UPDATE: Fix typos, PEP8 python format, correct declaration of return …

…type (#47)

CONTRIBUTING.md

@@ -7,10 +7,10 @@ We want to make contributing to this project as easy and transparent as possible
 - Proposing new features
 - Becoming a maintainer
 
-## We Develop with Github
-We use github to host code, to track issues and feature requests, as well as accept pull requests.
+## We Develop with GitHub
+We use GitHub to host code, to track issues and feature requests, as well as accept pull requests.
 
-## We Use [Github Flow](https://guides.github.com/introduction/flow/index.html), So All Code Changes Happen Through Pull Requests
+## We Use [GitHub Flow](https://guides.github.com/introduction/flow/index.html), So All Code Changes Happen Through Pull Requests
 Pull requests are the best way to propose changes to the codebase (we use [Github Flow](https://guides.github.com/introduction/flow/index.html)). We actively welcome your pull requests:
 
 1. Fork the repo and create your branch from `main`.
@@ -20,7 +20,7 @@ Pull requests are the best way to propose changes to the codebase (we use [Githu
 5. Make sure your code lints.
 6. Issue that pull request!
 
-## Report bugs using Github's [issues](https://github.com/nrel/graph-env/issues)
+## Report bugs using GitHub's [issues](https://github.com/nrel/graph-env/issues)
 We use GitHub issues to track public bugs. Report a bug by [opening a new issue](https://github.com/NREL/graph-env/issues/new).
 
 ## License

experiments/tsp/tsp_profiling.ipynb

@@ -4,7 +4,11 @@
    "cell_type": "code",
    "execution_count": 1,
    "id": "491b3ee4",
-   "metadata": {},
+   "metadata": {
+    "pycharm": {
+     "name": "#%%\n"
+    }
+   },
    "outputs": [
     {
      "name": "stdout",
@@ -30,7 +34,11 @@
    "cell_type": "code",
    "execution_count": 2,
    "id": "93a397e6",
-   "metadata": {},
+   "metadata": {
+    "pycharm": {
+     "name": "#%%\n"
+    }
+   },
    "outputs": [],
    "source": [
     "from graphenv.examples.tsp.graph_utils import make_complete_planar_graph\n",
@@ -46,7 +54,11 @@
    "cell_type": "code",
    "execution_count": 3,
    "id": "94e1fdae",
-   "metadata": {},
+   "metadata": {
+    "pycharm": {
+     "name": "#%%\n"
+    }
+   },
    "outputs": [],
    "source": [
     "G_fn = lambda : make_complete_planar_graph(N=40)\n",
@@ -59,7 +71,11 @@
    "cell_type": "code",
    "execution_count": 4,
    "id": "ee3fdbca",
-   "metadata": {},
+   "metadata": {
+    "pycharm": {
+     "name": "#%%\n"
+    }
+   },
    "outputs": [
     {
      "name": "stdout",
@@ -78,7 +94,11 @@
    "cell_type": "code",
    "execution_count": 5,
    "id": "b46c6cee",
-   "metadata": {},
+   "metadata": {
+    "pycharm": {
+     "name": "#%%\n"
+    }
+   },
    "outputs": [
     {
      "name": "stdout",
@@ -97,7 +117,11 @@
    "cell_type": "code",
    "execution_count": 6,
    "id": "7fb538bb",
-   "metadata": {},
+   "metadata": {
+    "pycharm": {
+     "name": "#%%\n"
+    }
+   },
    "outputs": [],
    "source": [
     "tsp_genv = GraphEnv({'state': tsp_state, 'max_num_children': 40})\n",
@@ -108,7 +132,11 @@
    "cell_type": "code",
    "execution_count": 7,
    "id": "ac8f3a04",
-   "metadata": {},
+   "metadata": {
+    "pycharm": {
+     "name": "#%%\n"
+    }
+   },
    "outputs": [
     {
      "name": "stdout",
@@ -130,7 +158,11 @@
    "cell_type": "code",
    "execution_count": 8,
    "id": "f5732a03",
-   "metadata": {},
+   "metadata": {
+    "pycharm": {
+     "name": "#%%\n"
+    }
+   },
    "outputs": [
     {
      "name": "stdout",
@@ -152,7 +184,11 @@
    "cell_type": "code",
    "execution_count": 9,
    "id": "587089c0",
-   "metadata": {},
+   "metadata": {
+    "pycharm": {
+     "name": "#%%\n"
+    }
+   },
    "outputs": [],
    "source": [
     "input_obs = tsp_genv.make_observation()\n",
@@ -163,7 +199,11 @@
    "cell_type": "code",
    "execution_count": 10,
    "id": "3db24ead",
-   "metadata": {},
+   "metadata": {
+    "pycharm": {
+     "name": "#%%\n"
+    }
+   },
    "outputs": [],
    "source": [
     "tsp_model = TSPModel._create_base_model(hidden_dim=256, embed_dim=256, num_nodes=40)\n",
@@ -174,7 +214,11 @@
    "cell_type": "code",
    "execution_count": 11,
    "id": "c1ca31da",
-   "metadata": {},
+   "metadata": {
+    "pycharm": {
+     "name": "#%%\n"
+    }
+   },
    "outputs": [
     {
      "name": "stdout",
@@ -195,7 +239,11 @@
    "cell_type": "code",
    "execution_count": 12,
    "id": "112041ff",
-   "metadata": {},
+   "metadata": {
+    "pycharm": {
+     "name": "#%%\n"
+    }
+   },
    "outputs": [
     {
      "name": "stdout",
@@ -234,4 +282,4 @@
  },
  "nbformat": 4,
  "nbformat_minor": 5
-}
+}

