When I want to execute Tensorboard with keras-rl (DQNAgent):
tb_callback = TensorBoard('/home/jose/TED/MLU_minimization/logs', update_freq=1)
dqn.fit(env, nb_steps=5000000, visualize=False, verbose=1, nb_max_episode_steps=None, log_interval=10000, callbacks=[tb_callback])
I obtain the next error:
AttributeError: 'TensorBoard' object has no attribute '_should_trace'. Did you mean: '_stop_trace'?
After the next:
Traceback (most recent call last): File "/home/jose/TED/MLU_minimization/DRL.py", line 303, in dqn.fit(env, nb_steps=5000000, visualize=False, verbose=1, nb_max_episode_steps=None, log_interval=10000, callbacks=[tb_callback])
File "/home/jose/TED/MLU_minimization/envDRL/lib/python3.10/site-packages/rl/core.py", line 100, in fit callbacks = CallbackList(callbacks)
File "/home/jose/TED/MLU_minimization/envDRL/lib/python3.10/site-packages/tensorflow/python/keras/callbacks.py", line 239, in init self._batch_hooks_support_tf_logs = all(
File "/home/jose/TED/MLU_minimization/envDRL/lib/python3.10/site-packages/tensorflow/python/keras/callbacks.py", line 242, in if cb._implements_train_batch_hooks() or cb
File "/home/jose/TED/MLU_minimization/envDRL/lib/python3.10/site-packages/keras/callbacks.py", line 2770, in _implements_train_batch_hooks return self._should_trace or self.write_steps_per_second
