[CLUSTER] 3x3_len10

manuel-delverme · Nov 19, 2021 · e531f74 · e531f74
1 parent e5eebc5
commit e531f74
Show file tree

Hide file tree

Showing 5 changed files with 50 additions and 60 deletions.
diff --git a/lstmDQN/custom_agent.py b/lstmDQN/custom_agent.py
@@ -14,10 +14,8 @@ def __init__(self, observation_space, action_space, device):
 
         self.action_vocab_size = len(action_space.vocab)
 
-        self.model = LSTMDQN(
-            len(observation_space.vocab), len(action_space.vocab), self.device, action_space.sequence_length,
-            config.embedding_size, config.encoder_rnn_hidden_size, config.action_scorer_hidden_dim,
-        )
+        self.model = LSTMDQN(len(observation_space.vocab), len(action_space.vocab), action_space.sequence_length, config.embedding_size, config.encoder_rnn_hidden_size,
+                             config.action_scorer_hidden_dim, )
 
         # obs_vocab_size, action_vocab_size, device, output_length: int,
         # embedding_size, encoder_rnn_hidden_size, action_scorer_hidden_dim, ):

diff --git a/lstmDQN/model.py b/lstmDQN/model.py
@@ -4,13 +4,10 @@
 class LSTMDQN(torch.nn.Module):
     batch_size = 1
 
-    def __init__(
-            self, obs_vocab_size, action_vocab_size, device, output_length: int,
-            embedding_size, encoder_rnn_hidden_size, action_scorer_hidden_dim, ):
+    def __init__(self, obs_vocab_size, action_vocab_size, output_length: int, embedding_size, encoder_rnn_hidden_size, action_scorer_hidden_dim, ):
         super(LSTMDQN, self).__init__()
-        self.device = device
 
-        self.word_embedding = torch.nn.Embedding(obs_vocab_size, embedding_size, device=self.device)
+        self.word_embedding = torch.nn.Embedding(obs_vocab_size, embedding_size)
         self.encoder = torch.nn.GRU(embedding_size, encoder_rnn_hidden_size, batch_first=True)
         self.Q_features = torch.nn.Sequential(
             torch.nn.Linear(encoder_rnn_hidden_size, action_scorer_hidden_dim),

diff --git a/requirements.txt b/requirements.txt
@@ -3,4 +3,4 @@ numpy
 gym
 stable-baselines3
 icecream
-git+https://github.com/ministry-of-silly-code/experiment_buddy.git@main#egg=experiment_buddy
+experiment_buddy
diff --git a/sweep.yml b/sweep.yml
@@ -8,50 +8,44 @@ parameters:
     distribution: log_uniform
     min: -7.76804212
     max: -4.76804212
-
-^replay_batch_size:
-  distribution: categorical
-  values:
-    - 32
-    - 64
-    - 128
-    - 256
-    - 1024
-    - 2048
-    - 4096
-
-^embedding_size:
-  distribution: categorical
-  values:
-    - 32
-    - 64
-    - 128
-    - 256
-
-^encoder_rnn_hidden_size:
-  distribution: categorical
-  values:
-    - 64
-    - 128
-    - 256
-    - 1024
-
-^action_socrer_hidden_dim:
-  distribution: categorical
-  values:
-    - 64
-    - 128
-    - 256
-
-^update_per_k_game_steps:
-  distribution: q_uniform
-  min: 1
-  max: 10
-  q: 1
-
-^epsilon_anneal_episodes:
-  distribution: q_uniform
-  min: 1000
-  max: 1000000
-  q: 1000
+  ^replay_batch_size:
+    distribution: categorical
+    values:
+      - 32
+      - 64
+      - 128
+      - 256
+      - 1024
+      - 2048
+      - 4096
+  ^embedding_size:
+    distribution: categorical
+    values:
+      - 32
+      - 64
+      - 128
+      - 256
+  ^encoder_rnn_hidden_size:
+    distribution: categorical
+    values:
+      - 64
+      - 128
+      - 256
+      - 1024
+  ^action_socrer_hidden_dim:
+    distribution: categorical
+    values:
+      - 64
+      - 128
+      - 256
+  ^update_per_k_game_steps:
+    distribution: q_uniform
+    min: 1
+    max: 10
+    q: 1
+  ^epsilon_anneal_episodes:
+    distribution: q_uniform
+    min: 1000
+    max: 1000000
+    q: 1000
 program: train.py
diff --git a/train.py b/train.py
@@ -72,9 +72,10 @@ def train(tb):
 
 if __name__ == '__main__':
     experiment_buddy.register_defaults(vars(config))
-    PROC_NUM = 1
+    PROC_NUM = 10
     # HOST = "mila" if config.user == "esac" else ""
-    HOST = ""
-    YAML_FILE = ""  # "sweep.yml"  # "env_suite.yml"
-    tb = experiment_buddy.deploy(host=HOST, sweep_yaml=YAML_FILE, proc_num=PROC_NUM, wandb_kwargs={"mode": "disabled" if config.DEBUG else "online", "entity": "rl-sql"})
+    HOST = "mila"
+    RUN_SWEEP = True
+    tb = experiment_buddy.deploy(host=HOST, sweep_yaml="sweep.yml" if RUN_SWEEP else "", proc_num=PROC_NUM,
+                                 wandb_kwargs={"mode": "disabled" if config.DEBUG else "online", "entity": "rl-sql"})
     train(tb)