net.new_node_covariate_int('r')[:] = 1 net.new_node_covariate_int('c')[:] = 1 data_model = FixedMargins(data_model, 'r', 'c') coverage_levels = np.append(0.0, np.cumsum(params['coverage_increments'])) traces = { 'wall_time': [], 'nll': [] } for rep in range(params['num_reps']): net.generate(data_model, arbitrary_init = params['arb_init']) wall_time_trace = [net.gen_info['wall_time']] nll_trace = [data_model.nll(net)] for coverage_inc in params['coverage_increments']: data_model.gibbs_improve_perm(net, net.as_dense(), coverage_inc) wall_time_trace.append(net.gen_info['wall_time']) nll_trace.append(data_model.nll(net)) traces['wall_time'].append(wall_time_trace) traces['nll'].append(nll_trace) plt.figure() plt.title('Computation time') plt.xlabel('Coverage level') plt.ylabel('Wall time (msec)') for rep in range(params['num_reps']): plt.plot(coverage_levels, traces['wall_time'][rep], '-') plt.figure()
net.new_node_covariate_int('r')[:] = 1 net.new_node_covariate_int('c')[:] = 1 data_model = FixedMargins(data_model, 'r', 'c') coverage_levels = np.append(0.0, np.cumsum(params['coverage_increments'])) traces = { 'wall_time': [], 'nll': [] } for rep in range(params['num_reps']): net.generate(data_model, arbitrary_init = params['arb_init']) wall_time_trace = [net.gen_info['wall_time']] nll_trace = [data_model.nll(net)] for coverage_inc in params['coverage_increments']: data_model.gibbs_improve_perm(net, net.adjacency_matrix(), coverage_inc) wall_time_trace.append(net.gen_info['wall_time']) nll_trace.append(data_model.nll(net)) traces['wall_time'].append(wall_time_trace) traces['nll'].append(nll_trace) plt.figure() plt.title('Computation time') plt.xlabel('Coverage level') plt.ylabel('Wall time (msec)') for rep in range(params['num_reps']): plt.plot(coverage_levels, traces['wall_time'][rep], '-') plt.figure()
# Specify data model as generation permuation networks net.new_node_covariate_int('r')[:] = 1 net.new_node_covariate_int('c')[:] = 1 data_model = FixedMargins(data_model, 'r', 'c') coverage_levels = np.append(0.0, np.cumsum(params['coverage_increments'])) traces = {'wall_time': [], 'nll': []} for rep in range(params['num_reps']): net.generate(data_model, arbitrary_init=params['arb_init']) wall_time_trace = [net.gen_info['wall_time']] nll_trace = [data_model.nll(net)] for coverage_inc in params['coverage_increments']: data_model.gibbs_improve_perm(net, net.as_dense(), coverage_inc) wall_time_trace.append(net.gen_info['wall_time']) nll_trace.append(data_model.nll(net)) traces['wall_time'].append(wall_time_trace) traces['nll'].append(nll_trace) plt.figure() plt.title('Computation time') plt.xlabel('Coverage level') plt.ylabel('Wall time (msec)') for rep in range(params['num_reps']): plt.plot(coverage_levels, traces['wall_time'][rep], '-') plt.figure()