hafnia 0.5.1__tar.gz → 0.5.3__tar.gz

{hafnia-0.5.1 → hafnia-0.5.3}/.github/workflows/build.yaml

@@ -21,8 +21,8 @@ jobs:
     outputs:
       package-version: ${{ steps.extract-version.outputs.package_version }}
     steps:
-      - uses: actions/checkout@v6.0.1
-      - uses: actions/setup-python@v6.1.0
+      - uses: actions/checkout@v6.0.2
+      - uses: actions/setup-python@v6.2.0
         with:
           python-version-file: ${{ inputs.python-version-file }}
 
@@ -38,7 +38,7 @@ jobs:
           version: 0.6.8
 
       - name: Cache dependencies
-        uses: actions/cache@v4
+        uses: actions/cache@v5
         with:
           path: ~/.cache/uv
           key: ${{ runner.os }}-uv-${{ hashFiles('**/pyproject.toml') }}

{hafnia-0.5.1 → hafnia-0.5.3}/.github/workflows/ci_cd.yaml

@@ -19,7 +19,7 @@ jobs:
     runs-on: ubuntu-latest
     needs: lint
     steps:
-      - uses: actions/checkout@v6.0.1
+      - uses: actions/checkout@v6.0.2
       - name: Run Trivy vulnerability scanner
         uses: aquasecurity/trivy-action@0.33.1
         with:

{hafnia-0.5.1 → hafnia-0.5.3}/.github/workflows/lint.yaml

@@ -10,8 +10,8 @@ jobs:
   lint:
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/checkout@v6.0.1
-      - uses: actions/setup-python@v6.1.0
+      - uses: actions/checkout@v6.0.2
+      - uses: actions/setup-python@v6.2.0
         with:
           python-version-file: ${{ inputs.python-version-file }}
       - uses: pre-commit/action@v3.0.1

{hafnia-0.5.1 → hafnia-0.5.3}/.github/workflows/publish_docker.yaml

@@ -24,8 +24,8 @@ jobs:
   build:
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/checkout@v6.0.1
-      - uses: actions/setup-python@v6.1.0
+      - uses: actions/checkout@v6.0.2
+      - uses: actions/setup-python@v6.2.0
         id: python
         with:
           python-version-file: ${{ inputs.python-version-file }}
@@ -57,7 +57,7 @@ jobs:
         uses: aws-actions/amazon-ecr-login@v2.0.1
 
       - name: Set up Docker Buildx
-        uses: docker/setup-buildx-action@v3.11.1
+        uses: docker/setup-buildx-action@v3.12.0
 
       - name: Build and push
         uses: docker/build-push-action@v6.18.0

{hafnia-0.5.1 → hafnia-0.5.3}/.github/workflows/tests.yaml

@@ -15,8 +15,8 @@ jobs:
       matrix:
         os: [ubuntu-latest, windows-latest]
     steps:
-      - uses: actions/checkout@v6.0.1
-      - uses: actions/setup-python@v6.1.0
+      - uses: actions/checkout@v6.0.2
+      - uses: actions/setup-python@v6.2.0
         with:
           python-version-file: ${{ inputs.python-version-file }}
       - name: Install uv

{hafnia-0.5.1 → hafnia-0.5.3}/.trivyignore

@@ -1,3 +1,6 @@
 # Ignore 'CVE-2024-37059' issue https://avd.aquasec.com/nvd/2024/cve-2024-37059/
 # The vulnerability does not apply to our platform as models are not loaded on our platform.
-CVE-2024-37059
+CVE-2024-37059
+
+# Temporarily ignore 'CVE-2026-0994' issue https://avd.aquasec.com/nvd/2026/cve-2026-0994/
+CVE-2026-0994

{hafnia-0.5.1 → hafnia-0.5.3}/.vscode/launch.json

@@ -77,9 +77,24 @@
                 "--dataset",
                 "mnist",
             ]
-        },    
+        },
+        {
+            "name": "cmd: 'hafnia trainer create [X]'",
+            "type": "debugpy",
+            "request": "launch",
+            "program": "${workspaceFolder}/src/hafnia_cli/__main__.py",
+            "args": [
+                "trainer",
+                "create",
+                "${workspaceFolder}/../trainer-classification",
+                "--name",
+                "Classification Trainer Package Launched in debug mode from vc-code",
+                "--description",
+                "A trainer package for image classification tasks launched in debug mode."
+            ]
+        },
         {
-            "name": "cmd: 'hafnia train-recipe [X]'",
+            "name": "cmd: 'hafnia trainer [X]'",
             "type": "debugpy",
             "request": "launch",
             "program": "${workspaceFolder}/src/hafnia_cli/__main__.py",

