dodox/metastable-baselines
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Removed old files (moved into sb3_trl/)

Browse Source
This commit is contained in:
Dominik Moritz Roth 2022-06-17 11:31:25 +02:00
parent 9199536166
commit e363021ed3
13 changed files with 0 additions and 2147 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

15
projections_orig/__init__.py
View File

@ -1,15 +0,0 @@
# Copyright (c) 2021 Robert Bosch GmbH
# Author: Fabian Otto
#
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU Affero General Public License as published
# by the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU Affero General Public License for more details.
#
# You should have received a copy of the GNU Affero General Public License
# along with this program. If not, see <https://www.gnu.org/licenses/>.

374
projections_orig/base_projection_layer.py
View File

@ -1,374 +0,0 @@
# Copyright (c) 2021 Robert Bosch GmbH
# Author: Fabian Otto
#
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU Affero General Public License as published
# by the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU Affero General Public License for more details.
#
# You should have received a copy of the GNU Affero General Public License
# along with this program. If not, see <https://www.gnu.org/licenses/>.
import copy
import math
import torch as ch
from typing import Tuple, Union
from trust_region_projections.models.policy.abstract_gaussian_policy import AbstractGaussianPolicy
from trust_region_projections.utils.network_utils import get_optimizer
from trust_region_projections.utils.projection_utils import gaussian_kl, get_entropy_schedule
from trust_region_projections.utils.torch_utils import generate_minibatches, select_batch, tensorize
def entropy_inequality_projection(policy: AbstractGaussianPolicy, p: Tuple[ch.Tensor, ch.Tensor],
beta: Union[float, ch.Tensor]):
"""
Projects std to satisfy an entropy INEQUALITY constraint.
Args:
policy: policy instance
p: current distribution
beta: target entropy for EACH std or general bound for all stds
Returns:
projected std that satisfies the entropy bound
"""
mean, std = p
k = std.shape[-1]
batch_shape = std.shape[:-2]
ent = policy.entropy(p)
mask = ent < beta
# if nothing has to be projected skip computation
if (~mask).all():
return p
alpha = ch.ones(batch_shape, dtype=std.dtype, device=std.device)
alpha[mask] = ch.exp((beta[mask] - ent[mask]) / k)
proj_std = ch.einsum('ijk,i->ijk', std, alpha)
return mean, ch.where(mask[..., None, None], proj_std, std)
def entropy_equality_projection(policy: AbstractGaussianPolicy, p: Tuple[ch.Tensor, ch.Tensor],
beta: Union[float, ch.Tensor]):
"""
Projects std to satisfy an entropy EQUALITY constraint.
Args:
policy: policy instance
p: current distribution
beta: target entropy for EACH std or general bound for all stds
Returns:
projected std that satisfies the entropy bound
"""
mean, std = p
k = std.shape[-1]
ent = policy.entropy(p)
alpha = ch.exp((beta - ent) / k)
proj_std = ch.einsum('ijk,i->ijk', std, alpha)
return mean, proj_std
def mean_projection(mean: ch.Tensor, old_mean: ch.Tensor, maha: ch.Tensor, eps: ch.Tensor):
"""
Projects the mean based on the Mahalanobis objective and trust region.
Args:
mean: current mean vectors
old_mean: old mean vectors
maha: Mahalanobis distance between the two mean vectors
eps: trust region bound
Returns:
projected mean that satisfies the trust region
"""
batch_shape = mean.shape[:-1]
mask = maha > eps
################################################################################################################
# mean projection maha
# if nothing has to be projected skip computation
if mask.any():
omega = ch.ones(batch_shape, dtype=mean.dtype, device=mean.device)
omega[mask] = ch.sqrt(maha[mask] / eps) - 1.
omega = ch.max(-omega, omega)[..., None]
m = (mean + omega * old_mean) / (1 + omega + 1e-16)
proj_mean = ch.where(mask[..., None], m, mean)
else:
proj_mean = mean
return proj_mean
class BaseProjectionLayer(object):
def __init__(self,
proj_type: str = "",
mean_bound: float = 0.03,
cov_bound: float = 1e-3,
trust_region_coeff: float = 0.0,
scale_prec: bool = True,
entropy_schedule: Union[None, str] = None,
action_dim: Union[None, int] = None,
total_train_steps: Union[None, int] = None,
target_entropy: float = 0.0,
temperature: float = 0.5,
entropy_eq: bool = False,
entropy_first: bool = False,
do_regression: bool = False,
regression_iters: int = 1000,
regression_lr: int = 3e-4,
optimizer_type_reg: str = "adam",
cpu: bool = True,
dtype: ch.dtype = ch.float32,
):
"""
Base projection layer, which can be used to compute metrics for non-projection approaches.
Args:
proj_type: Which type of projection to use. None specifies no projection and uses the TRPO objective.
mean_bound: projection bound for the step size w.r.t. mean
cov_bound: projection bound for the step size w.r.t. covariance matrix
trust_region_coeff: Coefficient for projection regularization loss term.
scale_prec: If true used mahalanobis distance for projections instead of euclidean with Sigma_old^-1.
entropy_schedule: Schedule type for entropy projection, one of 'linear', 'exp', None.
action_dim: number of action dimensions to scale exp decay correctly.
total_train_steps: total number of training steps to compute appropriate decay over time.
target_entropy: projection bound for the entropy of the covariance matrix
temperature: temperature decay for exponential entropy bound
entropy_eq: Use entropy equality constraints.
entropy_first: Project entropy before trust region.
do_regression: Conduct additional regression steps after the the policy steps to match projection and policy.
regression_iters: Number of regression steps.
regression_lr: Regression learning rate.
optimizer_type_reg: Optimizer for regression.
cpu: Compute on CPU only.
dtype: Data type to use, either of float32 or float64. The later might be necessary for higher
dimensions in order to learn the full covariance.
"""
# projection and bounds
self.proj_type = proj_type
self.mean_bound = tensorize(mean_bound, cpu=cpu, dtype=dtype)
self.cov_bound = tensorize(cov_bound, cpu=cpu, dtype=dtype)
self.trust_region_coeff = trust_region_coeff
self.scale_prec = scale_prec
# projection utils
assert (action_dim and total_train_steps) if entropy_schedule else True
self.entropy_proj = entropy_equality_projection if entropy_eq else entropy_inequality_projection
self.entropy_schedule = get_entropy_schedule(entropy_schedule, total_train_steps, dim=action_dim)
