Exemple #1
0
    def test_len_after_appending_to_loaded_memories(self):
        with TempDir() as temp_directory:
            replay_memories1_file_name = os.path.join(temp_directory, 'replay_memories.dat')
            replay_memories1 = ReplayMemories(replay_memories1_file_name, max_current_memories_in_ram=100)
            replay_memories1.append(self._create_replay_memory(consequent_reward=34.0))
            replay_memories1.append(self._create_replay_memory(consequent_reward=35.0))
            replay_memories1.save()

            replay_memories2 = ReplayMemories(replay_memories1_file_name, max_current_memories_in_ram=100)
            replay_memories2.append(self._create_replay_memory(consequent_reward=36.0))
            self.assertEquals(len(replay_memories2), 3)
Exemple #2
0
    def test_retrieval_after_appending_to_loaded_memories(self):
        with TempDir() as temp_directory:
            replay_memories1_file_name = os.path.join(temp_directory, 'replay_memories.dat')
            replay_memories1 = ReplayMemories(replay_memories1_file_name, max_current_memories_in_ram=100)
            replay_memory1 = self._create_replay_memory(action_index=11)
            replay_memory2 = self._create_replay_memory(action_index=22)
            replay_memories1.append(replay_memory1)
            replay_memories1.append(replay_memory2)
            replay_memories1.save()

            replay_memories2 = ReplayMemories(replay_memories1_file_name, max_current_memories_in_ram=100)
            replay_memory3 = self._create_replay_memory(action_index=33)
            replay_memories2.append(replay_memory3)
            expected_replay_memories = [replay_memory1, replay_memory2, replay_memory3]
            self.assertSequenceEqual(list(replay_memories2), expected_replay_memories)
Exemple #3
0
 def new_replay_memories(self):
     return ReplayMemories(
         previous_memories_path=self.replay_memories_path(),
         max_current_memories_in_ram=self.generic_hyperparameters.
         MAX_REPLAY_MEMORIES_IN_RAM,
         max_previous_memories_per_file=self.generic_hyperparameters.
         MAX_REPLAY_MEMORIES_PER_FILE)
Exemple #4
0
    def test_prefetch_max_memories_in_ram_less_than_total_memories(self):
        with TempDir() as temp_directory:
            replay_memories1_file_name = os.path.join(temp_directory, 'replay_memories.dat')
            replay_memories1 = ReplayMemories(replay_memories1_file_name, max_current_memories_in_ram=100)
            replay_memory1 = self._create_replay_memory(action_index=11)
            replay_memory2 = self._create_replay_memory(action_index=22)
            replay_memory3 = self._create_replay_memory(action_index=33)
            replay_memories1.append(replay_memory1)
            replay_memories1.append(replay_memory2)
            replay_memories1.append(replay_memory3)
            replay_memories1.save()

            replay_memories2 = ReplayMemories(replay_memories1_file_name, max_current_memories_in_ram=2, prefetch_into_ram=True)

            expected_replay_memories = [replay_memory1, replay_memory2, replay_memory3]
            self.assertSequenceEqual(list(replay_memories2), expected_replay_memories)
Exemple #5
0
    def test_save_after_max_current_memories(self):
        with TempDir() as temp_directory:
            replay_memories_file_name = os.path.join(temp_directory, 'replay_memories.dat')
            replay_memories = ReplayMemories(replay_memories_file_name, max_current_memories_in_ram=2)
            replay_memory1 = self._create_replay_memory(action_index=11)
            replay_memory2 = self._create_replay_memory(action_index=22)
            replay_memory3 = self._create_replay_memory(action_index=33)
            replay_memories.append(replay_memory1)
            replay_memories.append(replay_memory2)
            replay_memories.append(replay_memory3)
            replay_memories.save()