{hafnia-0.5.1 → hafnia-0.5.3}/PKG-INFO

@@ -1,12 +1,13 @@
 Metadata-Version: 2.4
 Name: hafnia
-Version: 0.5.1
+Version: 0.5.3
 Summary: Python SDK for communication with Hafnia platform.
 Author-email: Milestone Systems <hafniaplatform@milestone.dk>
 License-File: LICENSE
 Requires-Python: >=3.10
 Requires-Dist: boto3>=1.35.91
 Requires-Dist: click>=8.1.8
+Requires-Dist: docstring-parser>=0.17.0
 Requires-Dist: emoji>=2.14.1
 Requires-Dist: flatten-dict>=0.4.2
 Requires-Dist: keyring>=25.6.0
@@ -63,9 +64,16 @@ multiple GPUs and instances if needed.
 ## Getting started: Configuration
 To get started with Hafnia: 
 
-1. Install `hafnia` with your favorite python package manager. With pip do this:
+1. Install `hafnia` with your favorite python package manager:
+
+    ```bash
+    # With uv package manager 
+    uv add hafnia
+
+    # With pip
+    pip install hafnia
+    ```
 
-    `pip install hafnia`
 1. Sign in to the [Hafnia Platform](https://hafnia.milestonesys.com/). 
 1. Create an API KEY for Training aaS. For more instructions, follow this 
 [guide](https://hafnia.readme.io/docs/create-an-api-key). 
@@ -93,11 +101,9 @@ With Hafnia configured on your local machine, it is now possible to download
 and explore the dataset sample with a python script:
 
 ```python
-from hafnia.data import get_dataset_path
 from hafnia.dataset.hafnia_dataset import HafniaDataset
 
-# To download the sample dataset use:
-path_dataset = get_dataset_path("midwest-vehicle-detection")
+dataset = HafniaDataset.from_name("midwest-vehicle-detection")
 ```
 
 This will download the dataset sample `midwest-vehicle-detection` to the local `.data/datasets/` folder
@@ -123,11 +129,10 @@ midwest-vehicle-detection
 3 directories, 217 files
 ```
 
-You can interact with data as you want, but we also provide `HafniaDataset`
-for loading/saving, managing and interacting with the dataset.
+We provide the `HafniaDataset` format for loading/saving, managing and interacting with the dataset.
 
 We recommend the example script [examples/example_hafnia_dataset.py](examples/example_hafnia_dataset.py) 
-for a short introduction on the `HafniaDataset`.
+for a quick introduction on the `HafniaDataset`.
 
 Below is a short introduction to the `HafniaDataset` class.
 
@@ -135,7 +140,7 @@ Below is a short introduction to the `HafniaDataset` class.
 from hafnia.dataset.hafnia_dataset import HafniaDataset, Sample
 
 # Load dataset from path
-dataset = HafniaDataset.read_from_path(path_dataset)
+dataset = HafniaDataset.from_path(path_dataset)
 
 # Or get dataset directly by name
 dataset = HafniaDataset.from_name("midwest-vehicle-detection")

{hafnia-0.5.1 → hafnia-0.5.3}/README.md

@@ -34,9 +34,16 @@ multiple GPUs and instances if needed.
 ## Getting started: Configuration
 To get started with Hafnia: 
 
-1. Install `hafnia` with your favorite python package manager. With pip do this:
+1. Install `hafnia` with your favorite python package manager:
+
+    ```bash
+    # With uv package manager 
+    uv add hafnia
+
+    # With pip
+    pip install hafnia
+    ```
 
-    `pip install hafnia`
 1. Sign in to the [Hafnia Platform](https://hafnia.milestonesys.com/). 
 1. Create an API KEY for Training aaS. For more instructions, follow this 
 [guide](https://hafnia.readme.io/docs/create-an-api-key). 
@@ -64,11 +71,9 @@ With Hafnia configured on your local machine, it is now possible to download
 and explore the dataset sample with a python script:
 
 ```python
-from hafnia.data import get_dataset_path
 from hafnia.dataset.hafnia_dataset import HafniaDataset
 
-# To download the sample dataset use:
-path_dataset = get_dataset_path("midwest-vehicle-detection")
+dataset = HafniaDataset.from_name("midwest-vehicle-detection")
 ```
 
 This will download the dataset sample `midwest-vehicle-detection` to the local `.data/datasets/` folder
@@ -94,11 +99,10 @@ midwest-vehicle-detection
 3 directories, 217 files
 ```
 
-You can interact with data as you want, but we also provide `HafniaDataset`
-for loading/saving, managing and interacting with the dataset.
+We provide the `HafniaDataset` format for loading/saving, managing and interacting with the dataset.
 