graphenv/graph_model.py

@@ -214,7 +214,7 @@ def _stack_batch_dim(
 
 def _mask_and_split_values(
     flat_values: TensorType, obs: RepeatedValues, tensorlib: str = "tf"
-) -> Tuple[TensorType]:
+) -> Tuple[TensorType, TensorType]:
     """Returns the value for the current vertex (index 0 of values),
     and the masked values of the action vertices.
     Args:

graphenv/vertex.py

@@ -78,7 +78,7 @@ def _make_observation(self) -> any:
     def children(self) -> List[V]:
         """
         Gets the child vertices of this vertex.
-        Acts as a wrapper that memoizes calls to _get_children() and
+        Acts as a wrapper that memorizes calls to _get_children() and
         ensures that it is a list. If you would like a different behavior,
         such as stochastic child vertices, override this property.
 
@@ -93,7 +93,7 @@ def children(self) -> List[V]:
     def observation(self) -> any:
         """
         Gets the observation of this vertex.
-        Acts as a wrapper that memoizes calls to _make_observation().
+        Acts as a wrapper that memorizes calls to _make_observation().
         If you would like a different behavior,
         such as stochastic observations, override this property.
 

joss/paper.md

@@ -67,7 +67,7 @@ molecular optimization, for example, one can use domain knowledge alongside exis
 libraries to model the effect of adding a single bond or atom to an existing but incomplete
 molecule, and let the RL algorithm build a model of how good a given decision is by "experiencing"
 the simulated environment many times through.  In contrast, a model-based mathematical 
-formulation that fully expresses all of the chemical and physical constraints is intractable.
+formulation that fully expresses all the chemical and physical constraints is intractable.
 
 In recent  years, RL has emerged as an effective paradigm for optimizing searches over graphs 
 and led to state-of-the-art heuristics for games like Go and chess, as well as for classical CO

tests/test_graph_model_utils.py

@@ -78,4 +78,4 @@ def test_mask_and_split_values(obs, vals, tensorlib, request):
         dtype_min = torch.finfo(action_vals.dtype).min
 
     for length, val_row in zip([1, 4, 8, 2, 4], action_vals.numpy()):
-        assert np.allclose(val_row[length - 1 :], dtype_min)
+        assert np.allclose(val_row[length - 1:], dtype_min)

tests/test_hallway.py

@@ -75,7 +75,6 @@ def test_graphenv_step(hallway_env: GraphEnv):
 
 
 def test_rllib(ray_init, agent, caplog):
-
     trainer_fn, config, needs_q_model = agent
     model = HallwayQModel if needs_q_model else HallwayModel
 
@@ -103,7 +102,6 @@ def test_rllib(ray_init, agent, caplog):
 
 
 def test_rllib_torch(ray_init, agent, caplog):
-
     trainer_fn, config, needs_q_model = agent
     if needs_q_model:
         pytest.skip("DQN in torch not currently working")

tests/test_tsp.py

@@ -18,7 +18,6 @@ def G(N):
 
 
 def test_graphenv():
-
     env = GraphEnv(
         {
             "state": TSPState(
@@ -33,7 +32,6 @@ def test_graphenv():
 
 
 def test_rllib_base(ray_init, agent, N, G):
-
     trainer_fn, config, needs_q_model = agent
     model = TSPQModel if needs_q_model else TSPModel