Upload nle_hf_dataset.py

nle_hf_dataset.py

@@ -18,6 +18,7 @@ import h5py
 import json
 import os
 import datasets
+from glob import glob
 
 
 _CITATION = """\
@@ -38,8 +39,8 @@ _HOMEPAGE = ""
 _LICENSE = ""
 
 _URLS = {
-    "data": "https://huggingface.co/datasets/Howuhh/nle_hf_dataset/tree/main/data",
-    "metadata": "https://huggingface.co/datasets/Howuhh/nle_hf_dataset/tree/main/metadata",
+    "data": glob("https://huggingface.co/datasets/Howuhh/nle_hf_dataset/tree/main/data/*.json"),
+    "metadata": glob("https://huggingface.co/datasets/Howuhh/nle_hf_dataset/tree/main/metadata/*.hdf5"),
 }
 
 class NleHfDataset(datasets.GeneratorBasedBuilder):
@@ -50,7 +51,7 @@ class NleHfDataset(datasets.GeneratorBasedBuilder):
         datasets.BuilderConfig(name="data", version=VERSION, description="Data for all episodes"),
         datasets.BuilderConfig(name="metadata", version=VERSION, description="Metadata for all episodes"),
     ]
-    DEFAULT_CONFIG_NAME = "metadata"  # It's not mandatory to have a default configuration. Just use one if it make sense.
+    DEFAULT_CONFIG_NAME = "metadata"
 
     def _info(self):
         if self.config.name == "metadata":
@@ -107,24 +108,20 @@ class NleHfDataset(datasets.GeneratorBasedBuilder):
 
     def _split_generators(self, dl_manager):
         urls = _URLS[self.config.name]
-        data_dir = dl_manager.download_and_extract(urls)
+        filepaths = [dl_manager.download(url) for url in urls]
         return [
             datasets.SplitGenerator(
-                name=datasets.Split.TRAIN,
-                gen_kwargs={"filepath": data_dir, "split": "train"},
-            )
+                name=datasets.Split.TRAIN, gen_kwargs={"filepaths": filepaths})
         ]
 
-    def _generate_examples(self, filepath, split):
-        if self.config.name == "metadata":
-            for i in range(1, 5):
-                with open(os.path.join(filepath, f"{i}.json"), encoding="utf-8") as f:
+    def _generate_examples(self, filepaths):
+        for filepath in filepaths:
+            if self.config.name == "metadata":
+                with open(filepath, encoding="utf-8") as f:
                     data = json.loads(f.read())
-                
-                yield i, data
-        else:
-            for i in range(1, 5):
-                with h5py.File(os.path.join(filepath, f"{i}.hdf5"), "r") as f:
+                    yield i, data
+            else:
+                with h5py.File(filepath, "r") as f:
                     yield i, {
                         "tty_chars": f["tty_chars"][()],
                         "tty_colors": f["tty_colors"][()],
@@ -132,4 +129,4 @@ class NleHfDataset(datasets.GeneratorBasedBuilder):
                         "actions": f["actions"][()],
                         "rewards": f["rewards"][()],
                         "dones": f["dones"][()]
-                    }
+                    }