 We recommend the example script [examples/example_hafnia_dataset.py](examples/example_hafnia_dataset.py) 
-for a short introduction on the `HafniaDataset`.
+for a quick introduction on the `HafniaDataset`.
 
 Below is a short introduction to the `HafniaDataset` class.
 
@@ -106,7 +110,7 @@ Below is a short introduction to the `HafniaDataset` class.
 from hafnia.dataset.hafnia_dataset import HafniaDataset, Sample
 
 # Load dataset from path
-dataset = HafniaDataset.read_from_path(path_dataset)
+dataset = HafniaDataset.from_path(path_dataset)
 
 # Or get dataset directly by name
 dataset = HafniaDataset.from_name("midwest-vehicle-detection")

{hafnia-0.5.1 → hafnia-0.5.3}/examples/example_dataset_recipe.py

@@ -36,7 +36,7 @@ rprint(dataset_recipe)
 
 # Example: Saving and loading a dataset recipe from file.
 path_recipe = Path(".data/dataset_recipes/example_recipe.json")
-json_str: str = dataset_recipe.as_json_file(path_recipe)
+dataset_recipe.as_json_file(path_recipe)
 dataset_recipe_again: DatasetRecipe = DatasetRecipe.from_json_file(path_recipe)
 
 # Verify that the loaded recipe is identical to the original recipe.

{hafnia-0.5.1 → hafnia-0.5.3}/examples/example_torchvision_dataloader.py

@@ -6,7 +6,7 @@ import torchvision.transforms.functional
 from torch.utils.data import DataLoader
 from torchvision.transforms import v2
 
-from hafnia import torch_helpers
+from hafnia.dataset import torch_helpers
 from hafnia.dataset.hafnia_dataset import HafniaDataset
 
 if __name__ == "__main__":

{hafnia-0.5.1 → hafnia-0.5.3}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "hafnia"
-version = "0.5.1"
+version = "0.5.3"
 description = "Python SDK for communication with Hafnia platform."
 readme = "README.md"
 authors = [
@@ -29,16 +29,18 @@ dependencies = [
     "mlflow>=3.4.0",
     "sagemaker-mlflow>=0.1.0",
     "mcp>=1.23.0",
+    "docstring-parser>=0.17.0",          # Used to parse docstrings for CommandBuilder
 ]
 
 [dependency-groups]
 dev = [
     "ipykernel>=6.29.5",
-    "lark>=1.2.2",
     "torch>=2.6.0",
     "torchvision>=0.21.0",
     "flatten-dict>=0.4.2",
     "pytest-cov>=7.0.0",
+    "cyclopts>=4.5.0",     # Only to test "CommandBuilder" module
+    "typer>=0.21.1",       # Only to test "CommandBuilder" module
 ]
 
 test = ["pytest>=8.3.4", "pre-commit>=4.2.0", "ruff>=0.9.1"]

{hafnia-0.5.1 → hafnia-0.5.3}/src/hafnia/dataset/dataset_recipe/dataset_recipe.py

@@ -2,6 +2,7 @@ from __future__ import annotations
 
 import json
 import os
+import shutil
 from pathlib import Path
 from typing import Any, Callable, Dict, List, Optional, Tuple, Type, Union
 
@@ -12,6 +13,7 @@ from pydantic import (
 
 from hafnia import utils
 from hafnia.dataset.dataset_helpers import dataset_name_and_version_from_string
+from hafnia.dataset.dataset_names import FILENAME_RECIPE_JSON
 from hafnia.dataset.dataset_recipe import recipe_transforms
 from hafnia.dataset.dataset_recipe.recipe_types import (
     RecipeCreation,
@@ -22,6 +24,7 @@ from hafnia.dataset.hafnia_dataset import (
     HafniaDataset,
     available_dataset_versions_from_name,
 )
+from hafnia.dataset.hafnia_dataset_types import DatasetMetadataFilePaths
 from hafnia.dataset.primitives.primitive import Primitive
 from hafnia.log import user_logger
 
