def parse_args(args):
parser = argparse.ArgumentParser("""
Plot for the simulation convergence check: what if num features grows
""")
parser.add_argument('--result-folder',
type=str,
default="simulation_convergence_num_p")
parser.add_argument(
'--spinn-template',
type=str,
default="_output/seed_%d/num_train_200/num_p_%d/fitted_spinn.pkl")
parser.add_argument(
'--file-template',
type=str,
default="_output/seed_%d/num_train_200/num_p_%d/fitted_spinn.csv")
parser.add_argument('--num-ps',
type=str,
default=make_params([25, 50, 100, 200, 400]))
parser.add_argument('--lasso-ratio', type=float, default=0.1)
parser.add_argument('--max-relevant-idx', type=int, default=6)
parser.add_argument('--seeds', type=str, default=make_params(range(1, 21)))
parser.add_argument('--out-plot',
type=str,
default="_output/plot_simulation_mse.png")
parser.add_argument('--out-weight-plot',
type=str,
default="_output/plot_simulation_weights.png")
parser.set_defaults()
args = parser.parse_args(args)
args.seeds = process_params(args.seeds, int)
args.num_ps = process_params(args.num_ps, int)
return args
def parse_args(args):
''' parse command line arguments '''
parser = argparse.ArgumentParser(description=__doc__)
parser.add_argument('--seed', type=int, default=0)
parser.add_argument('--pi-alpha',
type=float,
default=0.1,
help="(1-pi_alpha) prediction interval")
parser.add_argument('--num-trains',
type=str,
default="1",
help="comma separated")
parser.add_argument('--coverage-files',
type=str,
default="_output/agg_coverages.pkl",
help="comma separated")
parser.add_argument('--plot-file',
type=str,
default="_output/coverage_vs_num_train.png")
parser.set_defaults()
args = parser.parse_args()
args.coverage_files = process_params(args.coverage_files, str)
args.num_trains = process_params(args.num_trains, int)
return args
def parse_args(args):
''' parse command line arguments '''
parser = argparse.ArgumentParser(description=__doc__)
parser.add_argument('--seed', type=int, default=0)
parser.add_argument('--data-file', type=str, default="_output/data.pkl")
parser.add_argument('--num-rand', type=int, default=20)
parser.add_argument('--num-test', type=int, default=6000)
parser.add_argument('--ci-alpha', type=float, default=0.1)
parser.add_argument('--fitted-files',
type=str,
default="_output/fitted.pkl")
parser.add_argument('--coverage-files',
type=str,
default="_output/recalibrated_coverages.pkl")
parser.add_argument('--out-file',
type=str,
default="_output/aggregate_coverages.pkl")
parser.add_argument('--log-file', type=str, default="_output/log_agg.txt")
parser.set_defaults()
args = parser.parse_args()
args.fitted_files = process_params(args.fitted_files, str)
args.coverage_files = process_params(args.coverage_files, str)
return args
def parse_args(args):
parser = argparse.ArgumentParser("Plot for the main simulations")
parser.add_argument(
'--result-folder',
type=str,
default="simulation_alpha")
parser.add_argument(
'--lasso',
type=str,
default=make_params(np.arange(-.5, -6, step=-.5)))
parser.add_argument(
'--group-lasso',
type=str,
default=make_params(np.arange(-.5, -6, step=-.5)))
parser.add_argument(
'--max-relevant-idx',
type=int,
default=6)
parser.add_argument(
'--seeds',
type=str,
default="4,5,6,7,8,9,10,11")
parser.add_argument(
'--file-template',
type=str,
default="_output/seed_%d/group_lasso_%.2f/lasso_%.2f/fitted_spinn.%s")
parser.add_argument(
'--out-mse-plot',
type=str,
default="_output/plot_alpha_mse.png")
parser.add_argument(
'--out-irrelev-weight-plot',
type=str,
default="_output/plot_alpha_irrelev_weight.png")
parser.add_argument(
'--out-relev-weight-plot',
type=str,
default="_output/plot_alpha_relev_weight.png")
parser.add_argument(
'--out-nonzero-hidden-plot',
type=str,
default="_output/plot_alpha_nonzero_hidden.png")
parser.add_argument(
'--out-nonzero-inputs-plot',
type=str,
default="_output/plot_alpha_nonzero_inputs.png")
parser.set_defaults()
args = parser.parse_args(args)
args.group_lasso = process_params(args.group_lasso, float)
args.lasso = process_params(args.lasso, float)
args.seeds = process_params(args.seeds, int)
return args
