eegdash 0.3.0.dev172943270__tar.gz → 0.3.1.dev50__tar.gz

{eegdash-0.3.0.dev172943270 → eegdash-0.3.1.dev50}/LICENSE

@@ -4,6 +4,8 @@ Copyright (C) 2024-2025
 
 Young Truong, UCSD, dt.young112@gmail.com
 Arnaud Delorme, UCSD, adelorme@ucsd.edu
+Aviv Dotan, BGU, avivdot@bgu.post.ac.il
+Oren Shriki, BGU, shrikio@bgu.ac.il
 Bruno Aristimunha, b.aristimunha@gmail.com
 
 This program is free software; you can redistribute it and/or modify

{eegdash-0.3.0.dev172943270 → eegdash-0.3.1.dev50}/PKG-INFO

@@ -1,14 +1,16 @@
 Metadata-Version: 2.4
 Name: eegdash
-Version: 0.3.0.dev172943270
+Version: 0.3.1.dev50
 Summary: EEG data for machine learning
-Author-email: Young Truong <dt.young112@gmail.com>, Arnaud Delorme <adelorme@gmail.com>, Bruno Aristimunha <b.aristimunha@gmail.com>
+Author-email: Young Truong <dt.young112@gmail.com>, Arnaud Delorme <adelorme@gmail.com>, Aviv Dotan <avivd220@gmail.com>, Oren Shriki <oren70@gmail.com>, Bruno Aristimunha <b.aristimunha@gmail.com>
 License: GNU General Public License
         
         Copyright (C) 2024-2025
         
         Young Truong, UCSD, dt.young112@gmail.com
         Arnaud Delorme, UCSD, adelorme@ucsd.edu
+        Aviv Dotan, BGU, avivdot@bgu.post.ac.il
+        Oren Shriki, BGU, shrikio@bgu.ac.il
         Bruno Aristimunha, b.aristimunha@gmail.com
         
         This program is free software; you can redistribute it and/or modify

{eegdash-0.3.0.dev172943270 → eegdash-0.3.1.dev50}/eegdash/__init__.py

@@ -5,4 +5,4 @@ from .utils import __init__mongo_client
 __init__mongo_client()
 
 __all__ = ["EEGDash", "EEGDashDataset", "EEGChallengeDataset"]
-__version__ = "0.3.0.dev172943270"
+__version__ = "0.3.1.dev50"

{eegdash-0.3.0.dev172943270 → eegdash-0.3.1.dev50}/eegdash/data_utils.py

@@ -55,16 +55,33 @@ class EEGDashBaseDataset(BaseDataset):
         self.cache_dir = Path(cache_dir)
         bids_kwargs = self.get_raw_bids_args()
 
+        if s3_bucket:
+            self.s3_bucket = s3_bucket
+            self.s3_open_neuro = False
+            bids_root = self.cache_dir
+            self.filecache = self.cache_dir / record["bidspath"]
+        else:
+            self.s3_bucket = self._AWS_BUCKET
+            self.s3_open_neuro = True
+            bids_root = self.cache_dir / record["dataset"]
+            self.filecache = self.cache_dir / record["bidspath"]
+
         self.bidspath = BIDSPath(
-            root=self.cache_dir / record["dataset"],
+            root=bids_root,
             datatype="eeg",
             suffix="eeg",
             **bids_kwargs,
         )
-        self.s3_bucket = s3_bucket if s3_bucket else self._AWS_BUCKET
+
         self.s3file = self.get_s3path(record["bidspath"])
-        self.filecache = self.cache_dir / record["bidspath"]
         self.bids_dependencies = record["bidsdependencies"]
+        # Temporary fix for BIDS dependencies path
+        # just to release to the competition
+        if not self.s3_open_neuro:
+            self.bids_dependencies = [
+                dep.split("/", 1)[1] for dep in self.bids_dependencies
+            ]
+
         self._raw = None
 
     def get_s3path(self, filepath: str) -> str:
@@ -72,11 +89,18 @@ class EEGDashBaseDataset(BaseDataset):
         return f"{self.s3_bucket}/{filepath}"
 
     def _download_s3(self) -> None:
-        """Fetch the given data from its S3 location and cache it locally."""
-        self.filecache.parent.mkdir(parents=True, exist_ok=True)
+        """Download function that gets the raw EEG data from S3."""
         filesystem = s3fs.S3FileSystem(
             anon=True, client_kwargs={"region_name": "us-east-2"}
         )
+        if not self.s3_open_neuro:
+            self.s3file = re.sub(r"(^|/)ds\d{6}/", r"\1", self.s3file, count=1)
+            self.filecache = re.sub(
+                r"(^|/)ds\d{6}/", r"\1", str(self.filecache), count=1
+            )
+            self.filecache = Path(self.filecache)
+
+        self.filecache.parent.mkdir(parents=True, exist_ok=True)
         filesystem.download(self.s3file, self.filecache)
         self.filenames = [self.filecache]
 