self.target_entropy = tensorize(target_entropy, cpu=cpu, dtype=dtype)
self.entropy_first = entropy_first
self.entropy_eq = entropy_eq
self.temperature = temperature
self._initial_entropy = None
# regression
self.do_regression = do_regression
self.regression_iters = regression_iters
self.lr_reg = regression_lr
self.optimizer_type_reg = optimizer_type_reg
def __call__(self, policy, p: Tuple[ch.Tensor, ch.Tensor], q, step, *args, **kwargs):
# entropy_bound = self.policy.entropy(q) - self.target_entropy
entropy_bound = self.entropy_schedule(self.initial_entropy, self.target_entropy, self.temperature,
step) * p[0].new_ones(p[0].shape[0])
return self._projection(policy, p, q, self.mean_bound, self.cov_bound, entropy_bound, **kwargs)
def _trust_region_projection(self, policy: AbstractGaussianPolicy, p: Tuple[ch.Tensor, ch.Tensor],
q: Tuple[ch.Tensor, ch.Tensor], eps: ch.Tensor, eps_cov: ch.Tensor, **kwargs):
"""
Hook for implementing the specific trust region projection
Args:
policy: policy instance
p: current distribution
q: old distribution
eps: mean trust region bound
eps_cov: covariance trust region bound
**kwargs:
Returns:
projected
"""
return p
# @final
def _projection(self, policy: AbstractGaussianPolicy, p: Tuple[ch.Tensor, ch.Tensor],
q: Tuple[ch.Tensor, ch.Tensor], eps: ch.Tensor, eps_cov: ch.Tensor, beta: ch.Tensor, **kwargs):
"""
Template method with hook _trust_region_projection() to encode specific functionality.
(Optional) entropy projection is executed before or after as specified by entropy_first.
Do not override this. For Python >= 3.8 you can use the @final decorator to enforce not overwriting.
Args:
policy: policy instance
p: current distribution
q: old distribution
eps: mean trust region bound
eps_cov: covariance trust region bound
beta: entropy bound
**kwargs:
Returns:
projected mean, projected std
"""
####################################################################################################################
# entropy projection in the beginning
if self.entropy_first:
p = self.entropy_proj(policy, p, beta)
####################################################################################################################
# trust region projection for mean and cov bounds
proj_mean, proj_std = self._trust_region_projection(policy, p, q, eps, eps_cov, **kwargs)
####################################################################################################################
# entropy projection in the end
if self.entropy_first:
return proj_mean, proj_std
return self.entropy_proj(policy, (proj_mean, proj_std), beta)
@property
def initial_entropy(self):
return self._initial_entropy
@initial_entropy.setter
def initial_entropy(self, entropy):
if self.initial_entropy is None:
self._initial_entropy = entropy
def trust_region_value(self, policy, p, q):
"""
Computes the KL divergence between two Gaussian distributions p and q.
Args:
policy: policy instance
p: current distribution
q: old distribution
Returns:
Mean and covariance part of the trust region metric.
"""
return gaussian_kl(policy, p, q)
def get_trust_region_loss(self, policy: AbstractGaussianPolicy, p: Tuple[ch.Tensor, ch.Tensor],
proj_p: Tuple[ch.Tensor, ch.Tensor]):
"""
Compute the trust region loss to ensure policy output and projection stay close.
Args:
policy: policy instance
proj_p: projected distribution
p: predicted distribution from network output
Returns:
trust region loss
"""
p_target = (proj_p[0].detach(), proj_p[1].detach())
mean_diff, cov_diff = self.trust_region_value(policy, p, p_target)
delta_loss = (mean_diff + cov_diff if policy.contextual_std else mean_diff).mean()
return delta_loss * self.trust_region_coeff
def compute_metrics(self, policy, p, q) -> dict:
"""
Returns dict with constraint metrics.
Args:
policy: policy instance
p: current distribution
q: old distribution
Returns:
dict with constraint metrics
"""
with ch.no_grad():
entropy_old = policy.entropy(q)
entropy = policy.entropy(p)
mean_kl, cov_kl = gaussian_kl(policy, p, q)
kl = mean_kl + cov_kl
mean_diff, cov_diff = self.trust_region_value(policy, p, q)
combined_constraint = mean_diff + cov_diff
entropy_diff = entropy_old - entropy
return {'kl': kl.detach().mean(),
'constraint': combined_constraint.mean(),
'mean_constraint': mean_diff.mean(),
'cov_constraint': cov_diff.mean(),
'entropy': entropy.mean(),
'entropy_diff': entropy_diff.mean(),
'kl_max': kl.max(),
'constraint_max': combined_constraint.max(),
'mean_constraint_max': mean_diff.max(),
'cov_constraint_max': cov_diff.max(),
'entropy_max': entropy.max(),
'entropy_diff_max': entropy_diff.max()
}
def trust_region_regression(self, policy: AbstractGaussianPolicy, obs: ch.Tensor, q: Tuple[ch.Tensor, ch.Tensor],
n_minibatches: int, global_steps: int):
"""
Take additional regression steps to match projection output and policy output.
The policy parameters are updated in-place.
Args:
policy: policy instance
obs: collected observations from trajectories
q: old distributions
n_minibatches: split the rollouts into n_minibatches.
global_steps: current number of steps, required for projection
Returns:
dict with mean of regession loss
"""
if not self.do_regression:
return {}
policy_unprojected = copy.deepcopy(policy)
optim_reg = get_optimizer(self.optimizer_type_reg, policy_unprojected.parameters(), learning_rate=self.lr_reg)
optim_reg.reset()
reg_losses = obs.new_tensor(0.)
# get current projected values --> targets for regression
p_flat = policy(obs)
p_target = self(policy, p_flat, q, global_steps)
for _ in range(self.regression_iters):
batch_indices = generate_minibatches(obs.shape[0], n_minibatches)
# Minibatches SGD
for indices in batch_indices:
batch = select_batch(indices, obs, p_target[0], p_target[1])
b_obs, b_target_mean, b_target_std = batch
proj_p = (b_target_mean.detach(), b_target_std.detach())
p = policy_unprojected(b_obs)
# invert scaling with coeff here as we do not have to balance with other losses
loss = self.get_trust_region_loss(policy, p, proj_p) / self.trust_region_coeff
optim_reg.zero_grad()
loss.backward()
optim_reg.step()
reg_losses += loss.detach()
policy.load_state_dict(policy_unprojected.state_dict())
if not policy.contextual_std:
# set policy with projection value.
# In non-contextual cases we have only one cov, so the projection is the same.
policy.set_std(p_target[1][0])
steps = self.regression_iters * (math.ceil(obs.shape[0] / n_minibatches))
return {"regression_loss": (reg_losses / steps).detach()}