def parse_args(args):
''' parse command line arguments '''
parser = argparse.ArgumentParser(description=__doc__)
parser.add_argument('--seed',
type=int,
default=0)
parser.add_argument('--num-test',
type=int,
default=2000)
parser.add_argument('--data-file',
type=str,
default="_output/data.pkl")
parser.add_argument('--data-split-file',
type=str,
default="_output/data_split.pkl")
parser.add_argument('--fitted-files',
type=str,
default="_output/fitted.pkl",
help="comma separated")
parser.add_argument('--plot-accept-region-file',
type=str,
default="_output/delta_accept_region.png")
parser.set_defaults()
args = parser.parse_args()
args.fitted_files = process_params(args.fitted_files, str)
return args
def parse_args(args):
""" parse command line arguments """
parser = argparse.ArgumentParser(description=__doc__)
parser.add_argument(
"--seed",
type=int,
help="Random number generator seed for replicability",
default=12,
)
parser.add_argument("--y-max", type=float, default=10)
parser.add_argument("--y-min", type=float, default=None)
parser.add_argument("--scale-loss", type=float, default=1)
parser.add_argument("--x-start", type=int, default=None)
parser.add_argument("--x-skip", type=int, default=None)
parser.add_argument("--plot-mean", action="store_true")
parser.add_argument("--history-files",
type=str,
help="comma separated pickle files")
parser.add_argument("--loss-plot",
type=str,
default="_output/approver_history_loss.png")
parser.add_argument("--human-plot",
type=str,
default="_output/approver_history_human.png")
parser.add_argument("--log-file", type=str, default="_output/log.txt")
parser.set_defaults()
args = parser.parse_args()
args.history_files = process_params(args.history_files, str)
return args
def parse_args(args):
parser = argparse.ArgumentParser("Plot riboflavin")
parser.add_argument('--result-folder', type=str, default="riboflavin")
parser.add_argument(
'--spinn-file-template',
type=str,
default="extractor_1/prop_0.20/seed_%d/relu_0/fitted_spinn.pkl")
parser.add_argument(
'--nn-file-template',
type=str,
#default="_output/seed_%d/relu_0/layer_10,10:2,10:4/fitted_%s.csv")
default="_output/seed_%d/relu_0/layer_3,10/fitted_%s.csv")
parser.add_argument('--file-template',
type=str,
default="_output/seed_%d/fitted_%s.csv")
parser.add_argument('--methods',
type=str,
default="lasso,trees,spinn,spam,ridge_nn")
parser.add_argument('--seeds',
type=str,
default=make_params(range(40, 70)))
parser.set_defaults()
args = parser.parse_args(args)
args.methods = args.methods.split(",")
args.seeds = process_params(args.seeds, int)
return args
def parse_args(args):
''' parse command line arguments '''
parser = argparse.ArgumentParser(description=__doc__)
parser.add_argument('--data-file',
type=str,
default="_output/data.pkl")
parser.add_argument('--data-split-file',
type=str,
default="_output/data_split.pkl")
parser.add_argument('--fitted-file',
type=str,
default="_output/fitted.pkl")
parser.add_argument('--out-file',
type=str,
default="_output/recalibrated_coverages.pkl")
parser.add_argument('--log-file',
type=str,
default="_output/recalibrated_log.txt")
parser.add_argument('--alphas',
type=str,
help="""
Specifies which alpha values to calibrate for
comma separated list
""",
default="0.05,0.1,0.2")
parser.set_defaults()
args = parser.parse_args()
args.alphas = process_params(args.alphas, float)
return args
def _setup_skeleton(self):
self.input_shape = process_params(self.prediction_layer_sizes[0], int, split_str="~")
self.num_p = self.input_shape[0]
# Input layers; create placeholders in tensorflow
self.x_concat_ph = tf.placeholder(tf.float32, [None] + self.input_shape, name="x")
self.y_concat_ph = tf.placeholder(tf.float32, [None, self.label_size], name="y")
self.is_data_ph = tf.placeholder(tf.float32, [None, 1], name="is_data")
self.min_y = None
def parse_args(args):
parser = argparse.ArgumentParser("Plot for the simulation convergence check")
parser.add_argument(
'--result-folder',
type=str,
default="simulation_convergence_num_obs")
parser.add_argument(
'--spinn-template',
type=str,
default="_output/seed_%d/n_train_%d/fitted_spinn.pkl")
