File size: 9,032 Bytes
079c32c |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 |
from typing import List, Dict, Any, Tuple, Union
from collections import namedtuple
import torch
import torch.nn as nn
from ding.torch_utils import Adam, to_device
from ding.model import model_wrap
from ding.utils import POLICY_REGISTRY
from ding.utils.data import default_collate, default_decollate
from .base_policy import Policy
FootballKaggle5thPlaceModel = None
@POLICY_REGISTRY.register('IL')
class ILPolicy(Policy):
r"""
Overview:
Policy class of Imitation learning algorithm
Interface:
__init__, set_setting, __repr__, state_dict_handle
Property:
learn_mode, collect_mode, eval_mode
"""
config = dict(
type='IL',
cuda=True,
# (bool) whether use on-policy training pipeline(behaviour policy and training policy are the same)
on_policy=False,
priority=False,
# (bool) Whether use Importance Sampling Weight to correct biased update. If True, priority must be True.
priority_IS_weight=False,
learn=dict(
# (int) collect n_episode data, train model n_iteration time
update_per_collect=20,
# (int) the number of data for a train iteration
batch_size=64,
# (float) gradient-descent step size
learning_rate=0.0002,
),
collect=dict(
# (int) collect n_sample data, train model n_iteration time
# n_sample=128,
# (float) discount factor for future reward, defaults int [0, 1]
discount_factor=0.99,
),
eval=dict(evaluator=dict(eval_freq=800, ), ),
other=dict(
replay_buffer=dict(
replay_buffer_size=100000,
# (int) max use count of data, if count is bigger than this value,
# the data will be removed from buffer
max_reuse=10,
),
command=dict(),
),
)
# TODO different collect model and learn model
def default_model(self) -> Tuple[str, List[str]]:
return 'football_iql', ['dizoo.gfootball.model.iql.iql_network']
def _init_learn(self) -> None:
r"""
Overview:
Learn mode init method. Called by ``self.__init__``.
Init optimizers, algorithm config, main and target models.
"""
# actor and critic optimizer
self._optimizer = Adam(self._model.parameters(), lr=self._cfg.learn.learning_rate)
# main and target models
self._learn_model = model_wrap(self._model, wrapper_name='base')
self._learn_model.train()
self._learn_model.reset()
self._forward_learn_cnt = 0 # count iterations
def _forward_learn(self, data: dict) -> Dict[str, Any]:
r"""
Overview:
Forward and backward function of learn mode.
Arguments:
- data (:obj:`dict`): Dict type data, including at least ['obs', 'action', 'reward', 'next_obs']
Returns:
- info_dict (:obj:`Dict[str, Any]`): Including at least actor and critic lr, different losses.
"""
data = default_collate(data, cat_1dim=False)
data['done'] = None
if self._cuda:
data = to_device(data, self._device)
loss_dict = {}
# ====================
# imitation learn forward
# ====================
obs = data.get('obs')
logit = data.get('logit')
assert isinstance(obs['processed_obs'], torch.Tensor), obs['processed_obs']
model_action_logit = self._learn_model.forward(obs['processed_obs'])['logit']
supervised_loss = nn.MSELoss(reduction='none')(model_action_logit, logit).mean()
self._optimizer.zero_grad()
supervised_loss.backward()
self._optimizer.step()
loss_dict['supervised_loss'] = supervised_loss
return {
'cur_lr': self._optimizer.defaults['lr'],
**loss_dict,
}
def _state_dict_learn(self) -> Dict[str, Any]:
return {
'model': self._learn_model.state_dict(),
'optimizer': self._optimizer.state_dict(),
}
def _load_state_dict_learn(self, state_dict: Dict[str, Any]) -> None:
self._learn_model.load_state_dict(state_dict['model'])
self._optimizer.load_state_dict(state_dict['optimizer'])
def _init_collect(self) -> None:
r"""
Overview:
Collect mode init method. Called by ``self.__init__``.