97
projections_orig/frob_projection_layer.py
View File

@ -1,97 +0,0 @@
# Copyright (c) 2021 Robert Bosch GmbH
# Author: Fabian Otto
#
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU Affero General Public License as published
# by the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU Affero General Public License for more details.
#
# You should have received a copy of the GNU Affero General Public License
# along with this program. If not, see <https://www.gnu.org/licenses/>.
import torch as ch
from typing import Tuple
from trust_region_projections.models.policy.abstract_gaussian_policy import AbstractGaussianPolicy
from trust_region_projections.projections.base_projection_layer import BaseProjectionLayer, mean_projection
from trust_region_projections.utils.projection_utils import gaussian_frobenius
class FrobeniusProjectionLayer(BaseProjectionLayer):
def _trust_region_projection(self, policy: AbstractGaussianPolicy, p: Tuple[ch.Tensor, ch.Tensor],
q: Tuple[ch.Tensor, ch.Tensor], eps: ch.Tensor, eps_cov: ch.Tensor, **kwargs):
"""
Runs Frobenius projection layer and constructs cholesky of covariance
Args:
policy: policy instance
p: current distribution
q: old distribution
eps: (modified) kl bound/ kl bound for mean part
eps_cov: (modified) kl bound for cov part
beta: (modified) entropy bound
**kwargs:
Returns: mean, cov cholesky
"""
mean, chol = p
old_mean, old_chol = q
batch_shape = mean.shape[:-1]
####################################################################################################################
# precompute mean and cov part of frob projection, which are used for the projection.
mean_part, cov_part, cov, cov_old = gaussian_frobenius(policy, p, q, self.scale_prec, True)
################################################################################################################
# mean projection maha/euclidean
proj_mean = mean_projection(mean, old_mean, mean_part, eps)
################################################################################################################
# cov projection frobenius
cov_mask = cov_part > eps_cov
if cov_mask.any():
# alpha = ch.where(fro_norm_sq > eps_cov, ch.sqrt(fro_norm_sq / eps_cov) - 1., ch.tensor(1.))
eta = ch.ones(batch_shape, dtype=chol.dtype, device=chol.device)
eta[cov_mask] = ch.sqrt(cov_part[cov_mask] / eps_cov) - 1.
eta = ch.max(-eta, eta)
new_cov = (cov + ch.einsum('i,ijk->ijk', eta, cov_old)) / (1. + eta + 1e-16)[..., None, None]
proj_chol = ch.where(cov_mask[..., None, None], ch.cholesky(new_cov), chol)
else:
proj_chol = chol
return proj_mean, proj_chol
def trust_region_value(self, policy, p, q):
"""
Computes the Frobenius metric between two Gaussian distributions p and q.
Args:
policy: policy instance
p: current distribution
q: old distribution
Returns:
mean and covariance part of Frobenius metric
"""
return gaussian_frobenius(policy, p, q, self.scale_prec)
def get_trust_region_loss(self, policy: AbstractGaussianPolicy, p: Tuple[ch.Tensor, ch.Tensor],
proj_p: Tuple[ch.Tensor, ch.Tensor]):
mean_diff, _ = self.trust_region_value(policy, p, proj_p)
if policy.contextual_std:
# Compute MSE here, because we found the Frobenius norm tends to generate values that explode for the cov
cov_diff = (p[1] - proj_p[1]).pow(2).sum([-1, -2])
delta_loss = (mean_diff + cov_diff).mean()
else:
delta_loss = mean_diff.mean()
return delta_loss * self.trust_region_coeff

101
projections_orig/kl_projection_layer.py
View File

@ -1,101 +0,0 @@
import cpp_projection
import numpy as np
import torch as ch
from typing import Any, Tuple
from trust_region_projections.models.policy.abstract_gaussian_policy import AbstractGaussianPolicy
from trust_region_projections.projections.base_projection_layer import BaseProjectionLayer, mean_projection
from trust_region_projections.utils.projection_utils import gaussian_kl
from trust_region_projections.utils.torch_utils import get_numpy
class KLProjectionLayer(BaseProjectionLayer):
def _trust_region_projection(self, policy: AbstractGaussianPolicy, p: Tuple[ch.Tensor, ch.Tensor],
q: Tuple[ch.Tensor, ch.Tensor], eps: ch.Tensor, eps_cov: ch.Tensor, **kwargs):
"""
Runs KL projection layer and constructs cholesky of covariance
Args:
policy: policy instance
p: current distribution
q: old distribution
eps: (modified) kl bound/ kl bound for mean part
eps_cov: (modified) kl bound for cov part
**kwargs:
Returns:
projected mean, projected cov cholesky
"""
mean, std = p
old_mean, old_std = q
if not policy.contextual_std:
# only project first one to reduce number of numerical optimizations
std = std[:1]
old_std = old_std[:1]
################################################################################################################
# project mean with closed form
mean_part, _ = gaussian_kl(policy, p, q)
proj_mean = mean_projection(mean, old_mean, mean_part, eps)
cov = policy.covariance(std)
old_cov = policy.covariance(old_std)
if policy.is_diag:
proj_cov = KLProjectionGradFunctionDiagCovOnly.apply(cov.diagonal(dim1=-2, dim2=-1),
old_cov.diagonal(dim1=-2, dim2=-1),
eps_cov)
proj_std = proj_cov.sqrt().diag_embed()
else:
raise NotImplementedError("The KL projection currently does not support full covariance matrices.")
if not policy.contextual_std:
# scale first std back to batchsize
proj_std = proj_std.expand(mean.shape[0], -1, -1)
return proj_mean, proj_std
class KLProjectionGradFunctionDiagCovOnly(ch.autograd.Function):
projection_op = None
@staticmethod
def get_projection_op(batch_shape, dim, max_eval=100):
if not KLProjectionGradFunctionDiagCovOnly.projection_op:
KLProjectionGradFunctionDiagCovOnly.projection_op = \
cpp_projection.BatchedDiagCovOnlyProjection(batch_shape, dim, max_eval=max_eval)
return KLProjectionGradFunctionDiagCovOnly.projection_op
@staticmethod
def forward(ctx: Any, *args: Any, **kwargs: Any) -> Any:
std, old_std, eps_cov = args
batch_shape = std.shape[0]
dim = std.shape[-1]
cov_np = get_numpy(std)
old_std = get_numpy(old_std)
eps = get_numpy(eps_cov) * np.ones(batch_shape)
# p_op = cpp_projection.BatchedDiagCovOnlyProjection(batch_shape, dim)
# ctx.proj = projection_op
p_op = KLProjectionGradFunctionDiagCovOnly.get_projection_op(batch_shape, dim)
ctx.proj = p_op
proj_std = p_op.forward(eps, old_std, cov_np)
return std.new(proj_std)
@staticmethod
def backward(ctx: Any, *grad_outputs: Any) -> Any:
projection_op = ctx.proj
d_std, = grad_outputs
d_std_np = get_numpy(d_std)
d_std_np = np.atleast_2d(d_std_np)
df_stds = projection_op.backward(d_std_np)
df_stds = np.atleast_2d(df_stds)
return d_std.new(df_stds), None, None

