mlforecast and lightgbm registered as part of dev requirement

Author: JQGoh

.github/workflows/ci.yaml

@@ -56,7 +56,7 @@ jobs:
           python-version: ${{ matrix.python-version }}
 
       - name: Install pip requirements
-        run: pip install uv && uv pip install --system ".[dev,distributed,nbconvert]"
+        run: pip install uv && uv pip install --system ".[dev,distributed]"
 
       - name: Run tests
         run: pytest --cov=nixtla --reruns 2 --reruns-delay 1 --only-rerun httpx.ConnectError nixtla_tests
@@ -86,7 +86,7 @@ jobs:
           python-version: ${{ matrix.python-version }}
 
       - name: Install pip requirements
-        run: pip install uv && uv pip install --prerelease=allow --system ".[dev,distributed]"
+        run: pip install uv && uv pip install --system ".[dev,distributed]"
 
       - name: Run tests
         run: pytest -m "not distributed_run" --reruns 2 --reruns-delay 1 --only-rerun httpx.ConnectError nixtla_tests

settings.ini

@@ -19,10 +19,10 @@ requirements = annotated-types httpx[zstd] orjson pandas pydantic>=1.10 tenacity
 dev_requirements = black datasetsforecast fire hierarchicalforecast ipython<=8.32.0 ipywidgets 
                    jupyterlab nbdev neuralforecast numpy<2 plotly polars pre-commit pyreadr<0.5.3 
                    python-dotenv pyyaml setuptools<70 statsforecast tabulate shap pytest pytest-cov pytest-rerunfailures
+                   mlforecast lightgbm
 distributed_requirements = fugue[dask,ray,spark]>=0.8.7 dask<=2024.12.1 pandas<2.2 ray<2.6.3
 plotting_requirements = utilsforecast[plotting]
 date_extra_requirements = holidays pandas_market_calendars
-nbconvert_requirements = mlforecast lightgbm statsforecast neuralforecast hierarchicalforecast
 nbs_path = nbs
 doc_path = _docs
 recursive = True