ready to merge with dev_1.10.2

.gitignore

@@ -403,6 +403,11 @@ grid2op/tests/requirements.txt
 grid2op/tests/venv_test_311/
 issue_577/
 junk.py
+grid2op/tests/20240429_failed_tests.txt
+grid2op/tests/20240429_failed_tests_small.txt
+grid2op/tests/20240429_teq_test.txt
+grid2op/tests/req_38_np121
+test_make_2_envs.py
 
 # profiling files
 **.prof

grid2op/Action/baseAction.py

@@ -6435,7 +6435,6 @@ def decompose_as_unary_actions(self,
                     tmp += a
             assert tmp == act
         
-        
         Parameters
         ----------
         group_topo : bool, optional
@@ -6499,12 +6498,15 @@ def decompose_as_unary_actions(self,
             self._aux_decompose_as_unary_actions_curtail(cls, group_curtail, res)
         return res
 
-    def _add_act_and_remove_line_status_only_set(self, other: "BaseAction"):
+    def _add_act_and_remove_line_status_only_set(self, other: "BaseAction") -> "BaseAction":
         """INTERNAL
         
         This is used by the environment when combining action in the "set state" in env.reset.
         
         It supposes both self and other are only "set" actions
+        
+        .. versionadded:: 1.10.2
+        
         """
         self += other
         cls = type(self)
@@ -6529,13 +6531,16 @@ def _add_act_and_remove_line_status_only_set(self, other: "BaseAction"):
             self._modif_set_status = True
         if (self._set_topo_vect != 0).any():
             self._modif_set_bus = True
+        return self
 
     def remove_change(self) -> "BaseAction":
         """This function will modify 'self' and remove all "change" action type. 
         
         It is mainly used in the environment, when removing the "change" type for setting the original
         state of the grid.
         
+        .. versionadded:: 1.10.2
+        
         """
         if self._change_bus_vect.any():
             warnings.warn("This action modified the buses with `change_bus` ")
@@ -6544,4 +6549,4 @@ def remove_change(self) -> "BaseAction":
         if self._switch_line_status.any():
             self._switch_line_status[:] = False
             self._modif_change_status = False
-        return self
+        return self