233
projections_orig/papi_projection.py
View File

@ -1,233 +0,0 @@
# Copyright (c) 2021 Robert Bosch GmbH
# Author: Fabian Otto
#
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU Affero General Public License as published
# by the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU Affero General Public License for more details.
#
# You should have received a copy of the GNU Affero General Public License
# along with this program. If not, see <https://www.gnu.org/licenses/>.
import logging
import copy
import numpy as np
import torch as ch
from typing import Tuple, Union
from trust_region_projections.utils.projection_utils import gaussian_kl
from trust_region_projections.models.policy.abstract_gaussian_policy import AbstractGaussianPolicy
from trust_region_projections.projections.base_projection_layer import BaseProjectionLayer
from trust_region_projections.utils.torch_utils import torch_batched_trace
logger = logging.getLogger("papi_projection")
class PAPIProjection(BaseProjectionLayer):
def __init__(self,
proj_type: str = "",
mean_bound: float = 0.015,
cov_bound: float = 0.0,
entropy_eq: bool = False,
entropy_first: bool = True,
cpu: bool = True,
dtype: ch.dtype = ch.float32,
**kwargs
):
"""
PAPI projection, which can be used after each training epoch to satisfy the trust regions.
Args:
proj_type: Which type of projection to use. None specifies no projection and uses the TRPO objective.
mean_bound: projection bound for the step size,
PAPI only has a joint KL constraint, mean and cov bound are summed for this bound.
cov_bound: projection bound for the step size,
PAPI only has a joint KL constraint, mean and cov bound are summed for this bound.
entropy_eq: Use entropy equality constraints.
entropy_first: Project entropy before trust region.
cpu: Compute on CPU only.
dtype: Data type to use, either of float32 or float64. The later might be necessary for higher
dimensions in order to learn the full covariance.
"""
assert entropy_first
super().__init__(proj_type, mean_bound, cov_bound, 0.0, False, None, None, None, 0.0, 0.0, entropy_eq,
entropy_first, cpu, dtype)
self.last_policies = []
def __call__(self, policy, p, q, step=0, *args, **kwargs):
if kwargs.get("obs"):
self._papi_steps(policy, q, **kwargs)
else:
return p
def _trust_region_projection(self, policy: AbstractGaussianPolicy, p: Tuple[ch.Tensor, ch.Tensor],
q: Tuple[ch.Tensor, ch.Tensor], eps: Union[ch.Tensor, float],
eps_cov: Union[ch.Tensor, float], **kwargs):
"""
runs papi projection layer and constructs sqrt of covariance
Args:
policy: policy instance
p: current distribution
q: old distribution
eps: (modified) kl bound/ kl bound for mean part
eps_cov: (modified) kl bound for cov part
**kwargs:
Returns:
mean, cov sqrt
"""
mean, chol = p
old_mean, old_chol = q
intermed_mean = kwargs.get('intermed_mean')
dtype = mean.dtype
device = mean.device
dim = mean.shape[-1]
################################################################################################################
# Precompute basic matrices
# Joint bound
eps += eps_cov
I = ch.eye(dim, dtype=dtype, device=device)
old_precision = ch.cholesky_solve(I, old_chol)[0]
logdet_old = policy.log_determinant(old_chol)
cov = policy.covariance(chol)
################################################################################################################
# compute expected KL
maha_part, cov_part = gaussian_kl(policy, p, q)
maha_part = maha_part.mean()
cov_part = cov_part.mean()
if intermed_mean is not None:
maha_intermediate = 0.5 * policy.maha(intermed_mean, old_mean, old_chol).mean()
mm = ch.min(maha_part, maha_intermediate)
################################################################################################################
# matrix rotation/rescaling projection
if maha_part + cov_part > eps + 1e-6:
old_cov = policy.covariance(old_chol)
maha_delta = eps if intermed_mean is None else (eps - mm)
eta_rot = maha_delta / ch.max(maha_part + cov_part, ch.tensor(1e-16, dtype=dtype, device=device))
new_cov = (1 - eta_rot) * old_cov + eta_rot * cov
proj_chol = ch.cholesky(new_cov)
# recompute covariance part of KL for new chol
trace_term = 0.5 * (torch_batched_trace(old_precision @ new_cov) - dim).mean() # rotation difference
entropy_diff = 0.5 * (logdet_old - policy.log_determinant(proj_chol)).mean()
cov_part = trace_term + entropy_diff
else:
proj_chol = chol
################################################################################################################
# mean interpolation projection
if maha_part + cov_part > eps + 1e-6:
if intermed_mean is not None:
a = 0.5 * policy.maha(mean, intermed_mean, old_chol).mean()
b = 0.5 * ((mean - intermed_mean) @ old_precision @ (intermed_mean - old_mean).T).mean()
c = maha_intermediate - ch.max(eps - cov_part, ch.tensor(0., dtype=dtype, device=device))
eta_mean = (-b + ch.sqrt(ch.max(b * b - a * c, ch.tensor(1e-16, dtype=dtype, device=device)))) / \
ch.max(a, ch.tensor(1e-16, dtype=dtype, device=device))
else:
eta_mean = ch.sqrt(
ch.max(eps - cov_part, ch.tensor(1e-16, dtype=dtype, device=device)) /
ch.max(maha_part, ch.tensor(1e-16, dtype=dtype, device=device)))
else:
eta_mean = ch.tensor(1., dtype=dtype, device=device)
return eta_mean, proj_chol
def _papi_steps(self, policy: AbstractGaussianPolicy, q: Tuple[ch.Tensor, ch.Tensor], obs: ch.Tensor, lr_schedule,
lr_schedule_vf=None):
"""
Take PAPI steps after PPO finished its steps. Policy parameters are updated in-place.
Args:
policy: policy instance
q: old distribution
obs: collected observations from trajectories
lr_schedule: lr schedule for policy
lr_schedule_vf: lr schedule for vf
Returns:
"""
assert not policy.contextual_std
# save latest policy in history
self.last_policies.append(copy.deepcopy(policy))
################################################################################################################
# policy backtracking: out of last n policies and current one find one that satisfies the kl constraint
intermed_policy = None
n_backtracks = 0
for i, pi in enumerate(reversed(self.last_policies)):
p_prime = pi(obs)
mean_part, cov_part = pi.kl_divergence(p_prime, q)
if (mean_part + cov_part).mean() <= self.mean_bound + self.cov_bound:
intermed_policy = pi
n_backtracks = i
break
################################################################################################################
# LR update
# reduce learning rate when appropriate policy not within the last 4 epochs
if n_backtracks >= 4 or intermed_policy is None:
# Linear learning rate annealing
lr_schedule.step()
if lr_schedule_vf:
lr_schedule_vf.step()
if intermed_policy is None:
# pop last policy and make it current one, as the updated one was poor
# do not keep last policy in history, otherwise we could stack the same policy multiple times.
if len(self.last_policies) >= 1:
policy.load_state_dict(self.last_policies.pop().state_dict())
logger.warning(f"No suitable policy found in backtracking of {len(self.last_policies)} policies.")
return
################################################################################################################
# PAPI iterations
# We assume only non contextual covariances here, therefore we only need to project for one
q = (q[0], q[1][:1]) # (means, covs[:1])
# This is A from Alg. 2 [Akrour et al., 2019]
intermed_weight = intermed_policy.get_last_layer().detach().clone()
# This is A @ phi(s)
intermed_mean = p_prime[0].detach().clone()
entropy = policy.entropy(q)
entropy_bound = obs.new_tensor([-np.inf]) if entropy / self.initial_entropy > 0.5 \
else entropy - (self.mean_bound + self.cov_bound)
for _ in range(20):
eta, proj_chol = self._projection(intermed_policy, (p_prime[0], p_prime[1][:1]), q,
self.mean_bound, self.cov_bound, entropy_bound,
intermed_mean=intermed_mean)
intermed_policy.papi_weight_update(eta, intermed_weight)
intermed_policy.set_std(proj_chol[0])
p_prime = intermed_policy(obs)
policy.load_state_dict(intermed_policy.state_dict())

