diff --git a/torchy_baselines/common/base_class.py b/torchy_baselines/common/base_class.py
index 597072d..00d8093 100644
--- a/torchy_baselines/common/base_class.py
+++ b/torchy_baselines/common/base_class.py
@@ -207,7 +207,7 @@ class BaseRLModel(object):
         """
         raise NotImplementedError()
 
-    def seed(self, seed=0):
+    def set_random_seed(self, seed=0):
         set_random_seed(seed, using_cuda=self.device == th.device('cuda'))
         self.action_space.seed(seed)
         if self.env is not None:
diff --git a/torchy_baselines/ppo/ppo.py b/torchy_baselines/ppo/ppo.py
index 76a6a73..5ef0793 100644
--- a/torchy_baselines/ppo/ppo.py
+++ b/torchy_baselines/ppo/ppo.py
@@ -40,7 +40,7 @@ class PPO(BaseRLModel):
                                   verbose, device, create_eval_env=create_eval_env, support_multi_env=True)
 
         self.learning_rate = learning_rate
-        self._seed = seed
+        self.seed = seed
         self.batch_size = batch_size
         self.n_optim = n_optim
         self.n_steps = n_steps
@@ -63,7 +63,7 @@ class PPO(BaseRLModel):
         state_dim, action_dim = self.observation_space.shape[0], self.action_space.shape[0]
         # TODO: different seed for each env when n_envs > 1
         if self.n_envs == 1:
-            self.seed(self._seed)
+            self.set_random_seed(self.seed)
 
         self.rollout_buffer = RolloutBuffer(self.n_steps, state_dim, action_dim, self.device,
                                             gamma=self.gamma, gae_lambda=self.gae_lambda, n_envs=self.n_envs)
diff --git a/torchy_baselines/sac/sac.py b/torchy_baselines/sac/sac.py
index 8fb3708..1465e3a 100644
--- a/torchy_baselines/sac/sac.py
+++ b/torchy_baselines/sac/sac.py
@@ -55,7 +55,7 @@ class SAC(BaseRLModel):
                  tau=0.005, ent_coef='auto', target_update_interval=1,
                  gradient_steps=1, target_entropy='auto', action_noise=None,
                  gamma=0.99, action_noise_std=0.0, create_eval_env=False,
-                 policy_kwargs=None, verbose=0, seed=0,
+                 policy_kwargs=None, verbose=0, seed=0, device='auto',
                  _init_setup_model=True):
 
         super(SAC, self).__init__(policy, env, SACPolicy, policy_kwargs, verbose, device,
@@ -64,7 +64,7 @@ class SAC(BaseRLModel):
         self.max_action = np.abs(self.action_space.high)
         self.action_noise_std = action_noise_std
         self.learning_rate = learning_rate
-        self._seed = seed
+        self.seed = seed
         self.target_entropy = target_entropy
         self.log_ent_coef = None
         # self.target_update_interval = target_update_interval
@@ -89,7 +89,7 @@ class SAC(BaseRLModel):
 
     def _setup_model(self):
         obs_dim, action_dim = self.observation_space.shape[0], self.action_space.shape[0]
-        self.seed(self._seed)
+        self.set_random_seed(self.seed)
 
         # Target entropy is used when learning the entropy coefficient
         if self.target_entropy == 'auto':
diff --git a/torchy_baselines/td3/td3.py b/torchy_baselines/td3/td3.py
index 7ede8e3..34f1051 100644
--- a/torchy_baselines/td3/td3.py
+++ b/torchy_baselines/td3/td3.py
@@ -57,7 +57,7 @@ class TD3(BaseRLModel):
         self.max_action = np.abs(self.action_space.high)
         self.action_noise_std = action_noise_std
         self.buffer_size = buffer_size
-        self._seed = seed
+        self.seed = seed
 
         self.buffer_size = buffer_size
         # TODO: accept callables
@@ -78,7 +78,7 @@ class TD3(BaseRLModel):
 
     def _setup_model(self):
         obs_dim, action_dim = self.observation_space.shape[0], self.action_space.shape[0]
-        self.seed(self._seed)
+        self.set_random_seed(self.seed)
         self.replay_buffer = ReplayBuffer(self.buffer_size, obs_dim, action_dim, self.device)
         self.policy = self.policy(self.observation_space, self.action_space,
                                   self.learning_rate, device=self.device, **self.policy_kwargs)