M_val, w, lbda, y) @@ -56,7 +56,7 @@ def type_lasso(M_val, w_val, lbda): if M_val.dtype == bool: M_val = M_val.astype('float32') - if type(lbda) == int: + if type(lbda) == int or type(lbda) == float: lbda = np.array(lbda) theta = tensor.row().T @@ -104,11 +104,11 @@ def diff_and_opt(theta, theta_, M, M_val, w, lbda, y): #cvxopt.solvers.options['feastol'] = 2e-5 #cvxopt.solvers.options['abstol'] = 2e-5 #cvxopt.solvers.options['maxiters'] = 100 - cvxopt.solvers.options['show_progress'] = True + cvxopt.solvers.options['show_progress'] = False try: theta = cvxopt.solvers.cp(F, G, h)['x'] except ArithmeticError: - print("ArithmeticError thrown, change initial point"+\ + print("ArithmeticError thrown, change initial point" " given to the solver") if cvxopt.solvers.options['show_progress']: @@ -121,9 +121,9 @@ def test(): """ unit test """ - lbda = 1 + lbda = .001 G = cascade_creation.InfluenceGraph(max_proba=.9) - G.erdos_init(n=10, p = .3) + G.erdos_init(n=100, p = .3) A = cascade_creation.generate_cascades(G, .1, 1000) M_val, w_val = cascade_creation.icc_matrixvector_for_node(A, 0) @@ -131,12 +131,13 @@ def test(): if 0: p_vec, theta = type_lasso(M_val, w_val, lbda) print(p_vec) + print(G.mat) #Sparse recovery if 1: p_vec, theta = sparse_recovery(M_val, w_val, lbda) - print(G.mat[0]) print(p_vec) + print(G.mat) if __name__=="__main__": test() |