parser.add_argument(
'--file-template',
type=str,
default="_output/seed_%d/n_train_%d/fitted_spinn.csv")
parser.add_argument(
'--n-trains',
type=str,
default=make_params([100, 200, 400, 800, 1600, 3200]))
parser.add_argument(
'--lasso-ratio',
type=float,
default=0.1)
parser.add_argument(
'--max-relevant-idx',
type=int,
default=6)
parser.add_argument(
'--seeds',
type=str,
default=make_params(range(11,31)))
parser.add_argument(
'--out-plot',
type=str,
default="_output/plot_simulation_mse.png")
parser.add_argument(
'--out-weight-plot',
type=str,
default="_output/plot_simulation_weights.png")
parser.set_defaults()
args = parser.parse_args(args)
args.seeds = process_params(args.seeds, int)
args.n_trains = process_params(args.n_trains, int)
return args
def parse_args(args):
parser = argparse.ArgumentParser("Plot for the main simulations")
parser.add_argument('--result-folder',
type=str,
default="simulation_univar_additive")
parser.add_argument(
'--spinn-template',
type=str,
#default="_output/seed_%d/n_train_%d/fitted_spinn.pkl")
default="_output/seed_%d/n_train_%d/fitted_spinn.pkl")
parser.add_argument('--file-template',
type=str,
default="_output/seed_%d/n_train_%d/fitted_%s.csv")
parser.add_argument(
'--methods',
type=str,
default="spam,lasso,spinn,gam,trees,ridge_nn,oracle_nn")
parser.add_argument('--max-relevant-idx', type=int, default=6)
parser.add_argument(
'--n-trains',
type=str,
#default=make_params([125, 250, 500, 1000]))
default=make_params([125, 250, 500, 1000, 2000, 4000, 8000]))
parser.add_argument('--seeds', type=str, default=make_params(range(2, 22)))
parser.add_argument('--out-plot',
type=str,
default="_output/plot_simulation_mse.png")
parser.add_argument('--out-weight-plot',
type=str,
default="_output/plot_simulation_weights.png")
parser.add_argument('--show-legend', action='store_true')
parser.set_defaults()
args = parser.parse_args(args)
args.methods = args.methods.split(",")
args.seeds = process_params(args.seeds, int)
args.n_trains = process_params(args.n_trains, int)
return args
def parse_args(args):
""" parse command line arguments """
parser = argparse.ArgumentParser(description=__doc__)
parser.add_argument(
"--seed",
type=int,
help="Random number generator seed for replicability",
default=12,
)
parser.add_argument(
"--history-files", type=str, help="comma separated pickle files"
)
parser.add_argument(
"--out-file", type=str, default="_output/approver_history_agg.pkl"
)
parser.set_defaults()
args = parser.parse_args()
args.history_files = process_params(args.history_files, str)
return args
def parse_args(args):
parser = argparse.ArgumentParser("Plot peptide results")
parser.add_argument(
'--result-folder',
type=str,
default="peptide_binding/_output")
parser.add_argument(
'--hla',
type=str,
default="A")
parser.add_argument(
'--spinn-file-template',
type=str,
default="extractor_1/prop_0.20/seed_%d/relu_0/fitted_spinn.pkl")
parser.add_argument(
'--nn-file-template',
type=str,
default="extractor_1/prop_0.20/seed_%d/relu_0/fitted_%s.csv")
#default="extractor_1/prop_0.20/seed_%d/relu_0/fitted_%s_final.csv")
parser.add_argument(
'--file-template',
type=str,
default="extractor_1/prop_0.20/seed_%d/fitted_%s.csv")
parser.add_argument(
'--methods',
type=str,
default="lasso,trees,spam,spinn,ridge_nn")
parser.add_argument(
'--seeds',
type=str,
default=make_params(range(3,23)))
parser.set_defaults()
args = parser.parse_args(args)
args.methods = args.methods.split(",")
args.seeds = process_params(args.seeds, int)
return args
def parse_args(args):
""" parse command line arguments """
parser = argparse.ArgumentParser(description=__doc__)
parser.add_argument(
"--seed",
type=int,
help="Random number generator seed for replicability",
default=0,
)
parser.add_argument(
"--density-parametric-form",
type=str,
default="bernoulli",
help="The parametric form we are going to use for Y|X",
choices=["gaussian", "bernoulli"],
)
parser.add_argument("--proposer-eps", type=float, default=1e-4)
parser.add_argument("--proposer-cv", type=int, default=3)
parser.add_argument("--proposer-batches", type=str, default="1")
