zipoop 0.1.0__py3-none-any.whl
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- zipoop/__init__.py +11 -0
- zipoop/cli.py +30 -0
- zipoop/data/encode-panel_1885.json +32502 -0
- zipoop/dataset.py +119 -0
- zipoop/decoder.py +62 -0
- zipoop/evaluation.py +77 -0
- zipoop/masks.py +73 -0
- zipoop/network.py +102 -0
- zipoop/panel.py +26 -0
- zipoop/shard.py +105 -0
- zipoop/teacher.py +43 -0
- zipoop-0.1.0.dist-info/METADATA +51 -0
- zipoop-0.1.0.dist-info/RECORD +17 -0
- zipoop-0.1.0.dist-info/WHEEL +5 -0
- zipoop-0.1.0.dist-info/entry_points.txt +2 -0
- zipoop-0.1.0.dist-info/licenses/LICENSE +21 -0
- zipoop-0.1.0.dist-info/top_level.txt +1 -0
zipoop/__init__.py
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
1
|
+
"""Zipoop: decoder and reader for the per-window teacher shards (merged per-chromosome npz)."""
|
|
2
|
+
from .dataset import Dataset, verify
|
|
3
|
+
from .decoder import Decoder
|
|
4
|
+
from .evaluation import calibrate, restore_real, summarize, window_pearson
|
|
5
|
+
from .panel import panel_tracks, zero_tracks
|
|
6
|
+
from .shard import Shard, read_shard
|
|
7
|
+
from .teacher import TeacherWindow, load_window
|
|
8
|
+
|
|
9
|
+
__version__ = "0.1.0"
|
|
10
|
+
__all__ = ["Dataset", "verify", "Decoder", "Shard", "read_shard", "TeacherWindow", "load_window",
|
|
11
|
+
"panel_tracks", "zero_tracks", "calibrate", "restore_real", "summarize", "window_pearson"]
|
zipoop/cli.py
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
1
|
+
import argparse
|
|
2
|
+
import sys
|
|
3
|
+
|
|
4
|
+
from .dataset import Dataset, verify
|
|
5
|
+
|
|
6
|
+
|
|
7
|
+
def main(argv=None):
|
|
8
|
+
parser = argparse.ArgumentParser(prog="zipoop", description="Zipoop teacher-shard dataset tools")
|
|
9
|
+
sub = parser.add_subparsers(dest="cmd", required=True)
|
|
10
|
+
v = sub.add_parser("verify", help="check <root>/*.npz against MANIFEST.tsv (size + sha256)")
|
|
11
|
+
v.add_argument("root")
|
|
12
|
+
v.add_argument("--chroms", default="", help="comma-separated subset")
|
|
13
|
+
v.add_argument("--workers", type=int, default=4, help="files hashed in parallel")
|
|
14
|
+
s = sub.add_parser("show", help="decode one window and print its shapes")
|
|
15
|
+
s.add_argument("root")
|
|
16
|
+
s.add_argument("window")
|
|
17
|
+
args = parser.parse_args(argv)
|
|
18
|
+
if args.cmd == "verify":
|
|
19
|
+
chroms = set(args.chroms.split(",")) if args.chroms else None
|
|
20
|
+
bad = verify(args.root, chroms, workers=args.workers, progress=lambda c, problem: print(f"{c}: {problem or 'ok'}", flush=True))
|
|
21
|
+
print(f"{len(bad)} problem(s)" if bad else "all files verified")
|
|
22
|
+
return 1 if bad else 0
|
|
23
|
+
with Dataset(args.root) as ds:
|
|
24
|
+
w = ds.load_window(args.window)
|
|
25
|
+
print(f"{w.key}: dna {tuple(w.dna.shape)}, signal {tuple(w.signal.shape)}, valid fraction {w.valid.float().mean():.4f}")
|
|
26
|
+
return 0
|
|
27
|
+
|
|
28
|
+
|
|
29
|
+
if __name__ == "__main__":
|
|
30
|
+
sys.exit(main())
|