@@ -117,6 +120,23 @@ class DatasetRecipe(Serializable):
         json_str = path_json.read_text(encoding="utf-8")
         return DatasetRecipe.from_json_str(json_str)
 
+    @staticmethod
+    def from_recipe_field(recipe_field: Union[str, Dict[str, Any]]) -> "DatasetRecipe":
+        """
+
+        Deserialize from a recipe field which can be either a string or a dictionary.
+
+        string: A dataset name and version string in the format 'name:version'.
+        dict: A dictionary representation of the DatasetRecipe.
+
+        """
+        if isinstance(recipe_field, str):
+            return DatasetRecipe.from_name_and_version_string(recipe_field)
+        elif isinstance(recipe_field, dict):
+            return DatasetRecipe.from_dict(recipe_field)
+
+        raise TypeError(f"Expected str or dict for recipe_field, got {type(recipe_field).__name__}.")
+
     @staticmethod
     def from_dict(data: Dict[str, Any]) -> "DatasetRecipe":
         """Deserialize from a dictionary."""
@@ -130,14 +150,9 @@ class DatasetRecipe(Serializable):
         from hafnia_cli.config import Config
 
         cfg = Config()
-        endpoint_dataset = cfg.get_platform_endpoint("dataset_recipes")
-        recipe_dict = get_dataset_recipe_by_id(recipe_id, endpoint=endpoint_dataset, api_key=cfg.api_key)
+        recipe_dict = get_dataset_recipe_by_id(recipe_id, cfg=cfg)
         recipe_dict = recipe_dict["template"]["body"]
-        if isinstance(recipe_dict, str):
-            return DatasetRecipe.from_implicit_form(recipe_dict)
-
-        recipe = DatasetRecipe.from_dict(recipe_dict)
-        return recipe
+        return DatasetRecipe.from_recipe_field(recipe_dict)
 
     @staticmethod
     def from_recipe_name(name: str) -> "DatasetRecipe":
@@ -146,8 +161,7 @@ class DatasetRecipe(Serializable):
         from hafnia_cli.config import Config
 
         cfg = Config()
-        endpoint_dataset = cfg.get_platform_endpoint("dataset_recipes")
-        recipe = get_dataset_recipe_by_name(name=name, endpoint=endpoint_dataset, api_key=cfg.api_key)
+        recipe = get_dataset_recipe_by_name(name=name, cfg=cfg)
         if not recipe:
             raise ValueError(f"Dataset recipe '{name}' not found.")
         recipe_id = recipe["id"]
@@ -168,82 +182,6 @@ class DatasetRecipe(Serializable):
 
         return DatasetRecipe.from_name(name=dataset_name, version=version)
 