parser.add_argument("--proposer-alphas", type=int, default=30)
parser.add_argument("--proposer-offset-scale", type=float, default=0)
parser.add_argument("--min-y", type=float, default=-1)
parser.add_argument("--max-y", type=float, default=1)
parser.add_argument("--log-file",
type=str,
default="_output/proposer_log.txt")
parser.add_argument("--out-file", type=str, default="_output/proposer.pkl")
parser.set_defaults()
args = parser.parse_args()
assert args.min_y < args.max_y
args.proposer_batches = process_params(args.proposer_batches, int)
assert np.all(np.array(args.proposer_batches) > 0)
return args
def parse_args(args):
''' parse command line arguments '''
parser = argparse.ArgumentParser(description=__doc__)
parser.add_argument('--seed',
type=int,
help='Random number generator seed for replicability',
default=1)
parser.add_argument('--max-iters',
type=int,
help="max training iters",
default=20)
parser.add_argument('--num-inits',
type=int,
help="num random initializations",
default=1)
parser.add_argument('--data-file',
type=str,
help="input data file name, needs to be a pickle file containing a Dataset obj (see data_generator.py)",
default="_output/data.pkl")
parser.add_argument('--data-split-file',
type=str,
help="data split file",
default="_output/data_split.pkl")
parser.add_argument('--density-weight-params',
type=str,
help="""
Ridge penalty of the coefficients in the neural network parameterizing the conditioanl density
comma separated list of weight params (for tuning over)
""",
default="0.1")
parser.add_argument('--weight-penalty-type',
type=str,
help="""
Weight type lasso or ridge
""",
default="ridge",
choices=["ridge", "group_lasso"])
parser.add_argument('--density-layer-sizes',
type=str,
help="""
NN structure for density function
comma-colon separated lists of +-separated lists of layer sizes (for tuning over)
does not include the output layer
Example: 6+2+1,6+2+2+1
""",
default="4+10")
parser.add_argument('--density-parametric-form',
type=str,
default="gaussian",
help="The parametric form we are going to use for Y|X",
choices=["gaussian", "bernoulli", "multinomial"])
parser.add_argument('--dropout-rate',
type=str,
help="""
dropout rate (0 means no dropout)
comma separated list (for tuning over)
""",
default="0")
parser.add_argument('--num-ensemble',
type=int,
help="num ensemble",
default=1)
parser.add_argument('--act-func',
type=str,
help="activation function in all the NNs",
default="tanh",
choices=["tanh", "relu"])
parser.add_argument('--learning-rate',
type=float,
help="learning rate for adam",
default=0.002)
parser.add_argument('--cv',
type=int,
help="num cross-validation folds",
default=2)
parser.add_argument('--do-distributed',
action="store_true")
parser.add_argument('--log-file',
type=str,
default="_output/log.txt")
parser.add_argument('--fitted-file',
type=str,
default="_output/fitted.pkl")
parser.add_argument('--scratch-dir',
type=str,
default="scratch")
parser.set_defaults()
args = parser.parse_args()
# Parse parameters to CV over
args.density_weight_params = process_params(args.density_weight_params, float)
args.density_layer_sizes = [process_params(substr, str, "+") for substr in args.density_layer_sizes.split(",")]
args.dropout_rate = process_params(args.dropout_rate, float)
return args
def parse_args(args):
''' parse command line arguments '''
parser = argparse.ArgumentParser(description=__doc__)
parser.add_argument('--seed',
type=int,
help='Random number generator seed for replicability',
default=1)
parser.add_argument('--max-iters',
type=int,
help="max training iters",
default=2000)
parser.add_argument('--num-inits',
type=int,
help="num random initializations",
default=1)
parser.add_argument(
'--support-sim-num',
type=int,
help="number of points to sample from the support (uniformly)",
default=100)
parser.add_argument(
'--data-file',
type=str,
help=
"input data file name, needs to be a pickle file containing a Dataset obj (see data_generator.py)",
default="_output/data.pkl")
parser.add_argument('--data-split-file',
type=str,
help="data split file",
default="_output/data_split.pkl")
parser.add_argument('--do-no-harm-params',