@@ -90,6 +114,8 @@ class EEGDashBaseDataset(BaseDataset):
         for dep in self.bids_dependencies:
             s3path = self.get_s3path(dep)
             filepath = self.cache_dir / dep
+            # here, we download the dependency and it is fine
+            # in the case of the competition.
             if not filepath.exists():
                 filepath.parent.mkdir(parents=True, exist_ok=True)
                 filesystem.download(s3path, filepath)
@@ -108,7 +134,7 @@ class EEGDashBaseDataset(BaseDataset):
                 self._download_dependencies()
             self._download_s3()
         if self._raw is None:
-            self._raw = mne.io.read_raw(self.bidspath, verbose=False)
+            self._raw = mne.io.read_raw(fname=self.bidspath, verbose=False)
 
     # === BaseDataset and PyTorch Dataset interface ===
 
@@ -215,6 +241,7 @@ class EEGDashBaseRaw(BaseRaw):
         )
 
     def get_s3path(self, filepath):
+        print(f"Getting S3 path for {filepath}")
         return f"{self._AWS_BUCKET}/{filepath}"
 
     def _download_s3(self):

{eegdash-0.3.0.dev172943270 → eegdash-0.3.1.dev50}/eegdash/dataset.py

@@ -1,5 +1,19 @@
 from .api import EEGDashDataset
 
+RELEASE_TO_OPENNEURO_DATASET_MAP = {
+    "R11": "ds005516",
+    "R10": "ds005515",
+    "R9": "ds005514",
+    "R8": "ds005512",
+    "R7": "ds005511",
+    "R6": "ds005510",
+    "R4": "ds005508",
+    "R5": "ds005509",
+    "R3": "ds005507",
+    "R2": "ds005506",
+    "R1": "ds005505",
+}
+
 
 class EEGChallengeDataset(EEGDashDataset):
     def __init__(
@@ -7,7 +21,7 @@ class EEGChallengeDataset(EEGDashDataset):
         release: str = "R5",
         query: dict | None = None,
         cache_dir: str = ".eegdash_cache",
-        s3_bucket: str | None = "s3://nmdatasets/NeurIPS25/",
+        s3_bucket: str | None = "s3://nmdatasets/NeurIPS25",
         **kwargs,
     ):
         """Create a new EEGDashDataset from a given query or local BIDS dataset directory
@@ -32,25 +46,11 @@ class EEGChallengeDataset(EEGDashDataset):
             constructor.
 
         """
-        dsnumber_release_map = {
-            "R11": "ds005516",
-            "R10": "ds005515",
-            "R9": "ds005514",
-            "R8": "ds005512",
-            "R7": "ds005511",
-            "R6": "ds005510",
-            "R4": "ds005508",
-            "R5": "ds005509",
-            "R3": "ds005507",
-            "R2": "ds005506",
-            "R1": "ds005505",
-        }
-
         self.release = release
-        if release not in dsnumber_release_map:
+        if release not in RELEASE_TO_OPENNEURO_DATASET_MAP:
             raise ValueError(f"Unknown release: {release}")
 
-        dataset = dsnumber_release_map[release]
+        dataset = RELEASE_TO_OPENNEURO_DATASET_MAP[release]
         if query is None:
             query = {"dataset": dataset}
         elif "dataset" not in query:

{eegdash-0.3.0.dev172943270 → eegdash-0.3.1.dev50}/eegdash/features/feature_bank/complexity.py