grid2op/tests/test_RunnerFast.py

@@ -0,0 +1,125 @@
+# Copyright (c) 2019-2020, RTE (https://www.rte-france.com)
+# See AUTHORS.txt
+# This Source Code Form is subject to the terms of the Mozilla Public License, version 2.0.
+# If a copy of the Mozilla Public License, version 2.0 was not distributed with this file,
+# you can obtain one at http://mozilla.org/MPL/2.0/.
+# SPDX-License-Identifier: MPL-2.0
+# This file is part of Grid2Op, Grid2Op a testbed platform to model sequential decision making in power systems.
+
+import warnings
+import unittest
+
+from grid2op.tests.helper_path_test import *
+
+PATH_ADN_CHRONICS_FOLDER = os.path.abspath(
+    os.path.join(PATH_CHRONICS, "test_multi_chronics")
+)
+PATH_PREVIOUS_RUNNER = os.path.join(data_test_dir, "runner_data")
+
+import grid2op
+from grid2op.Runner import Runner
+from grid2op.dtypes import dt_float
+
+warnings.simplefilter("error")
+
+
+class TestRunner(HelperTests, unittest.TestCase):
+    def setUp(self):
+        super().setUp()
+        self.init_grid_path = os.path.join(PATH_DATA_TEST_PP, "test_case14.json")
+        self.path_chron = PATH_ADN_CHRONICS_FOLDER
+        self.parameters_path = None
+        self.max_iter = 10
+        self.real_reward = dt_float(7748.425 / 12.)
+        self.real_reward_li = [self.real_reward, dt_float(7786.8955 / 12.)]  # 7786.89599609375
+
+        self.all_real_rewards = [
+            dt_float(el / 12.)
+            for el in [
+                761.3295,
+                768.10144,
+                770.2673,
+                767.767,
+                768.69,
+                768.71246,
+                779.1029,
+                783.2737,
+                788.7833,
+                792.39764,
+            ]
+        ]
+        with warnings.catch_warnings():
+            warnings.filterwarnings("ignore")
+            self.env = grid2op.make("l2rpn_case14_sandbox", test=True, _add_to_name=type(self).__name__)
+        self.runner = Runner(**self.env.get_params_for_runner())
+
+    def test_one_episode(self):
+        with warnings.catch_warnings():
+            warnings.filterwarnings("ignore")
+            _, cum_reward, timestep, max_ts = self.runner.run_one_episode(
+            max_iter=self.max_iter
+        )
+        assert int(timestep) == self.max_iter
+        assert np.abs(cum_reward - self.real_reward) <= self.tol_one, f"{cum_reward} != {self.real_reward}"
+
+    def test_one_episode_detailed(self):
+        with warnings.catch_warnings():
+            warnings.filterwarnings("ignore")
+            _, cum_reward, timestep, max_ts, episode_data = self.runner.run_one_episode(
+            max_iter=self.max_iter, detailed_output=True
+        )
+        assert int(timestep) == self.max_iter
+        assert np.abs(cum_reward - self.real_reward) <= self.tol_one
+        for j in range(len(self.all_real_rewards)):
+            assert (
+                np.abs(episode_data.rewards[j] - self.all_real_rewards[j])
+                <= self.tol_one
+            ), f"{episode_data.rewards[j]} != {self.all_real_rewards[j]}"
+
+    def test_2episode(self):
+        with warnings.catch_warnings():
+            warnings.filterwarnings("ignore")
+            res = self.runner._run_sequential(nb_episode=2, max_iter=self.max_iter)
+        assert len(res) == 2
+        for i, (stuff, _, cum_reward, timestep, total_ts) in enumerate(res):
+            assert int(timestep) == self.max_iter
+            assert np.abs(cum_reward - self.real_reward_li[i]) <= self.tol_one, f"for iter {i}: {cum_reward} != {self.real_reward_li[i]}"
+
+    def test_init_from_env(self):
+        with warnings.catch_warnings():
+            warnings.filterwarnings("ignore")
+            with grid2op.make("rte_case14_test", test=True, _add_to_name=type(self).__name__) as env:
+                runner = Runner(**env.get_params_for_runner())
+                res = runner.run(nb_episode=1, max_iter=self.max_iter)
+        for i, _, cum_reward, timestep, total_ts in res:
+            assert int(timestep) == self.max_iter, f"{timestep} != {self.max_iter}"
+
+    def test_seed_seq(self):
+        with warnings.catch_warnings():
+            warnings.filterwarnings("ignore")
+            with grid2op.make("rte_case14_test", test=True, _add_to_name=type(self).__name__) as env:
+                runner = Runner(**env.get_params_for_runner())
+                res = runner.run(
+                    nb_episode=1, max_iter=self.max_iter, env_seeds=[1], agent_seeds=[2]
+                )
+        for i, _, cum_reward, timestep, total_ts in res:
+            assert int(timestep) == self.max_iter, f"{timestep} != {self.max_iter}"
+
+    def test_seed_par(self):
+        with warnings.catch_warnings():
+            warnings.filterwarnings("ignore")
+            with grid2op.make("rte_case14_test", test=True, _add_to_name=type(self).__name__) as env:
+                runner = Runner(**env.get_params_for_runner())
+                res = runner.run(
+                    nb_episode=2,
+                    nb_process=2,
+                    max_iter=self.max_iter,
+                    env_seeds=[1, 2],
+                    agent_seeds=[3, 4],
+                )
+        for i, _, cum_reward, timestep, total_ts in res:
+            assert int(timestep) == self.max_iter
+
+
+if __name__ == "__main__":
+    unittest.main()

grid2op/tests/test_action_set_orig_state.py

@@ -41,8 +41,6 @@
 
 
 # TODO test "change" is deactivated
-# TODO test grid2Op compat mode (storage units)
-# TODO test with redispatching and curtailment actions
 # TODO test with "names_orig_to_backend"