54
projections_orig/projection_factory.py
View File

@ -1,54 +0,0 @@
# Copyright (c) 2021 Robert Bosch GmbH
# Author: Fabian Otto
#
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU Affero General Public License as published
# by the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU Affero General Public License for more details.
#
# You should have received a copy of the GNU Affero General Public License
# along with this program. If not, see <https://www.gnu.org/licenses/>.
from trust_region_projections.projections.base_projection_layer import BaseProjectionLayer
from trust_region_projections.projections.frob_projection_layer import FrobeniusProjectionLayer
from trust_region_projections.projections.kl_projection_layer import KLProjectionLayer
from trust_region_projections.projections.papi_projection import PAPIProjection
from trust_region_projections.projections.w2_projection_layer import WassersteinProjectionLayer
def get_projection_layer(proj_type: str = "", **kwargs) -> BaseProjectionLayer:
"""
Factory to generate the projection layers for all projections.
Args:
proj_type: One of None/' ', 'ppo', 'papi', 'w2', 'w2_non_com', 'frob', 'kl', or 'entropy'
**kwargs: arguments for projection layer
Returns:
"""
if not proj_type or proj_type.isspace() or proj_type.lower() in ["ppo", "sac", "td3", "mpo", "entropy"]:
return BaseProjectionLayer(proj_type, **kwargs)
elif proj_type.lower() == "w2":
return WassersteinProjectionLayer(proj_type, **kwargs)
elif proj_type.lower() == "frob":
return FrobeniusProjectionLayer(proj_type, **kwargs)
elif proj_type.lower() == "kl":
return KLProjectionLayer(proj_type, **kwargs)
elif proj_type.lower() == "papi":
# papi has a different approach compared to our projections.
# It has to be applied after the training with PPO.
return PAPIProjection(proj_type, **kwargs)
else:
raise ValueError(
f"Invalid projection type {proj_type}."
f" Choose one of None/' ', 'ppo', 'papi', 'w2', 'w2_non_com', 'frob', 'kl', or 'entropy'.")

84
projections_orig/w2_projection_layer.py
View File

@ -1,84 +0,0 @@
# Copyright (c) 2021 Robert Bosch GmbH
# Author: Fabian Otto
#
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU Affero General Public License as published
# by the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU Affero General Public License for more details.
#
# You should have received a copy of the GNU Affero General Public License
# along with this program. If not, see <https://www.gnu.org/licenses/>.
import torch as ch
from typing import Tuple
from trust_region_projections.models.policy.abstract_gaussian_policy import AbstractGaussianPolicy
from trust_region_projections.projections.base_projection_layer import BaseProjectionLayer, mean_projection
from trust_region_projections.utils.projection_utils import gaussian_wasserstein_commutative
class WassersteinProjectionLayer(BaseProjectionLayer):
def _trust_region_projection(self, policy: AbstractGaussianPolicy, p: Tuple[ch.Tensor, ch.Tensor],
q: Tuple[ch.Tensor, ch.Tensor], eps: ch.Tensor, eps_cov: ch.Tensor, **kwargs):
"""
Runs commutative Wasserstein projection layer and constructs sqrt of covariance
Args:
policy: policy instance
p: current distribution
q: old distribution
eps: (modified) kl bound/ kl bound for mean part
eps_cov: (modified) kl bound for cov part
**kwargs:
Returns:
mean, cov sqrt
"""
mean, sqrt = p
old_mean, old_sqrt = q
batch_shape = mean.shape[:-1]
####################################################################################################################
# precompute mean and cov part of W2, which are used for the projection.
# Both parts differ based on precision scaling.
# If activated, the mean part is the maha distance and the cov has a more complex term in the inner parenthesis.
mean_part, cov_part = gaussian_wasserstein_commutative(policy, p, q, self.scale_prec)
####################################################################################################################
# project mean (w/ or w/o precision scaling)
proj_mean = mean_projection(mean, old_mean, mean_part, eps)
####################################################################################################################
# project covariance (w/ or w/o precision scaling)
cov_mask = cov_part > eps_cov
if cov_mask.any():
# gradient issue with ch.where, it executes both paths and gives NaN gradient.
eta = ch.ones(batch_shape, dtype=sqrt.dtype, device=sqrt.device)
eta[cov_mask] = ch.sqrt(cov_part[cov_mask] / eps_cov) - 1.
eta = ch.max(-eta, eta)
new_sqrt = (sqrt + ch.einsum('i,ijk->ijk', eta, old_sqrt)) / (1. + eta + 1e-16)[..., None, None]
proj_sqrt = ch.where(cov_mask[..., None, None], new_sqrt, sqrt)
else:
proj_sqrt = sqrt
return proj_mean, proj_sqrt
def trust_region_value(self, policy, p, q):
"""
Computes the Wasserstein distance between two Gaussian distributions p and q.
Args:
policy: policy instance
p: current distribution
q: old distribution
Returns:
mean and covariance part of Wasserstein distance
"""
return gaussian_wasserstein_commutative(policy, p, q, scale_prec=self.scale_prec)

2
trl_pg/__init__.py
View File

@ -1,2 +0,0 @@
from sb3.trl_pg.policies import CnnPolicy, MlpPolicy, MultiInputPolicy
from sb3.trl_pg.trl_pg import TRL_PG

7
trl_pg/policies.py
View File

@ -1,7 +0,0 @@
# This file is here just to define MlpPolicy/CnnPolicy
# that work for TRL_PG
from stable_baselines3.common.policies import ActorCriticCnnPolicy, ActorCriticPolicy, MultiInputActorCriticPolicy
MlpPolicy = ActorCriticPolicy
CnnPolicy = ActorCriticCnnPolicy
MultiInputPolicy = MultiInputActorCriticPolicy