type=str,
help="""
The weight param for the do-no-harm penalty
comma separated list (for tuning over)
""",
default="0.1")
parser.add_argument('--log-barrier-params',
type=str,
help="""
The weight param for the log barrier penalty
comma separated list(for tuning over)
""",
default="0.05")
parser.add_argument('--interval-alpha',
type=float,
help="Train intervals to have coverage (1 - alpha)",
default=0.1)
parser.add_argument('--cost-decline',
type=float,
help="The cost of declining to make a prediction",
default=0.1)
parser.add_argument('--interval-weight-params',
type=str,
help="""
Ridge penalty of the coefficients in the neural network parameterizing the interval
comma separated list of weight params (for tuning over)
""",
default="0.1")
parser.add_argument('--decision-weight-params',
type=str,
help="""
Ridge penalty of the coefficients in the neural network parameterizing the decision function
comma separated list of weight params (for tuning over)
""",
default="0.1")
parser.add_argument('--weight-penalty-type',
type=str,
help="""
Weight type lasso or ridge
""",
default="ridge",
choices=["ridge", "lasso"])
parser.add_argument('--decision-layer-sizes',
type=str,
help="""
NN structure for decision function
comma-separated lists of +-separated lists of layer sizes (for tuning over)
Example: 6+2+1,6+2+2+1
""",
default=None)
parser.add_argument('--interval-layer-sizes',
type=str,
help="""
NN structure for interval function
comma-colon separated lists of +-separated lists of layer sizes (for tuning over)
does not include the output layer
Example: 6+2+1,6+2+2+1
""",
default="4+10")
parser.add_argument('--act-func',
type=str,
help="activation function in all the NNs",
default="tanh",
choices=["tanh", "relu"])
parser.add_argument('--learning-rate',
type=float,
help="learning rate for adam",
default=0.002)
parser.add_argument('--cv',
type=int,
help="num cross-validation folds",
default=2)
parser.add_argument('--log-file', type=str, default="_output/log.txt")
parser.add_argument('--fitted-file',
type=str,
default="_output/fitted.pkl")
parser.set_defaults()
args = parser.parse_args()
# Parse parameters to CV over
args.do_no_harm_params = process_params(args.do_no_harm_params, float)
args.log_barrier_params = process_params(args.log_barrier_params, float)
args.interval_weight_params = process_params(args.interval_weight_params,
float)
args.decision_weight_params = process_params(args.decision_weight_params,
float)
args.interval_layer_sizes = [
process_params(substr, str, "+")
for substr in args.interval_layer_sizes.split(",")
]
if args.decision_layer_sizes is not None:
args.decision_layer_sizes = [
process_params(substr, str, "+")
for substr in args.decision_layer_sizes.split(",")
]
else:
args.decision_layer_sizes = [None]
return args
def parse_args(args):
''' parse command line arguments '''
parser = argparse.ArgumentParser(description=__doc__)
parser.add_argument('--seed',
type=int,
help='Random number generator seed for replicability',
default=1)
parser.add_argument('--max-iters',
type=int,
help="max training iters",
default=2000)
parser.add_argument('--num-ensemble',
type=int,
default=1)
parser.add_argument('--num-inits',
type=int,
help="num random initializations",
default=1)
parser.add_argument('--support-sim-num',
type=int,
help="number of points to sample from the support (uniformly)",
default=100)
parser.add_argument('--data-file',
type=str,
help="input data file name, needs to be a pickle file containing a Dataset obj (see data_generator.py)",
default="_output/data.pkl")
parser.add_argument('--data-split-file',
type=str,
help="data split file",
default="_output/data_split.pkl")
parser.add_argument('--log-barrier-params',
type=str,
help="""
The weight param for the log barrier penalty
comma separated list (for tuning over)
""",
default="0.01")
parser.add_argument('--do-no-harm-params',
type=str,
help="""
The weight param for the do-no-harm penalty
comma separated list (for tuning over)
""",
default="0.1")
parser.add_argument('--cost-decline',
type=float,
help=""""
The cost of declining to make a prediction.