            expected_replay_memories = [replay_memory1, replay_memory2, replay_memory3]
            self.assertSequenceEqual(list(replay_memories), expected_replay_memories)
Exemple #6
0
    def test_prefetch(self):
        with TempDir() as temp_directory:
            replay_memories1_file_name = os.path.join(temp_directory, 'replay_memories.dat')
            replay_memories1 = ReplayMemories(replay_memories1_file_name, max_current_memories_in_ram=100)
            replay_memory1 = self._create_replay_memory(action_index=11)
            replay_memory2 = self._create_replay_memory(action_index=22)
            replay_memory3 = self._create_replay_memory(action_index=33)
            replay_memories1.append(replay_memory1)
            replay_memories1.append(replay_memory2)
            replay_memories1.append(replay_memory3)
            replay_memories1.save()

            replay_memories2 = ReplayMemories(replay_memories1_file_name, max_current_memories_in_ram=100, prefetch_into_ram=True)

        # This block should be idented outside the TempDir block, since we want to test that memories are not fetched from disk
        expected_replay_memories = [replay_memory1, replay_memory2, replay_memory3]
        print(list(replay_memories2))
        print(expected_replay_memories)
        self.assertSequenceEqual(list(replay_memories2), expected_replay_memories)
Exemple #7
0
    def test_retrieval_when_max_current_memories_not_reached(self):
        with TempDir() as temp_directory:
            replay_memories1_file_name = os.path.join(temp_directory, 'replay_memories.dat')
            replay_memories1 = ReplayMemories(replay_memories1_file_name, max_current_memories_in_ram=4)
            replay_memory1 = self._create_replay_memory(action_index=11)
            replay_memory2 = self._create_replay_memory(action_index=22)
            replay_memory3 = self._create_replay_memory(action_index=33)
            replay_memories1.append(replay_memory1)
            replay_memories1.append(replay_memory2)
            replay_memories1.append(replay_memory3)

            replay_memories2 = ReplayMemories(replay_memories1_file_name, max_current_memories_in_ram=100)
            expected_replay_memories = []
            self.assertSequenceEqual(list(replay_memories2), expected_replay_memories)
Exemple #8
0
    def test_sample_loaded_memories(self):
        with TempDir() as temp_directory:
            replay_memories_file_name = os.path.join(temp_directory, 'replay_memories.dat')
            replay_memories1 = ReplayMemories(replay_memories_file_name, max_current_memories_in_ram=100)
            replay_memory1 = self._create_replay_memory(action_index=11)
            replay_memory2 = self._create_replay_memory(action_index=22)
            replay_memory3 = self._create_replay_memory(action_index=32)
            replay_memory4 = self._create_replay_memory(action_index=42)
            replay_memory5 = self._create_replay_memory(action_index=52)
            replay_memory6 = self._create_replay_memory(action_index=62)
            replay_memory7 = self._create_replay_memory(action_index=72)
            replay_memories1.append(replay_memory1)
            replay_memories1.append(replay_memory2)
            replay_memories1.append(replay_memory3)
            replay_memories1.append(replay_memory4)
            replay_memories1.append(replay_memory5)
            replay_memories1.append(replay_memory6)
            replay_memories1.append(replay_memory7)
            replay_memories1.save()

            replay_memories2 = ReplayMemories(replay_memories_file_name, max_current_memories_in_ram=100)
            sampled_replay_memories = replay_memories2.sample(5, seed=3)

            expected_replay_memories = [replay_memory2, replay_memory4, replay_memory7, replay_memory3, replay_memory5]
            self.assertItemsEqual(sampled_replay_memories, expected_replay_memories)
Exemple #9
0
    def test_sample_size_smaller_than_recent_memories(self):
        with TempDir() as temp_directory:
            replay_memories_file_name = os.path.join(temp_directory, 'replay_memories.dat')
            replay_memories1 = ReplayMemories(replay_memories_file_name, max_current_memories_in_ram=100)
            replay_memory1 = self._create_replay_memory(action_index=11)
            replay_memory2 = self._create_replay_memory(action_index=22)
            replay_memory3 = self._create_replay_memory(action_index=32)
            replay_memory4 = self._create_replay_memory(action_index=42)
            replay_memories1.append(replay_memory1)
            replay_memories1.append(replay_memory2)
            replay_memories1.append(replay_memory3)
            replay_memories1.save()