338
trl_pg/trl_pg.py
View File

@ -1,338 +0,0 @@
import warnings
from typing import Any, Dict, Optional, Type, Union
import numpy as np
import torch as th
from gym import spaces
from torch.nn import functional as F
from stable_baselines3.common.on_policy_algorithm import OnPolicyAlgorithm
from stable_baselines3.common.policies import ActorCriticCnnPolicy, ActorCriticPolicy, BasePolicy, MultiInputActorCriticPolicy
from stable_baselines3.common.type_aliases import GymEnv, MaybeCallback, Schedule
from stable_baselines3.common.utils import explained_variance, get_schedule_fn
class TRL_PG(OnPolicyAlgorithm):
"""
Differential Trust Region Layer (TRL) for Policy Gradient (PG)
Paper: https://arxiv.org/abs/2101.09207
Code: This implementation borrows (/steals most) code from SB3's PPO implementation https://github.com/DLR-RM/stable-baselines3/blob/master/stable_baselines3/ppo/ppo.py
The implementation of the TRL-specific parts borrows from https://github.com/boschresearch/trust-region-layers/blob/main/trust_region_projections/algorithms/pg/pg.py
:param policy: The policy model to use (MlpPolicy, CnnPolicy, ...)
:param env: The environment to learn from (if registered in Gym, can be str)
:param learning_rate: The learning rate, it can be a function
of the current progress remaining (from 1 to 0)
:param n_steps: The number of steps to run for each environment per update
(i.e. rollout buffer size is n_steps * n_envs where n_envs is number of environment copies running in parallel)
NOTE: n_steps * n_envs must be greater than 1 (because of the advantage normalization)
See https://github.com/pytorch/pytorch/issues/29372
:param batch_size: Minibatch size
:param n_epochs: Number of epoch when optimizing the surrogate loss
:param gamma: Discount factor
:param gae_lambda: Factor for trade-off of bias vs variance for Generalized Advantage Estimator
:param clip_range: Clipping parameter, it can be a function of the current progress
remaining (from 1 to 0).
:param clip_range_vf: Clipping parameter for the value function,
it can be a function of the current progress remaining (from 1 to 0).
This is a parameter specific to the OpenAI implementation. If None is passed (default),
no clipping will be done on the value function.
IMPORTANT: this clipping depends on the reward scaling.
:param normalize_advantage: Whether to normalize or not the advantage
:param ent_coef: Entropy coefficient for the loss calculation
:param vf_coef: Value function coefficient for the loss calculation
:param max_grad_norm: The maximum value for the gradient clipping
:param use_sde: Whether to use generalized State Dependent Exploration (gSDE)
instead of action noise exploration (default: False)
:param sde_sample_freq: Sample a new noise matrix every n steps when using gSDE
Default: -1 (only sample at the beginning of the rollout)
:param target_kl: Limit the KL divergence between updates,
because the clipping is not enough to prevent large update
see issue #213 (cf https://github.com/hill-a/stable-baselines/issues/213)
By default, there is no limit on the kl div.
:param tensorboard_log: the log location for tensorboard (if None, no logging)
:param create_eval_env: Whether to create a second environment that will be
used for evaluating the agent periodically. (Only available when passing string for the environment)
:param policy_kwargs: additional arguments to be passed to the policy on creation
:param verbose: the verbosity level: 0 no output, 1 info, 2 debug
:param seed: Seed for the pseudo random generators
:param device: Device (cpu, cuda, ...) on which the code should be run.
Setting it to auto, the code will be run on the GPU if possible.
:param _init_setup_model: Whether or not to build the network at the creation of the instance
"""
policy_aliases: Dict[str, Type[BasePolicy]] = {
"MlpPolicy": ActorCriticPolicy,
"CnnPolicy": ActorCriticCnnPolicy,
"MultiInputPolicy": MultiInputActorCriticPolicy,
}
def __init__(
self,
policy: Union[str, Type[ActorCriticPolicy]],
env: Union[GymEnv, str],
learning_rate: Union[float, Schedule] = 3e-4,
n_steps: int = 2048,
batch_size: int = 64,
n_epochs: int = 10,
gamma: float = 0.99,
gae_lambda: float = 0.95,
clip_range: Union[float, Schedule] = 0.2,
clip_range_vf: Union[None, float, Schedule] = None,
normalize_advantage: bool = True,
ent_coef: float = 0.0,
vf_coef: float = 0.5,
max_grad_norm: float = 0.5,
use_sde: bool = False,
sde_sample_freq: int = -1,
target_kl: Optional[float] = None,
tensorboard_log: Optional[str] = None,
create_eval_env: bool = False,
policy_kwargs: Optional[Dict[str, Any]] = None,
verbose: int = 0,
seed: Optional[int] = None,
device: Union[th.device, str] = "auto",
# Different from PPO:
projection: BaseProjectionLayer = None,
_init_setup_model: bool = True,
):
super().__init__(
policy,
env,
learning_rate=learning_rate,
n_steps=n_steps,
gamma=gamma,
gae_lambda=gae_lambda,
ent_coef=ent_coef,
vf_coef=vf_coef,
max_grad_norm=max_grad_norm,
use_sde=use_sde,
sde_sample_freq=sde_sample_freq,
tensorboard_log=tensorboard_log,
policy_kwargs=policy_kwargs,
verbose=verbose,
device=device,
create_eval_env=create_eval_env,
seed=seed,
_init_setup_model=False,
supported_action_spaces=(
spaces.Box,
spaces.Discrete,
spaces.MultiDiscrete,
spaces.MultiBinary,
),
)
# Sanity check, otherwise it will lead to noisy gradient and NaN
# because of the advantage normalization
if normalize_advantage:
assert (
batch_size > 1
), "`batch_size` must be greater than 1. See https://github.com/DLR-RM/stable-baselines3/issues/440"
if self.env is not None:
# Check that `n_steps * n_envs > 1` to avoid NaN
# when doing advantage normalization
buffer_size = self.env.num_envs * self.n_steps
assert (
buffer_size > 1
), f"`n_steps * n_envs` must be greater than 1. Currently n_steps={self.n_steps} and n_envs={self.env.num_envs}"
# Check that the rollout buffer size is a multiple of the mini-batch size
untruncated_batches = buffer_size // batch_size
if buffer_size % batch_size > 0:
warnings.warn(
f"You have specified a mini-batch size of {batch_size},"
f" but because the `RolloutBuffer` is of size `n_steps * n_envs = {buffer_size}`,"
f" after every {untruncated_batches} untruncated mini-batches,"
f" there will be a truncated mini-batch of size {buffer_size % batch_size}\n"
f"We recommend using a `batch_size` that is a factor of `n_steps * n_envs`.\n"