If is zero, we assume that we actually want an accept ALL model (not accept none).
""",
default=0.1)
parser.add_argument('--density-weight-params',
type=str,
help="""
Ridge penalty of the coefficients in the neural network parameterizing the conditioanl density
comma separated list of weight params (for tuning over)
""",
default="0.1")
parser.add_argument('--decision-weight-params',
type=str,
help="""
Ridge penalty of the coefficients in the neural network parameterizing the decision function
comma separated list of weight params (for tuning over)
""",
default="0.1")
parser.add_argument('--weight-penalty-type',
type=str,
help="""
Weight type lasso or ridge
""",
default="ridge",
choices=["ridge", "group_lasso"])
parser.add_argument('--density-layer-sizes',
type=str,
help="""
NN structure for density function
comma-colon separated lists of +-separated lists of layer sizes (for tuning over)
does not include the output layer
Example: 6+2+1,6+2+2+1
""",
default="4+10")
parser.add_argument('--decision-layer-sizes',
type=str,
help="""
NN structure for density function
comma-colon separated lists of +-separated lists of layer sizes (for tuning over)
does not include the output layer
Example: 6+2+1,6+2+2+1
""",
default=None)
parser.add_argument('--dropout-rate',
type=str,
help="""
dropout rate (0 means no dropout)
comma separated list (for tuning over)
""",
default="0")
parser.add_argument('--density-parametric-form',
type=str,
default="gaussian",
help="The parametric form we are going to use for Y|X",
choices=["gaussian", "bernoulli", "multinomial"])
parser.add_argument('--act-func',
type=str,
help="activation function in all the NNs",
default="tanh",
choices=["tanh", "relu"])
parser.add_argument('--learning-rate',
type=float,
help="learning rate for adam",
default=0.002)
parser.add_argument('--cv',
type=int,
help="num cross-validation folds",
default=2)
parser.add_argument('--use-train-data-support',
action="store_true")
parser.add_argument('--empirical-support-file',
type=str,
default="",
help="support empirical distribution")
parser.add_argument('--support-empirical-scale',
type=float,
default=1,
help="How much to scale the gaussian perturbation around the empirical support distirbution")
parser.add_argument('--log-file',
type=str,
default="_output/log.txt")
parser.add_argument('--fitted-file',
type=str,
default="_output/fitted.pkl")
parser.add_argument('--do-distributed',
action="store_true")
parser.add_argument('--scratch-dir',
type=str,
default="scratch")
parser.set_defaults()
args = parser.parse_args()
# Parse parameters to CV over
args.do_no_harm_params = process_params(args.do_no_harm_params, float)
args.log_barrier_params = process_params(args.log_barrier_params, float)
args.density_weight_params = process_params(args.density_weight_params, float)
args.decision_weight_params = process_params(args.decision_weight_params, float)
args.dropout_rate = process_params(args.dropout_rate, float)
args.density_layer_sizes = [process_params(substr, str, "+") for substr in args.density_layer_sizes.split(",")]
if args.decision_layer_sizes is not None:
args.decision_layer_sizes = [process_params(substr, str, "+") for substr in args.decision_layer_sizes.split(",")]
else:
args.decision_layer_sizes = [None]
return args