dtiarks / ddpg_agent

Tensorflow implementation of the DDPG algorithm

Tensorflow implementation of the DDPG algorithm

This is a reimplementation of the work presented in

Paper

Continuous control with deep reinforcement learning https://arxiv.org/abs/1509.02971

Some examples:

• Hopper environment

• Walker environment

• Lunar Lander environment

ddpg_agent's Issues

AttributeError: 'DDPGAgent' object has no attribute 'train_writer'

When i run ddpg_agent.py i get an error:
2019-03-27 01:12:47.773075: I tensorflow/core/platform/cpu_feature_guard.cc:141] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2
2019-03-27 01:12:47.776984: I tensorflow/core/platform/profile_utils/cpu_utils.cc:94] CPU Frequency: 2904000000 Hz
2019-03-27 01:12:47.777137: I tensorflow/compiler/xla/service/service.cc:150] XLA service 0x555af9008f00 executing computations on platform Host. Devices:
2019-03-27 01:12:47.777191: I tensorflow/compiler/xla/service/service.cc:158] StreamExecutor device (0): ,
WARNING:tensorflow:From /home/greta/anaconda3/envs/ddpg/lib/python3.7/site-packages/tensorflow/python/framework/op_def_library.py:263: colocate_with (from tensorflow.python.framework.ops) is deprecated and will be removed in a future version.
Instructions for updating:
Colocations handled automatically by placer.
WARNING:tensorflow:From ddpg_agent.py:151: to_float (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version.
Instructions for updating:
Use tf.cast instead.
Traceback (most recent call last):
File "ddpg_agent.py", line 360, in
ddpga=DDPGAgent(sess,env,params)
File "ddpg_agent.py", line 76, in init
os.mkdir(self.traindir)
FileNotFoundError: [Errno 2] No such file or directory: './train_dir/run_270319_011248'
Exception ignored in: <function DDPGAgent.del at 0x7fc5a0973950>
Traceback (most recent call last):
File "ddpg_agent.py", line 104, in del
self.train_writer.close()
AttributeError: 'DDPGAgent' object has no attribute 'train_writer'
Can't figure out what is the issue.