add splits + remove accessory dataset.py

.gitattributes

@@ -56,6 +56,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.webp filter=lfs diff=lfs merge=lfs -text
 # Video files - compressed
 *.mp4 filter=lfs diff=lfs merge=lfs -text
-*.webm filter=lfs diff=lfs merge=lfs -text
-
-training_mvalues.parquet preview=1
+*.webm filter=lfs diff=lfs merge=lfs -text

.gitignore

@@ -0,0 +1,5 @@
+# dev files
+_*
+
+# large files - upload with hf_transfer or huggingface-cli
+*.parquet

dataset.py

@@ -1,64 +0,0 @@
-import datasets
-import pandas as pd
-
-_CITATION = """\
-@article{kim2024methylation_atlas,
-  title   = {Ontology-aware DNA methylation classification with a curated atlas of human tissues and cell types},
-  author  = {Kim, Mirae and Dannenfelser, Ruth and Cui, Yufei and Allen, Genevera and Yao, Vicky},
-  journal = {bioRxiv preprint},
-  year    = {2025},
-  doi     = {10.1101/2024.XX.XXXXXX}
-}
-"""
-
-_DESCRIPTION = ""
-_HOMEPAGE = "https://github.com/ylaboratory/methylation-classification"
-_LICENSE = "cc-by-4.0"
-
-class MethylClassification(datasets.GeneratorBasedBuilder):
-    VERSION = datasets.Version("1.0.0")
-
-    def _info(self):
-        return datasets.DatasetInfo(
-            description=_DESCRIPTION,
-            features=None,
-            supervised_keys=None,
-            homepage=_HOMEPAGE,
-            license=_LICENSE,
-            citation=_CITATION,
-        )
-
-    def _split_generators(self, dl_manager):
-        files = dl_manager.download_and_extract({
-            "train_mvalues": "training_mvalues.parquet",
-            "train_meta": "training_meta.parquet",
-            "test_mvalues": "label_transfer_mvalues.parquet",
-            "test_meta": "label_transfer_meta.parquet",
-        })
-
-        return [
-            datasets.SplitGenerator(
-                name=datasets.Split.TRAIN,
-                gen_kwargs={
-                    "mvalues_path": files["train_mvalues"],
-                    "meta_path": files["train_meta"]
-                }
-            ),
-            datasets.SplitGenerator(
-                name=datasets.Split.TEST,
-                gen_kwargs={
-                    "mvalues_path": files["test_mvalues"],
-                    "meta_path": files["test_meta"]
-                }
-            )
-        ]
-
-    def _generate_examples(self, mvalues_path, meta_path):
-        mvalues_df = pd.read_parquet(mvalues_path)
-        meta_df = pd.read_parquet(meta_path)
-
-        # Combine the DataFrames by index (assuming the index is the sample ID and aligned)
-        combined_df = pd.merge(meta_df, mvalues_df, left_index=True, right_index=True)
-
-        for idx, row in combined_df.iterrows():
-            yield idx, row.to_dict()

label_transfer_mvalues.parquet

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:5347f27cff73f022f47facecc6c5d3f769b7e780b3defa50c389f39a38376722
-size 18954632144

training_mvalues.parquet

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:7aef4a03f8d89c4406afda205d0a8e19d12f08189dfd317db69cb571a67524ee
-size 30091856136