ENH ref to benchopt template

README.rst

@@ -141,7 +141,7 @@ Each solver derive from the [`benchopt.BaseSolver` class](https://benchopt.githu
 
 2 - How to add a new problem?
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
-In this benchmark, each problem is defined by a [Dataset class](https://benchopt.github.io/user_guide/generated/benchopt.BaseDataset.html) in the [datasets](datasets) folder. 
+In this benchmark, each problem is defined by a [Dataset class](https://benchopt.github.io/user_guide/generated/benchopt.BaseDataset.html) in the [datasets](datasets) folder. A [template](datasets/template_dataset.py) is provided.
 
 Cite
 ----

datasets/template_dataset.py

@@ -29,6 +29,12 @@ class Dataset(BaseDataset):
     """Hyperparameter optimization with IJCNN1 dataset."""
     # Name to select the dataset in the CLI and to display the results.
     name = "ijcnn1"
+    """How to add a new problem to the benchmark?
+
+    This template dataset is an adaptation of the dataset from the benchopt
+    template benchmark (https://github.com/benchopt/template_benchmark/) to
+    the bilevel setting.
+    """
 
     install_cmd = 'conda'
     # List of packages needed to run the dataset. See the corresponding

solvers/template_solver.py

@@ -25,7 +25,7 @@ class Solver(BaseSolver):
     # Name to select the solver in the CLI and to display the results.
     name = 'jaxopt_GD'
 
-    """How to add a new stochastic solver to the benchmark?
+    """How to add a new  solver to the benchmark?
 
     This template solver is an adaptation of the solver from the benchopt
     template benchmark (https://github.com/benchopt/template_benchmark/) to