def main(): # These ranges likely contain the global maximum. Find it. center = [50, 100, 150, 200, 250] spread = np.arange(50, 1001, 50) params = {'centers': center, 'spread': spread} run_optimization(get_rbf_nn_prediction, params, 'optimal_rbf.shelf', 'RBF')
def main(): alphas = list(np.arange(0.01, 1.01, 0.01)) params = {'alpha': alphas} run_optimization(get_lasso_prediction, params, 'optimal_lasso.shelf', 'LASSO', backend=PARALLEL_BACKEND)
def main(): alphas = list(np.arange(0.05, 1.00, 0.05)) l1_ratios = list(np.arange(0.05, 1.00, 0.05)) params = {'alpha': alphas, 'l1_ratio': l1_ratios} run_optimization(get_en_prediction, params, 'optimal_en.shelf', 'EN', backend=PARALLEL_BACKEND)
def main(): nums = (1e10, 1e5, 1e1, 0, -1e1, -1e5, -1e10) alpha_1 = nums alpha_2 = nums lambda_1 = nums lambda_2 = nums params = { 'alpha_1': alpha_1, 'alpha_2': alpha_2 , 'lambda_1': lambda_1, 'lambda_2': lambda_2 } run_optimization(get_brr_prediction, params, 'optimal_brr.shelf', 'BRR', backend=PARALLEL_BACKEND)
def main(): params = { 'hidden': HIDDEN , 'epochs' : (EPOCHS,) } backend = PARALLEL_BACKEND if get_is_on_gpu() or get_is_time_stats() or get_should_plot(): backend = SINGLE_CORE_BACKEND run_optimization( get_net_prediction, params, 'optimal_nn.shelf', 'N' , sample_size_multiplier=DOUBLE_MULTIPLIER , backend=backend, retry_nans=True)
def main(): params = { 'hidden': HIDDEN , 'dropout_prob': DROPOUT , 'weight_decay': WEIGHT_DECAY , 'epochs': (EPOCHS,) } backend = PARALLEL_BACKEND if get_is_on_gpu() or get_is_time_stats() or get_should_plot(): backend = SINGLE_CORE_BACKEND run_optimization(get_net_prediction, params, 'optimal_wddonn.shelf', 'NWDDO', backend=backend, retry_nans=True)
def main(): nums = (1e10, 1e5, 1e1, 0, -1e1, -1e5, -1e10) alpha_1 = nums alpha_2 = nums lambda_1 = nums lambda_2 = nums params = { 'alpha_1': alpha_1, 'alpha_2': alpha_2, 'lambda_1': lambda_1, 'lambda_2': lambda_2 } run_optimization(get_brr_prediction, params, 'optimal_brr.shelf', 'BRR', backend=PARALLEL_BACKEND)
def main(): # alphas = list(np.logspace(-8., 8., base=10, num=17)) # Wide search alphas = list(np.linspace(10, 1000, num=100)) # Narrow Search params = {'alpha': alphas} run_optimization(get_rr_prediction, params, 'optimal_rr.shelf', 'RR') # Already parallel.
def main(): params = {} run_optimization(get_lr_prediction, params, 'optimal_lr.shelf', 'OLS')
def main(): alphas = list(np.arange(0.01, 1.01,0.01)) params = {'alpha': alphas} run_optimization( get_lasso_prediction, params , 'optimal_lasso.shelf', 'LASSO', backend=PARALLEL_BACKEND)