Merge branch 'main' into video

.github/conda/meta.yaml

@@ -24,7 +24,7 @@ requirements:
     - dataclasses
     - multiprocess
     - fsspec
-    - huggingface_hub >=0.22.0,<1.0.0
+    - huggingface_hub >=0.23.0,<1.0.0
     - packaging
     - aiohttp
   run:
@@ -41,7 +41,7 @@ requirements:
     - dataclasses
     - multiprocess
     - fsspec
-    - huggingface_hub >=0.22.0,<1.0.0
+    - huggingface_hub >=0.23.0,<1.0.0
     - packaging
     - aiohttp
 

.github/workflows/ci.yml

@@ -62,7 +62,7 @@ jobs:
         run: uv pip install --system --upgrade pyarrow huggingface-hub "dill<0.3.9"
       - name: Install dependencies (minimum versions)
         if: ${{ matrix.deps_versions != 'deps-latest' }}
-        run: uv pip install --system pyarrow==15.0.0 huggingface-hub==0.22.0 transformers dill==0.3.1.1
+        run: uv pip install --system pyarrow==15.0.0 huggingface-hub==0.23.5 transformers dill==0.3.1.1
       - name: Test with pytest
         run: |
           python -m pytest -rfExX -m ${{ matrix.test }} -n 2 --dist loadfile -sv ./tests/

docs/source/image_dataset.mdx

@@ -2,7 +2,7 @@
 
 There are two methods for creating and sharing an image dataset. This guide will show you how to:
 
-* Create an audio dataset from local files in python with [`Dataset.push_to_hub`]. This is an easy way that requires only a few steps in python.
+* Create an image dataset from local files in python with [`Dataset.push_to_hub`]. This is an easy way that requires only a few steps in python.
 
 * Create an image dataset with `ImageFolder` and some metadata. This is a no-code solution for quickly creating an image dataset with several thousand images.
 

setup.py

@@ -133,7 +133,7 @@
     # for data streaming via http
     "aiohttp",
     # To get datasets from the Datasets Hub on huggingface.co
-    "huggingface-hub>=0.22.0",
+    "huggingface-hub>=0.23.0",
     # Utilities from PyPA to e.g., compare versions
     "packaging",
     # To parse YAML metadata from dataset cards
@@ -235,7 +235,7 @@
 
 setup(
     name="datasets",
-    version="3.0.2.dev0",  # expected format is one of x.y.z.dev0, or x.y.z.rc1 or x.y.z (no to dashes, yes to dots)
+    version="3.0.3.dev0",  # expected format is one of x.y.z.dev0, or x.y.z.rc1 or x.y.z (no to dashes, yes to dots)
     description="HuggingFace community-driven open-source library of datasets",
     long_description=open("README.md", encoding="utf-8").read(),
     long_description_content_type="text/markdown",

src/datasets/__init__.py

@@ -12,7 +12,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-__version__ = "3.0.2.dev0"
+__version__ = "3.0.3.dev0"
 
 from .arrow_dataset import Dataset
 from .arrow_reader import ReadInstruction

src/datasets/download/download_manager.py

@@ -189,7 +189,11 @@ def _download_batched(
             download_func = partial(self._download_single, download_config=download_config)
 
             fs: fsspec.AbstractFileSystem
-            fs, path = url_to_fs(url_or_filenames[0], **download_config.storage_options)
+            path = str(url_or_filenames[0])
+            if is_relative_path(path):
+                # append the relative path to the base_path
+                path = url_or_path_join(self._base_path, path)
+            fs, path = url_to_fs(path, **download_config.storage_options)
             size = 0
             try:
                 size = fs.info(path).get("size", 0)