-    @staticmethod
-    def from_implicit_form(recipe: Any) -> DatasetRecipe:
-        """
-        Recursively convert from implicit recipe to explicit form.
-        Handles mixed implicit/explicit recipes.
-
-        Conversion rules:
-        - str: Will get a dataset by name -> DatasetRecipeFromName
-        - Path: Will get a dataset from path -> DatasetRecipeFromPath
-        - tuple: Will merge datasets specified in the tuple -> RecipeMerger
-        - list: Will define a list of transformations -> RecipeTransforms
-
-        Example: DataRecipe from dataset name:
-        ```python
-        recipe_implicit = "mnist"
-        recipe_explicit = DatasetRecipe.from_implicit_form(recipe_implicit)
-        >>> recipe_explicit
-        DatasetRecipeFromName(dataset_name='mnist', force_redownload=False)
-        ```
-
-        Example: DataRecipe from tuple (merging multiple recipes):
-        ```python
-        recipe_implicit = ("dataset1", "dataset2")
-        recipe_explicit = DatasetRecipe.from_implicit_form(recipe_implicit)
-        >>> recipe_explicit
-        RecipeMerger(
-            recipes=[
-                DatasetRecipeFromName(dataset_name='dataset1', force_redownload=False),
-                DatasetRecipeFromName(dataset_name='dataset2', force_redownload=False)
-            ]
-        )
-
-        Example: DataRecipe from list (recipe and transformations):
-        ```python
-        recipe_implicit = ["mnist", SelectSamples(n_samples=20), Shuffle(seed=123)]
-        recipe_explicit = DatasetRecipe.from_implicit_form(recipe_implicit)
-        >>> recipe_explicit
-        Transforms(
-            recipe=DatasetRecipeFromName(dataset_name='mnist', force_redownload=False),
-            transforms=[SelectSamples(n_samples=20), Shuffle(seed=123)]
-        )
-        ```
-
-        """
-        if isinstance(recipe, DatasetRecipe):  # type: ignore
-            # It is possible to do an early return if recipe is a 'DataRecipe'-type even for nested and
-            # potentially mixed recipes. If you (really) think about it, this might surprise you,
-            # as this will bypass the conversion logic for nested recipes.
-            # However, this is not a problem as 'DataRecipe' classes are also pydantic models,
-            # so if a user introduces a 'DataRecipe'-class in the recipe (in potentially
-            # some nested and mixed implicit/explicit form) it will (due to pydantic validation) force
-            # the user to specify all nested recipes to be converted to explicit form.
-            return recipe
-
-        if isinstance(recipe, str):  # str-type is convert to DatasetFromName
-            return DatasetRecipe.from_name_and_version_string(string=recipe, resolve_missing_version=True)
-
-        if isinstance(recipe, Path):  # Path-type is convert to DatasetFromPath
-            return DatasetRecipe.from_path(path_folder=recipe)
-
-        if isinstance(recipe, tuple):  # tuple-type is convert to DatasetMerger
-            recipes = [DatasetRecipe.from_implicit_form(item) for item in recipe]
-            return DatasetRecipe.from_merger(recipes=recipes)
-
-        if isinstance(recipe, list):  # list-type is convert to Transforms
-            if len(recipe) == 0:
-                raise ValueError("List of recipes cannot be empty")
-
-            dataset_recipe = recipe[0]  # First element is the dataset recipe
-            loader = DatasetRecipe.from_implicit_form(dataset_recipe)
-
-            transforms = recipe[1:]  # Remaining items are transformations
-            return DatasetRecipe(creation=loader.creation, operations=transforms)
-
-        raise ValueError(f"Unsupported recipe type: {type(recipe)}")
-
     ### Upload, store and recipe conversions ###
     def as_python_code(self, keep_default_fields: bool = False, as_kwargs: bool = True) -> str:
         str_operations = [self.creation.as_python_code(keep_default_fields=keep_default_fields, as_kwargs=as_kwargs)]
@@ -285,17 +223,10 @@ class DatasetRecipe(Serializable):
         from hafnia.platform.dataset_recipe import get_or_create_dataset_recipe
         from hafnia_cli.config import Config
 
-        recipe = self.as_dict()
         cfg = Config()
-        endpoint_dataset = cfg.get_platform_endpoint("dataset_recipes")
-        recipe_dict = get_or_create_dataset_recipe(
-            recipe=recipe,
-            endpoint=endpoint_dataset,
-            api_key=cfg.api_key,
-            name=recipe_name,
-            overwrite=overwrite,
-        )
 
+        recipe = self.as_dict()