            replay_memories2 = ReplayMemories(replay_memories_file_name, max_current_memories_in_ram=100)
            replay_memories2.append(replay_memory4)

            sampled_replay_memories = replay_memories2.sample(2, recent_memories_span=3, seed=4)

            expected_replay_memories = [replay_memory2, replay_memory4]
            self.assertItemsEqual(sampled_replay_memories, expected_replay_memories)
Exemple #10
0
    def test_sample_recent_memories_after_appending_to_loaded_memories(self):
        with TempDir() as temp_directory:
            replay_memories_file_name = os.path.join(temp_directory, 'replay_memories.dat')
            replay_memories1 = ReplayMemories(replay_memories_file_name, max_current_memories_in_ram=100)
            replay_memory1 = self._create_replay_memory(action_index=11)
            replay_memory2 = self._create_replay_memory(action_index=22)
            replay_memory3 = self._create_replay_memory(action_index=32)
            replay_memory4 = self._create_replay_memory(action_index=42)
            replay_memory5 = self._create_replay_memory(action_index=52)
            replay_memory6 = self._create_replay_memory(action_index=62)
            replay_memory7 = self._create_replay_memory(action_index=72)
            replay_memories1.append(replay_memory1)
            replay_memories1.append(replay_memory2)
            replay_memories1.append(replay_memory3)
            replay_memories1.save()

            replay_memories2 = ReplayMemories(replay_memories_file_name, max_current_memories_in_ram=100)
            replay_memories2.append(replay_memory4)
            replay_memories2.append(replay_memory5)
            replay_memories2.append(replay_memory6)
            replay_memories2.append(replay_memory7)

            sampled_replay_memories = replay_memories2.sample(5, recent_memories_span=5)

            self.assertTrue(replay_memory3 in sampled_replay_memories)
            self.assertTrue(replay_memory4 in sampled_replay_memories)
            self.assertTrue(replay_memory5 in sampled_replay_memories)
            self.assertTrue(replay_memory6 in sampled_replay_memories)
            self.assertTrue(replay_memory7 in sampled_replay_memories)
Exemple #11
0
    def test_sample_recent_memories(self):
        with TempDir() as temp_directory:
            replay_memories_file_name = os.path.join(temp_directory, 'replay_memories.dat')
            replay_memories = ReplayMemories(replay_memories_file_name, max_current_memories_in_ram=100)
            replay_memory1 = self._create_replay_memory(action_index=11)
            replay_memory2 = self._create_replay_memory(action_index=22)
            replay_memory3 = self._create_replay_memory(action_index=32)
            replay_memory4 = self._create_replay_memory(action_index=42)
            replay_memory5 = self._create_replay_memory(action_index=52)
            replay_memory6 = self._create_replay_memory(action_index=62)
            replay_memory7 = self._create_replay_memory(action_index=72)
            replay_memories.append(replay_memory1)
            replay_memories.append(replay_memory2)
            replay_memories.append(replay_memory3)
            replay_memories.append(replay_memory4)
            replay_memories.append(replay_memory5)
            replay_memories.append(replay_memory6)
            replay_memories.append(replay_memory7)

            sampled_replay_memories = replay_memories.sample(2, recent_memories_span=2)

            print(sampled_replay_memories)

            self.assertTrue(replay_memory6 in sampled_replay_memories)
            self.assertTrue(replay_memory7 in sampled_replay_memories)
Exemple #12
0
 def test_len(self):
     replay_memories = ReplayMemories("non_existing_previous_memories.dat", max_current_memories_in_ram=100)
     replay_memories.append(self._create_replay_memory())
     replay_memories.append(self._create_replay_memory())
     self.assertEquals(len(replay_memories), 2)