search

cs3243-ay1819s2 / PokerAI

Reinforcement-Learning AI that plays Texas Hold'em
1 stars 0 forks source link

Running with options -a 2 has the following error #6

Open jeffkwoh opened 5 years ago

jeffkwoh commented 5 years ago 
Game:20, epsilon:1.0Traceback (most recent call last):
  File "main.py", line 177, in <module>
    new_agent, final_state, winner_counts, n_games_played = run_episode(agents)
  File "main.py", line 134, in run_episode
    agents[0].replay(BATCH_SIZE)
  File "/Users/base/Documents/Github2/PokerAI/agent/dqn_agent.py", line 84, in replay
    target_f = self.model.predict(state)
  File "/usr/local/lib/python3.7/site-packages/keras/engine/training.py", line 1149, in predict
    x, _, _ = self._standardize_user_data(x)
  File "/usr/local/lib/python3.7/site-packages/keras/engine/training.py", line 751, in _standardize_user_data
    exception_prefix='input')
  File "/usr/local/lib/python3.7/site-packages/keras/engine/training_utils.py", line 138, in standardize_input_data
    str(data_shape))
ValueError: Error when checking input: expected simple_rnn_5_input to have shape (1, 134) but got array with shape (1, 128)