+        recipe_dict = get_or_create_dataset_recipe(recipe=recipe, name=recipe_name, overwrite=overwrite, cfg=cfg)
         return recipe_dict
 
     ### Dataset Recipe Transformations ###
@@ -428,13 +359,6 @@ def unique_name_from_recipe(recipe: DatasetRecipe) -> str:
     return unique_name
 
 
-def get_dataset_path_from_recipe(recipe: DatasetRecipe, path_datasets: Optional[Union[Path, str]] = None) -> Path:
-    path_datasets = path_datasets or utils.PATH_DATASETS
-    path_datasets = Path(path_datasets)
-    unique_dataset_name = unique_name_from_recipe(recipe)
-    return path_datasets / unique_dataset_name
-
-
 class FromPath(RecipeCreation):
     path_folder: Path
     check_for_images: bool = True
@@ -525,3 +449,34 @@ class FromMerger(RecipeCreation):
         for recipe in self.recipes:
             names.extend(recipe.creation.get_dataset_names())
         return names
+
+
+def get_dataset_path_from_recipe(recipe: DatasetRecipe, path_datasets: Optional[Union[Path, str]] = None) -> Path:
+    path_datasets = path_datasets or utils.PATH_DATASETS
+    path_datasets = Path(path_datasets)
+    unique_dataset_name = unique_name_from_recipe(recipe)
+    return path_datasets / unique_dataset_name
+
+
+def get_or_create_dataset_path_from_recipe(
+    dataset_recipe: DatasetRecipe,
+    force_redownload: bool = False,
+    path_datasets: Optional[Union[Path, str]] = None,
+) -> Path:
+    path_dataset = get_dataset_path_from_recipe(dataset_recipe, path_datasets=path_datasets)
+
+    if force_redownload:
+        shutil.rmtree(path_dataset, ignore_errors=True)
+
+    dataset_metadata_files = DatasetMetadataFilePaths.from_path(path_dataset)
+    if dataset_metadata_files.exists(raise_error=False):
+        return path_dataset
+
+    path_dataset.mkdir(parents=True, exist_ok=True)
+    path_recipe_json = path_dataset / FILENAME_RECIPE_JSON
+    path_recipe_json.write_text(dataset_recipe.model_dump_json(indent=4))
+
+    dataset: HafniaDataset = dataset_recipe.build()
+    dataset.write(path_dataset)
+
+    return path_dataset

{hafnia-0.5.1 → hafnia-0.5.3}/src/hafnia/dataset/hafnia_dataset.py

@@ -1,7 +1,6 @@
 from __future__ import annotations
 
 import copy
-import shutil
 from dataclasses import dataclass
 from pathlib import Path
 from random import Random
@@ -14,7 +13,6 @@ from hafnia import utils
 from hafnia.dataset import dataset_helpers
 from hafnia.dataset.dataset_helpers import is_valid_version_string, version_from_string
 from hafnia.dataset.dataset_names import (
-    FILENAME_RECIPE_JSON,
     TAG_IS_SAMPLE,
     PrimitiveField,
     SampleField,
@@ -146,17 +144,6 @@ class HafniaDataset:
         table = table_transformations.add_dataset_name_if_missing(table, dataset_name=info.dataset_name)
         return HafniaDataset(info=info, samples=table)
 
-    @staticmethod
-    def from_recipe(dataset_recipe: Any) -> "HafniaDataset":
-        """
-        Load a dataset from a recipe. The recipe can be a string (name of the dataset), a dictionary, or a DataRecipe object.
-        """
-        from hafnia.dataset.dataset_recipe.dataset_recipe import DatasetRecipe
-
-        recipe_explicit = DatasetRecipe.from_implicit_form(dataset_recipe)
-
-        return recipe_explicit.build()  # Build dataset from the recipe
-
     @staticmethod
     def from_merge(dataset0: "HafniaDataset", dataset1: "HafniaDataset") -> "HafniaDataset":
         return HafniaDataset.merge(dataset0, dataset1)
@@ -172,6 +159,8 @@ class HafniaDataset:
         If the dataset is already cached, it will be loaded from the cache.
         """
 
+        from hafnia.dataset.dataset_recipe.dataset_recipe import get_or_create_dataset_path_from_recipe
+
         path_dataset = get_or_create_dataset_path_from_recipe(
             dataset_recipe,
             path_datasets=path_datasets,
@@ -245,7 +234,7 @@ class HafniaDataset:
 
         Example: Defining split ratios and applying the transformation
 
-        >>> dataset = HafniaDataset.read_from_path(Path("path/to/dataset"))
+        >>> dataset = HafniaDataset.from_path(Path("path/to/dataset"))
         >>> split_ratios = {SplitName.TRAIN: 0.8, SplitName.VAL: 0.1, SplitName.TEST: 0.1}
         >>> dataset_with_splits = splits_by_ratios(dataset, split_ratios, seed=42)
         Or use the function as a
@@ -270,7 +259,7 @@ class HafniaDataset:
         splits based on the provided ratios.
 
         Example: Defining split ratios and applying the transformation
-        >>> dataset = HafniaDataset.read_from_path(Path("path/to/dataset"))
+        >>> dataset = HafniaDataset.from_path(Path("path/to/dataset"))
         >>> split_name = SplitName.TEST