@@ -71,7 +71,7 @@ def complexity_svd_entropy(x, m=10, tau=1):
 @FeaturePredecessor(*SIGNAL_PREDECESSORS)
 @univariate_feature
 @nb.njit(cache=True, fastmath=True)
-def complexity_lempel_ziv(x, threshold=None):
+def complexity_lempel_ziv(x, threshold=None, normalize=True):
     lzc = np.empty(x.shape[:-1])
     for i in np.ndindex(x.shape[:-1]):
         t = np.median(x[i]) if threshold is None else threshold
@@ -97,4 +97,6 @@ def complexity_lempel_ziv(x, threshold=None):
                     j, k, k_max = 0, 1, 1
                 else:
                     k = 1
+        if normalize:
+            lzc[i] *= np.log2(n) / n
     return lzc

{eegdash-0.3.0.dev172943270 → eegdash-0.3.1.dev50}/eegdash/features/feature_bank/csp.py

@@ -97,5 +97,5 @@ class CommonSpatialPattern(TrainableFeature):
                 + "all weights were filtered out."
             )
         proj = (self.transform_input(x) - self._mean) @ w
-        proj = proj.reshape(x.shape[0], x.shape[2], -1).mean(axis=1)
+        proj = proj.reshape(x.shape[0], x.shape[2], -1).var(axis=1)
         return {f"{i}": proj[:, i] for i in range(proj.shape[-1])}

{eegdash-0.3.0.dev172943270 → eegdash-0.3.1.dev50}/eegdash.egg-info/PKG-INFO

@@ -1,14 +1,16 @@
 Metadata-Version: 2.4
 Name: eegdash
-Version: 0.3.0.dev172943270
+Version: 0.3.1.dev50
 Summary: EEG data for machine learning
-Author-email: Young Truong <dt.young112@gmail.com>, Arnaud Delorme <adelorme@gmail.com>, Bruno Aristimunha <b.aristimunha@gmail.com>
+Author-email: Young Truong <dt.young112@gmail.com>, Arnaud Delorme <adelorme@gmail.com>, Aviv Dotan <avivd220@gmail.com>, Oren Shriki <oren70@gmail.com>, Bruno Aristimunha <b.aristimunha@gmail.com>
 License: GNU General Public License
         
         Copyright (C) 2024-2025
         
         Young Truong, UCSD, dt.young112@gmail.com
         Arnaud Delorme, UCSD, adelorme@ucsd.edu
+        Aviv Dotan, BGU, avivdot@bgu.post.ac.il
+        Oren Shriki, BGU, shrikio@bgu.ac.il
         Bruno Aristimunha, b.aristimunha@gmail.com
         
         This program is free software; you can redistribute it and/or modify

{eegdash-0.3.0.dev172943270 → eegdash-0.3.1.dev50}/pyproject.toml

@@ -8,6 +8,8 @@ name = "eegdash"
 authors = [
   { name = "Young Truong",    email = "dt.young112@gmail.com" },
   { name = "Arnaud Delorme",  email = "adelorme@gmail.com" },
+  { name = "Aviv Dotan",  email = "avivd220@gmail.com" },
+  { name = "Oren Shriki",  email = "oren70@gmail.com" },
   { name = "Bruno Aristimunha", email = "b.aristimunha@gmail.com"}
 ]
 description     = "EEG data for machine learning"

{eegdash-0.3.0.dev172943270 → eegdash-0.3.1.dev50}/tests/test_dataset.py

@@ -30,7 +30,7 @@ def test_eeg_challenge_dataset_initialization():
     dataset = EEGChallengeDataset(release="R5")
 
     release = "R5"
-    expected_bucket_prefix = f"s3://nmdatasets/NeurIPS25//{release}_L100"
+    expected_bucket_prefix = f"s3://nmdatasets/NeurIPS25/{release}_L100"
     assert dataset.s3_bucket == expected_bucket_prefix, (
         f"Unexpected s3_bucket: {dataset.s3_bucket} (expected {expected_bucket_prefix})"
     )
@@ -79,4 +79,10 @@ def test_mongodb_load_under_slo(release):
     start_time = time.perf_counter()
     _ = EEGChallengeDataset(release=release)
     duration = time.perf_counter() - start_time
-    assert duration < 10, f"{release} took {duration:.2f}s"
+    assert duration < 30, f"{release} took {duration:.2f}s"
+
+
+def test_consuming_data_r5():
+    dataset_obj = EEGChallengeDataset(release="R5", query=dict(task="RestingState"))
+    raw = dataset_obj.datasets[0].raw
+    assert raw is not None