From 99f1f7cfc149b5e5d8a59796b5f1ada27fd0ae64 Mon Sep 17 00:00:00 2001
From: Czar Echavez <Czar.Echavez@envitia.com>
Date: Thu, 6 Jul 2023 12:10:26 +0100
Subject: [PATCH] #1386: remove setting of global seed + running pre-commit
 checks

---
 src/primaite/agents/agent.py                  |  1 -
 src/primaite/agents/sb3.py                    |  7 ++----
 src/primaite/environment/primaite_env.py      |  6 -----
 .../test_seeding_and_deterministic_session.py | 22 ++++++-------------
 4 files changed, 9 insertions(+), 27 deletions(-)

diff --git a/src/primaite/agents/agent.py b/src/primaite/agents/agent.py
index 5b76c36b..4b39839a 100644
--- a/src/primaite/agents/agent.py
+++ b/src/primaite/agents/agent.py
@@ -248,7 +248,6 @@ class AgentSessionABC(ABC):
 
             agent.session_path = path
 
-
             return agent
 
         else:
diff --git a/src/primaite/agents/sb3.py b/src/primaite/agents/sb3.py
index 0d031c10..9d295c6f 100644
--- a/src/primaite/agents/sb3.py
+++ b/src/primaite/agents/sb3.py
@@ -59,7 +59,7 @@ class SB3Agent(AgentSessionABC):
             verbose=self.sb3_output_verbose_level,
             n_steps=self._training_config.num_steps,
             tensorboard_log=str(self._tensorboard_log_path),
-            seed=self._training_config.seed
+            seed=self._training_config.seed,
         )
 
     def _save_checkpoint(self):
@@ -121,10 +121,7 @@ class SB3Agent(AgentSessionABC):
             obs = self._env.reset()
 
             for step in range(time_steps):
-                action, _states = self._agent.predict(
-                    obs,
-                    deterministic=self._training_config.deterministic
-                )
+                action, _states = self._agent.predict(obs, deterministic=self._training_config.deterministic)
                 if isinstance(action, np.ndarray):
                     action = np.int64(action)
                 obs, rewards, done, info = self._env.step(action)
diff --git a/src/primaite/environment/primaite_env.py b/src/primaite/environment/primaite_env.py
index cd2ff9e5..03c23f93 100644
--- a/src/primaite/environment/primaite_env.py
+++ b/src/primaite/environment/primaite_env.py
@@ -2,7 +2,6 @@
 """Main environment module containing the PRIMmary AI Training Evironment (Primaite) class."""
 import copy
 import logging
-import random
 import uuid as uuid
 from pathlib import Path
 from random import choice, randint, sample, uniform
@@ -241,11 +240,6 @@ class Primaite(Env):
         self.episode_av_reward_writer = SessionOutputWriter(self, transaction_writer=False, learning_session=True)
         self.transaction_writer = SessionOutputWriter(self, transaction_writer=True, learning_session=True)
 
-        # set the seed globally if there is one
-        if self.training_config.seed:
-            random.seed(self.training_config.seed)
-            np.random.seed(self.training_config.seed)
-
     @property
     def actual_episode_count(self) -> int:
         """Shifts the episode_count by -1 for RLlib."""
diff --git a/tests/test_seeding_and_deterministic_session.py b/tests/test_seeding_and_deterministic_session.py
index 0e420459..34cb43fb 100644
--- a/tests/test_seeding_and_deterministic_session.py
+++ b/tests/test_seeding_and_deterministic_session.py
@@ -6,15 +6,11 @@ from tests import TEST_CONFIG_ROOT
 
 @pytest.mark.parametrize(
     "temp_primaite_session",
-    [[
-        TEST_CONFIG_ROOT / "ppo_seeded_training_config.yaml",
-        dos_very_basic_config_path()
-    ]],
+    [[TEST_CONFIG_ROOT / "ppo_seeded_training_config.yaml", dos_very_basic_config_path()]],
     indirect=True,
 )
 def test_seeded_learning(temp_primaite_session):
     """Test running seeded learning produces the same output when ran twice."""
-
     expected_mean_reward_per_episode = {
         1: -90.703125,
         2: -91.15234375,
@@ -25,26 +21,22 @@ def test_seeded_learning(temp_primaite_session):
         7: -88.984375,
         8: -88.3203125,
         9: -112.79296875,
-        10: -100.01953125
+        10: -100.01953125,
     }
     with temp_primaite_session as session:
-        assert session._training_config.seed == 67890, \
-            "Expected output is based upon a agent that was trained with " \
-            "seed 67890"
+        assert session._training_config.seed == 67890, (
+            "Expected output is based upon a agent that was trained with " "seed 67890"
+        )
         session.learn()
         actual_mean_reward_per_episode = session.learn_av_reward_per_episode()
 
     assert actual_mean_reward_per_episode == expected_mean_reward_per_episode
 
 
-@pytest.mark.skip(reason="Inconsistent results. Needs someone with RL "
-                         "knowledge to investigate further.")
+@pytest.mark.skip(reason="Inconsistent results. Needs someone with RL " "knowledge to investigate further.")
 @pytest.mark.parametrize(
     "temp_primaite_session",
-    [[
-        TEST_CONFIG_ROOT / "ppo_seeded_training_config.yaml",
-        dos_very_basic_config_path()
-    ]],
+    [[TEST_CONFIG_ROOT / "ppo_seeded_training_config.yaml", dos_very_basic_config_path()]],
     indirect=True,
 )
 def test_deterministic_evaluation(temp_primaite_session):