def save(self):
"""Save."""
state_dict = {
'pi': self.pi.state_dict(),
'qf1': self.qf1.state_dict(),
'qf2': self.qf2.state_dict(),
'vf': self.vf.state_dict(),
'opt_pi': self.opt_pi.state_dict(),
'opt_qf1': self.opt_qf1.state_dict(),
'opt_qf2': self.opt_qf2.state_dict(),
'opt_vf': self.opt_vf.state_dict(),
'log_alpha': (self.log_alpha if self.automatic_entropy_tuning
else None),
'opt_alpha': (self.opt_alpha.state_dict()
if self.automatic_entropy_tuning else None),
'env': misc.env_state_dict(self.env),
't': self.t
}
buffer_dict = self.buffer.state_dict()
state_dict['buffer_format'] = nest.get_structure(buffer_dict)
self.ckptr.save(state_dict, self.t)
# save buffer seperately and only once (because it can be huge)
np.savez(os.path.join(self.ckptr.ckptdir, 'buffer.npz'),
**{f'{i:04d}': x for i, x in
enumerate(nest.flatten(buffer_dict))})
def save(self):
"""State dict."""
state_dict = {
'pi': self.pi.state_dict(),
'opt': self.opt.state_dict(),
'env': misc.env_state_dict(self.env),
't': self.t
}
self.ckptr.save(state_dict, self.t)
def save(self):
"""State dict."""
state_dict = {
'pi': self.pi.state_dict(),
'opt': self.opt.state_dict(),
'lambda_': self.log_lambda_,
'opt_l': self.opt_l.state_dict(),
'env': misc.env_state_dict(self.env),
'_actor': self._actor.state_dict(),
't': self.t
}
self.ckptr.save(state_dict, self.t)
def save(self):
"""State dict."""
state_dict = {
'pi': self.pi.state_dict(),
'vf': self.vf.state_dict(),
'opt_pi': self.opt_pi.state_dict(),
'opt_vf': self.opt_vf.state_dict(),
'kl_weight': self.kl_weight,
'env': misc.env_state_dict(self.env),
'_actor': self._actor.state_dict(),
't': self.t
}
self.ckptr.save(state_dict, self.t)
def save(self):
"""Save."""
state_dict = {
'qf': self.qf.state_dict(),
'qf_targ': self.qf.state_dict(),
'opt': self.opt.state_dict(),
'_actor': self._actor.state_dict(),
'env': misc.env_state_dict(self.env),
't': self.t
}
buffer_dict = self.buffer.state_dict()
state_dict['buffer_format'] = nest.get_structure(buffer_dict)
self.ckptr.save(state_dict, self.t)
# save buffer seperately and only once (because it can be huge)
np.savez(os.path.join(self.ckptr.ckptdir, 'buffer.npz'),
**{f'{i:04d}': x for i, x in
enumerate(nest.flatten(buffer_dict))})