Replace nobreakspace by normal spaces

.github/workflows/build.yml

@@ -267,7 +267,7 @@ jobs:
           echo "AUTODOC_BINDER_ENV_GH_BRANCH=${AUTODOC_BINDER_ENV_GH_BRANCH}" >> $GITHUB_ENV
           echo "AUTODOC_NOTEBOOKS_REPO_URL=${AUTODOC_NOTEBOOKS_REPO_URL}" >> $GITHUB_ENV
           echo "AUTODOC_NOTEBOOKS_BRANCH=${AUTODOC_NOTEBOOKS_BRANCH}" >> $GITHUB_ENV
-          # check computed variables
+          # check computed variables
           echo "Binder env: ${AUTODOC_BINDER_ENV_GH_REPO_NAME}/${AUTODOC_BINDER_ENV_GH_BRANCH}"
           echo "Notebooks source: ${AUTODOC_NOTEBOOKS_REPO_URL}/tree/${AUTODOC_NOTEBOOKS_BRANCH}"
 

README.md

@@ -34,7 +34,7 @@ pip install scikit-decide[all]
 ```
 For more details, see the [online documentation](https://airbus.github.io/scikit-decide/install).
 
-## Documentation
+## Documentation
 
 The latest documentation is available [online](https://airbus.github.io/scikit-decide).
 

docs/contribute.md

@@ -171,7 +171,7 @@ Then generate and serve locally the documentation with:
 poetry run yarn docs:dev
 ```
 
-NB: The above command will call  `python docs/autodoc.py` hence the use of `poetry run`.
+NB: The above command will call `python docs/autodoc.py` hence the use of `poetry run`.
 
 
 Open your web browser to access the documentation (by default on http://localhost:8080/scikit-decide/).

notebooks/11_maze_tuto.ipynb

@@ -665,7 +665,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## Conclusion"
+    "## Conclusion"
    ]
   },
   {

notebooks/12_gym_tuto.ipynb

@@ -85,7 +85,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## About Continuous Mountain Car problem"
+    "## About Continuous Mountain Car problem"
    ]
   },
   {

notebooks/14_benchmarking_tuto.ipynb

@@ -151,7 +151,7 @@
     "Now it is time to run the benchmark. \n",
     "\n",
     "Some remarks: \n",
-    "- By default, one free CPU will be allocated for each solver trial, but you can customize allocated CPUs/GPUs using the  `resources_per_trial` argument. \n",
+    "- By default, one free CPU will be allocated for each solver trial, but you can customize allocated CPUs/GPUs using the `resources_per_trial` argument. \n",
     "- Some solvers will fail for various reasons (e.g. missing required arguments, as logged in induvidual error.txt files under ~/ray_results arborescence), but this will not stop the benchmarck from running the other ones. So do not be afraid of the numerous red lines below!\n",
     "- You could fix most of the failing solvers by specifying the missing arguments thanks to `solver_args` option as shown below for `StableBaseline`.\n",
     "- To avoid a very long output, we use here a progress reporter adapted to Jupyter notebooks that will update in place the status of different jobs. As a side effect, error messages of failing solvers may be overwritten. But you can still have a look to the error files afterwards (see \"error file\" column in the second table below)."

notebooks/maze_utils.py

@@ -66,15 +66,15 @@ def generate_random_maze(width: int, height: int) -> "Maze":
             # pick a cell
             current_cell = stack.pop()
             i, j = current_cell
-            #  find unvisited neighbours
+            # find unvisited neighbours
             neighbours = [(i + di, j + dj) for (di, dj) in _deltas_neighbour]
             neighbours = [
                 (i, j)
                 for (i, j) in neighbours
                 if (i > 0) and (i < height) and (j > 0) and (j < width)
             ]
             unvisited_neighbours = [cell for cell in neighbours if cell not in visited]
-            #  remove a wall towards an unvisited cell
+            # remove a wall towards an unvisited cell
             if len(unvisited_neighbours) > 0:
                 stack.append(current_cell)
                 next_cell = random.choice(unvisited_neighbours)