Init traj and unroll length, collect model.
"""
self._collect_model = model_wrap(FootballKaggle5thPlaceModel(), wrapper_name='base')
self._gamma = self._cfg.collect.discount_factor
self._collect_model.eval()
self._collect_model.reset()
def _forward_collect(self, data: dict) -> dict:
r"""
Overview:
Forward function of collect mode.
Arguments:
- data (:obj:`Dict[str, Any]`): Dict type data, stacked env data for predicting policy_output(action), \
values are torch.Tensor or np.ndarray or dict/list combinations, keys are env_id indicated by integer.
Returns:
- output (:obj:`Dict[int, Any]`): Dict type data, including at least inferred action according to input obs.
ReturnsKeys
- necessary: ``action``
- optional: ``logit``
"""
data_id = list(data.keys())
data = default_collate(list(data.values()))
if self._cuda:
data = to_device(data, self._device)
with torch.no_grad():
output = self._collect_model.forward(default_decollate(data['obs']['raw_obs']))
if self._cuda:
output = to_device(output, 'cpu')
output = default_decollate(output)
return {i: d for i, d in zip(data_id, output)}
def _process_transition(self, obs: Any, model_output: dict, timestep: namedtuple) -> Dict[str, Any]:
r"""
Overview:
Generate dict type transition data from inputs.
Arguments:
- obs (:obj:`Any`): Env observation
- model_output (:obj:`dict`): Output of collect model, including at least ['action']
- timestep (:obj:`namedtuple`): Output after env step, including at least ['obs', 'reward', 'done'] \
(here 'obs' indicates obs after env step, i.e. next_obs).
Return:
- transition (:obj:`Dict[str, Any]`): Dict type transition data.
"""
transition = {
'obs': obs,
'action': model_output['action'],
'logit': model_output['logit'],
'reward': timestep.reward,
'done': timestep.done,
}
return transition
def _get_train_sample(self, origin_data: list) -> Union[None, List[Any]]:
datas = []
pre_rew = 0
for i in range(len(origin_data) - 1, -1, -1):
data = {}
data['obs'] = origin_data[i]['obs']
data['action'] = origin_data[i]['action']
cur_rew = origin_data[i]['reward']
pre_rew = cur_rew + (pre_rew * self._gamma)
# sample uniformly
data['priority'] = 1
data['logit'] = origin_data[i]['logit']
datas.append(data)
return datas
def _init_eval(self) -> None:
r"""
Overview:
Evaluate mode init method. Called by ``self.__init__``.
Init eval model. Unlike learn and collect model, eval model does not need noise.
"""
self._eval_model = model_wrap(self._model, wrapper_name='argmax_sample')
self._eval_model.reset()
def _forward_eval(self, data: dict) -> dict:
r"""
Overview:
Forward function of eval mode, similar to ``self._forward_collect``.
Arguments:
- data (:obj:`Dict[str, Any]`): Dict type data, stacked env data for predicting policy_output(action), \
values are torch.Tensor or np.ndarray or dict/list combinations, keys are env_id indicated by integer.
Returns:
- output (:obj:`Dict[int, Any]`): The dict of predicting action for the interaction with env.
ReturnsKeys
- necessary: ``action``
- optional: ``logit``
"""
data_id = list(data.keys())
data = default_collate(list(data.values()))
if self._cuda:
data = to_device(data, self._device)
with torch.no_grad():
output = self._eval_model.forward(data['obs']['processed_obs'])
if self._cuda:
output = to_device(output, 'cpu')
output = default_decollate(output)
return {i: d for i, d in zip(data_id, output)}
def _monitor_vars_learn(self) -> List[str]:
r"""
Overview:
Return variables' name if variables are to used in monitor.
Returns:
- vars (:obj:`List[str]`): Variables' name list.
"""
return ['cur_lr', 'supervised_loss']
|