f"Info: (n_steps={self.n_steps} and n_envs={self.env.num_envs})"
)
self.batch_size = batch_size
self.n_epochs = n_epochs
self.clip_range = clip_range
self.clip_range_vf = clip_range_vf
self.normalize_advantage = normalize_advantage
self.target_kl = target_kl
# Different from PPO:
self.projection = projection
if _init_setup_model:
self._setup_model()
def _setup_model(self) -> None:
super()._setup_model()
# Initialize schedules for policy/value clipping
self.clip_range = get_schedule_fn(self.clip_range)
if self.clip_range_vf is not None:
if isinstance(self.clip_range_vf, (float, int)):
assert self.clip_range_vf > 0, "`clip_range_vf` must be positive, " "pass `None` to deactivate vf clipping"
self.clip_range_vf = get_schedule_fn(self.clip_range_vf)
def train(self) -> None:
"""
Update policy using the currently gathered rollout buffer.
"""
# Switch to train mode (this affects batch norm / dropout)
self.policy.set_training_mode(True)
# Update optimizer learning rate
self._update_learning_rate(self.policy.optimizer)
# Compute current clip range
clip_range = self.clip_range(self._current_progress_remaining)
# Optional: clip range for the value function
if self.clip_range_vf is not None:
clip_range_vf = self.clip_range_vf(self._current_progress_remaining)
surrogate_losses = []
entropy_losses = []
trust_region_losses = []
pg_losses, value_losses = [], []
clip_fractions = []
continue_training = True
# train for n_epochs epochs
for epoch in range(self.n_epochs):
approx_kl_divs = []
# Do a complete pass on the rollout buffer
for rollout_data in self.rollout_buffer.get(self.batch_size):
actions = rollout_data.actions
if isinstance(self.action_space, spaces.Discrete):
# Convert discrete action from float to long
actions = rollout_data.actions.long().flatten()
# Re-sample the noise matrix because the log_std has changed
if self.use_sde:
self.policy.reset_noise(self.batch_size)
values, log_prob, entropy = self.policy.evaluate_actions(rollout_data.observations, actions)
values = values.flatten()
# Normalize advantage
advantages = rollout_data.advantages
if self.normalize_advantage:
advantages = (advantages - advantages.mean()) / (advantages.std() + 1e-8)
# ratio between old and new policy, should be one at the first iteration
ratio = th.exp(log_prob - rollout_data.old_log_prob)
# Difference from PPO: We renamed 'policy_loss' to 'surrogate_loss'
# clipped surrogate loss
surrogate_loss_1 = advantages * ratio
surrogate_loss_2 = advantages * th.clamp(ratio, 1 - clip_range, 1 + clip_range)
surrogate_loss = -th.min(policy_loss_1, policy_loss_2).mean()
surrogate_losses.append(surrogate_loss.item())
# Logging
pg_losses.append(policy_loss.item())
clip_fraction = th.mean((th.abs(ratio - 1) > clip_range).float()).item()
clip_fractions.append(clip_fraction)
if self.clip_range_vf is None:
# No clipping
values_pred = values
else:
# Clip the different between old and new value
# NOTE: this depends on the reward scaling
values_pred = rollout_data.old_values + th.clamp(
values - rollout_data.old_values, -clip_range_vf, clip_range_vf
)
# Value loss using the TD(gae_lambda) target
value_loss = F.mse_loss(rollout_data.returns, values_pred)
value_losses.append(value_loss.item())
# Entropy loss favor exploration
if entropy is None:
# Approximate entropy when no analytical form
entropy_loss = -th.mean(-log_prob)
else:
entropy_loss = -th.mean(entropy)
entropy_losses.append(entropy_loss.item())
# Difference to PPO: Added trust_region_loss; policy_loss includes entropy_loss + trust_region_loss
trust_region_loss = self.projection.get_trust_region_loss()#TODO: params
trust_region_losses.append(trust_region_loss.item())
policy_loss = surrogate_loss + self.ent_coef * entropy_loss + trust_region_loss
loss = policy_loss + self.vf_coef * value_loss
# Calculate approximate form of reverse KL Divergence for early stopping
# see issue #417: https://github.com/DLR-RM/stable-baselines3/issues/417
# and discussion in PR #419: https://github.com/DLR-RM/stable-baselines3/pull/419
# and Schulman blog: http://joschu.net/blog/kl-approx.html
with th.no_grad():
log_ratio = log_prob - rollout_data.old_log_prob
approx_kl_div = th.mean((th.exp(log_ratio) - 1) - log_ratio).cpu().numpy()
approx_kl_divs.append(approx_kl_div)
if self.target_kl is not None and approx_kl_div > 1.5 * self.target_kl:
continue_training = False
if self.verbose >= 1:
print(f"Early stopping at step {epoch} due to reaching max kl: {approx_kl_div:.2f}")
break
# Optimization step
self.policy.optimizer.zero_grad()
loss.backward()
# Clip grad norm
th.nn.utils.clip_grad_norm_(self.policy.parameters(), self.max_grad_norm)
self.policy.optimizer.step()
if not continue_training:
break
self._n_updates += self.n_epochs
explained_var = explained_variance(self.rollout_buffer.values.flatten(), self.rollout_buffer.returns.flatten())
# Logs
self.logger.record("train/surrogate_loss", np.mean(surrogate_losses))
self.logger.record("train/entropy_loss", np.mean(entropy_losses))
self.logger.record("train/trust_region_loss", np.mean(trust_region_losses))
self.logger.record("train/policy_gradient_loss", np.mean(pg_losses))
self.logger.record("train/value_loss", np.mean(value_losses))
self.logger.record("train/approx_kl", np.mean(approx_kl_divs))
self.logger.record("train/clip_fraction", np.mean(clip_fractions))
self.logger.record("train/loss", loss.item())
self.logger.record("train/explained_variance", explained_var)
if hasattr(self.policy, "log_std"):
self.logger.record("train/std", th.exp(self.policy.log_std).mean().item())
self.logger.record("train/n_updates", self._n_updates, exclude="tensorboard")
self.logger.record("train/clip_range", clip_range)
if self.clip_range_vf is not None:
self.logger.record("train/clip_range_vf", clip_range_vf)
def learn(
self,
total_timesteps: int,
callback: MaybeCallback = None,
log_interval: int = 1,
eval_env: Optional[GymEnv] = None,
eval_freq: int = -1,
n_eval_episodes: int = 5,
tb_log_name: str = "PPO",
eval_log_path: Optional[str] = None,
reset_num_timesteps: bool = True,
) -> "PPO":
return super().learn(
total_timesteps=total_timesteps,
callback=callback,
log_interval=log_interval,
eval_env=eval_env,
eval_freq=eval_freq,
n_eval_episodes=n_eval_episodes,
tb_log_name=tb_log_name,
eval_log_path=eval_log_path,
reset_num_timesteps=reset_num_timesteps,
)

