def __init__(
self,
size,
obs_shape,
act_shape,
pi_info_shapes,
gamma=0.99,
lam=0.95,
cost_gamma=0.99,
cost_lam=0.95,
):
self.obs_buf = np.zeros(combined_shape(size, obs_shape),
dtype=np.float32)
self.act_buf = np.zeros(combined_shape(size, act_shape),
dtype=np.float32)
self.adv_buf = np.zeros(size, dtype=np.float32)
self.rew_buf = np.zeros(size, dtype=np.float32)
self.ret_buf = np.zeros(size, dtype=np.float32)
self.val_buf = np.zeros(size, dtype=np.float32)
self.cadv_buf = np.zeros(size, dtype=np.float32) # cost advantage
self.cost_buf = np.zeros(size, dtype=np.float32) # costs
self.cret_buf = np.zeros(size, dtype=np.float32) # cost return
self.cval_buf = np.zeros(size, dtype=np.float32) # cost value
self.logp_buf = np.zeros(size, dtype=np.float32)
self.pi_info_bufs = {
k: np.zeros([size] + list(v), dtype=np.float32)
for k, v in pi_info_shapes.items()
}
self.sorted_pi_info_keys = keys_as_sorted_list(self.pi_info_bufs)
self.gamma, self.lam = gamma, lam
self.cost_gamma, self.cost_lam = cost_gamma, cost_lam
self.ptr, self.path_start_idx, self.max_size = 0, 0, size
def placeholder(dim=None):
return tf.placeholder(dtype=tf.float32, shape=combined_shape(None, dim))