         >>> split_ratios = {SplitName.TEST: 0.8, SplitName.VAL: 0.2}
         >>> dataset_with_splits = split_into_multiple_splits(dataset, split_name, split_ratios)
@@ -543,7 +532,7 @@ class HafniaDataset:
         primitive: Type[Primitive],
         task_name: Optional[str] = None,
         keep_sample_data: bool = False,
-    ) -> pl.DataFrame:
+    ) -> Optional[pl.DataFrame]:
         return table_transformations.create_primitive_table(
             samples_table=self.samples,
             PrimitiveType=primitive,
@@ -741,36 +730,6 @@ def check_hafnia_dataset_from_path(path_dataset: Path) -> None:
     dataset.check_dataset()
 
 
-def get_or_create_dataset_path_from_recipe(
-    dataset_recipe: Any,
-    force_redownload: bool = False,
-    path_datasets: Optional[Union[Path, str]] = None,
-) -> Path:
-    from hafnia.dataset.dataset_recipe.dataset_recipe import (
-        DatasetRecipe,
-        get_dataset_path_from_recipe,
-    )
-
-    recipe: DatasetRecipe = DatasetRecipe.from_implicit_form(dataset_recipe)
-    path_dataset = get_dataset_path_from_recipe(recipe, path_datasets=path_datasets)
-
-    if force_redownload:
-        shutil.rmtree(path_dataset, ignore_errors=True)
-
-    dataset_metadata_files = DatasetMetadataFilePaths.from_path(path_dataset)
-    if dataset_metadata_files.exists(raise_error=False):
-        return path_dataset
-
-    path_dataset.mkdir(parents=True, exist_ok=True)
-    path_recipe_json = path_dataset / FILENAME_RECIPE_JSON
-    path_recipe_json.write_text(recipe.model_dump_json(indent=4))
-
-    dataset: HafniaDataset = recipe.build()
-    dataset.write(path_dataset)
-
-    return path_dataset
-
-
 def available_dataset_versions_from_name(dataset_name: str) -> Dict[Version, "DatasetMetadataFilePaths"]:
     credentials: ResourceCredentials = get_read_credentials_by_name(dataset_name=dataset_name)
     return available_dataset_versions(credentials=credentials)
@@ -795,12 +754,13 @@ def select_version_from_available_versions(
 
     if version is None:
         str_versions = [str(v) for v in available_versions]
-        raise ValueError(f"Version must be specified. Available versions: {str_versions}")
-    elif version == "latest":
+        raise ValueError(f"Version must be specified. Available versions: {str_versions}. ")
+
+    if version == "latest":
         version_casted = max(available_versions)
         user_logger.info(f"'latest' version '{version_casted}' has been selected")
     else:
-        version_casted = version_from_string(version)
+        version_casted = version_from_string(version, raise_error=True)
 
     if version_casted not in available_versions:
         raise ValueError(f"Selected version '{version}' not found in available versions: {available_versions}")

{hafnia-0.5.1 → hafnia-0.5.3}/src/hafnia/dataset/hafnia_dataset_types.py

@@ -470,7 +470,7 @@ class Sample(BaseModel):
         return image
 
     def draw_annotations(self, image: Optional[np.ndarray] = None) -> np.ndarray:
-        from hafnia.visualizations import image_visualizations
+        from hafnia.dataset import image_visualizations
 
         if image is None:
             image = self.read_image()

{hafnia-0.5.1 → hafnia-0.5.3}/src/hafnia/dataset/operations/dataset_stats.py

@@ -104,7 +104,8 @@ def calculate_primitive_counts(dataset: HafniaDataset) -> Dict[str, int]:
         name = task.primitive.__name__
         if task.name != task.primitive.default_task_name():
             name = f"{name}.{task.name}"
-        annotation_counts[name] = len(objects)
+        n_objects = 0 if objects is None else len(objects)
+        annotation_counts[name] = n_objects
     return annotation_counts
 
 

{hafnia-0.5.1 → hafnia-0.5.3}/src/hafnia/dataset/primitives/classification.py

@@ -39,7 +39,7 @@ class Classification(Primitive):
     def draw(self, image: np.ndarray, inplace: bool = False, draw_label: bool = True) -> np.ndarray:
         if draw_label is False:
             return image
-        from hafnia.visualizations import image_visualizations
+        from hafnia.dataset import image_visualizations
 
         class_name = self.get_class_name()
         if self.task_name == self.default_task_name():

{hafnia-0.5.1 → hafnia-0.5.3}/src/hafnia/dataset/primitives/segmentation.py

@@ -4,9 +4,9 @@ import cv2
 import numpy as np
 from pydantic import Field
 
+from hafnia.dataset.colors import get_n_colors
 from hafnia.dataset.primitives.primitive import Primitive
 from hafnia.dataset.primitives.utils import get_class_name
-from hafnia.visualizations.colors import get_n_colors
 
 
 class Segmentation(Primitive):