"""Run the complete PCA-IV-AR(1) factor/path/mode comparison matrix."""
from __future__ import annotations
import argparse
import json
from pathlib import Path
import torch
from quantdata.dysanos.grid.torch_simulator import GaussianSpotImportanceConfig
from .comparison import write_aggregate_comparison
from .experiment import ImpliedPCAExperiment
from .run_experiment import run
_MODES = ("one-step", "one-step-weekly", "path", "weekly")
[docs]
def main() -> None:
parser = argparse.ArgumentParser(description=__doc__)
parser.add_argument(
"--path-counts", type=int, nargs="+", default=(100_000, 1_000_000)
)
parser.add_argument("--modes", choices=_MODES, nargs="+", default=_MODES)
parser.add_argument("--num-alpha", choices=("5", "all"), nargs="+", default=("5", "all"))
parser.add_argument("--seed", type=int, default=700_000)
parser.add_argument("--block-size", type=int, default=10_000)
parser.add_argument("--num-steps", type=int, default=10)
parser.add_argument("--direct-p", action="store_true")
parser.add_argument("--skip-local-tail", action="store_true")
parser.add_argument("--device", choices=("auto", "cpu", "cuda"), default="auto")
parser.add_argument("--invariant-samples", type=int, default=100_000)
parser.add_argument("--static-mesh-points", type=int, default=201)
parser.add_argument(
"--output-root", default="quantdata/dysanos/implied/reports"
)
parser.add_argument(
"--cache-root",
default="/media/storage/tmp/quantdata/dysanos/implied-pca-universes",
)
parser.add_argument("--sanos-reports", nargs="*", default=())
args = parser.parse_args()
importance = None if args.direct_p else GaussianSpotImportanceConfig()
pca_reports = []
for alpha_name in args.num_alpha:
alpha = None if alpha_name == "all" else 5
for paths in args.path_counts:
for mode in args.modes:
experiment = ImpliedPCAExperiment(
num_samples=paths,
mode=mode,
num_alpha=alpha,
num_steps=args.num_steps,
seed=args.seed + paths + 1_009 * _MODES.index(mode),
block_size=args.block_size,
importance_sampling=importance,
static_mesh_points=args.static_mesh_points,
)
pca_reports.append(
run(
experiment,
output_root=args.output_root,
cache_root=args.cache_root,
invariant_samples=args.invariant_samples,
run_local_tail=not args.skip_local_tail,
local_tail_device=(
None
if args.device == "auto"
else torch.device(args.device)
),
)
)
comparison_path = Path(args.output_root) / "pca-sanos-comparison.json"
comparison = write_aggregate_comparison(
pca_reports=pca_reports,
sanos_reports=args.sanos_reports,
output=comparison_path,
)
print(
json.dumps(
{
"status": comparison["status"],
"PCA_runs": len(pca_reports),
"report": str(comparison_path),
},
indent=2,
)
)
if __name__ == "__main__":
main()