2
trl_sac/__init__.py
View File

@ -1,2 +0,0 @@
from sb3_trl.trl_sac.policies import CnnPolicy, MlpPolicy, MultiInputPolicy
from sb3_trl.trl_sac.trl_sac import TRL_SAC

516
trl_sac/policies.py
View File

@ -1,516 +0,0 @@
import warnings
from typing import Any, Dict, List, Optional, Tuple, Type, Union
import gym
import torch as th
from torch import nn
from stable_baselines3.common.distributions import SquashedDiagGaussianDistribution, StateDependentNoiseDistribution
from stable_baselines3.common.policies import BasePolicy, ContinuousCritic
from stable_baselines3.common.preprocessing import get_action_dim
from stable_baselines3.common.torch_layers import (
BaseFeaturesExtractor,
CombinedExtractor,
FlattenExtractor,
NatureCNN,
create_mlp,
get_actor_critic_arch,
)
from stable_baselines3.common.type_aliases import Schedule
# CAP the standard deviation of the actor
LOG_STD_MAX = 2
LOG_STD_MIN = -20
class Actor(BasePolicy):
"""
Actor network (policy) for SAC.
:param observation_space: Obervation space
:param action_space: Action space
:param net_arch: Network architecture
:param features_extractor: Network to extract features
(a CNN when using images, a nn.Flatten() layer otherwise)
:param features_dim: Number of features
:param activation_fn: Activation function
:param use_sde: Whether to use State Dependent Exploration or not
:param log_std_init: Initial value for the log standard deviation
:param full_std: Whether to use (n_features x n_actions) parameters
for the std instead of only (n_features,) when using gSDE.
:param sde_net_arch: Network architecture for extracting features
when using gSDE. If None, the latent features from the policy will be used.
Pass an empty list to use the states as features.
:param use_expln: Use ``expln()`` function instead of ``exp()`` when using gSDE to ensure
a positive standard deviation (cf paper). It allows to keep variance
above zero and prevent it from growing too fast. In practice, ``exp()`` is usually enough.
:param clip_mean: Clip the mean output when using gSDE to avoid numerical instability.
:param normalize_images: Whether to normalize images or not,
dividing by 255.0 (True by default)
"""
def __init__(
self,
observation_space: gym.spaces.Space,
action_space: gym.spaces.Space,
net_arch: List[int],
features_extractor: nn.Module,
features_dim: int,
activation_fn: Type[nn.Module] = nn.ReLU,
use_sde: bool = False,
log_std_init: float = -3,
full_std: bool = True,
sde_net_arch: Optional[List[int]] = None,
use_expln: bool = False,
clip_mean: float = 2.0,
normalize_images: bool = True,
):
super().__init__(
observation_space,
action_space,
features_extractor=features_extractor,
normalize_images=normalize_images,
squash_output=True,
)
# Save arguments to re-create object at loading
self.use_sde = use_sde
self.sde_features_extractor = None
self.net_arch = net_arch
self.features_dim = features_dim
self.activation_fn = activation_fn
self.log_std_init = log_std_init
self.sde_net_arch = sde_net_arch
self.use_expln = use_expln
self.full_std = full_std
self.clip_mean = clip_mean
if sde_net_arch is not None:
warnings.warn("sde_net_arch is deprecated and will be removed in SB3 v2.4.0.", DeprecationWarning)
action_dim = get_action_dim(self.action_space)
latent_pi_net = create_mlp(features_dim, -1, net_arch, activation_fn)
self.latent_pi = nn.Sequential(*latent_pi_net)
last_layer_dim = net_arch[-1] if len(net_arch) > 0 else features_dim
if self.use_sde:
self.action_dist = StateDependentNoiseDistribution(
action_dim, full_std=full_std, use_expln=use_expln, learn_features=True, squash_output=True
)
self.mu, self.log_std = self.action_dist.proba_distribution_net(
latent_dim=last_layer_dim, latent_sde_dim=last_layer_dim, log_std_init=log_std_init
)
# Avoid numerical issues by limiting the mean of the Gaussian
# to be in [-clip_mean, clip_mean]
if clip_mean > 0.0:
self.mu = nn.Sequential(self.mu, nn.Hardtanh(min_val=-clip_mean, max_val=clip_mean))
else:
self.action_dist = SquashedDiagGaussianDistribution(action_dim)
self.mu = nn.Linear(last_layer_dim, action_dim)
self.log_std = nn.Linear(last_layer_dim, action_dim)
def _get_constructor_parameters(self) -> Dict[str, Any]:
data = super()._get_constructor_parameters()
data.update(
dict(
net_arch=self.net_arch,
features_dim=self.features_dim,
activation_fn=self.activation_fn,
use_sde=self.use_sde,
log_std_init=self.log_std_init,
full_std=self.full_std,
use_expln=self.use_expln,
features_extractor=self.features_extractor,
clip_mean=self.clip_mean,
)
)
return data
def get_std(self) -> th.Tensor:
"""
Retrieve the standard deviation of the action distribution.
Only useful when using gSDE.
It corresponds to ``th.exp(log_std)`` in the normal case,
but is slightly different when using ``expln`` function
(cf StateDependentNoiseDistribution doc).
:return:
"""
msg = "get_std() is only available when using gSDE"
assert isinstance(self.action_dist, StateDependentNoiseDistribution), msg
return self.action_dist.get_std(self.log_std)
def reset_noise(self, batch_size: int = 1) -> None:
"""
Sample new weights for the exploration matrix, when using gSDE.
:param batch_size:
"""
msg = "reset_noise() is only available when using gSDE"
assert isinstance(self.action_dist, StateDependentNoiseDistribution), msg
self.action_dist.sample_weights(self.log_std, batch_size=batch_size)
def get_action_dist_params(self, obs: th.Tensor) -> Tuple[th.Tensor, th.Tensor, Dict[str, th.Tensor]]:
"""
Get the parameters for the action distribution.
:param obs:
:return:
Mean, standard deviation and optional keyword arguments.
"""
features = self.extract_features(obs)
latent_pi = self.latent_pi(features)
mean_actions = self.mu(latent_pi)
if self.use_sde:
return mean_actions, self.log_std, dict(latent_sde=latent_pi)
# Unstructured exploration (Original implementation)
log_std = self.log_std(latent_pi)
# Original Implementation to cap the standard deviation
log_std = th.clamp(log_std, LOG_STD_MIN, LOG_STD_MAX)
return mean_actions, log_std, {}
def forward(self, obs: th.Tensor, deterministic: bool = False) -> th.Tensor:
mean_actions, log_std, kwargs = self.get_action_dist_params(obs)
# Note: the action is squashed
return self.action_dist.actions_from_params(mean_actions, log_std, deterministic=deterministic, **kwargs)
def action_log_prob(self, obs: th.Tensor) -> Tuple[th.Tensor, th.Tensor]:
mean_actions, log_std, kwargs = self.get_action_dist_params(obs)
# return action and associated log prob
return self.action_dist.log_prob_from_params(mean_actions, log_std, **kwargs)
def _predict(self, observation: th.Tensor, deterministic: bool = False) -> th.Tensor:
return self(observation, deterministic)
class SACPolicy(BasePolicy):
"""
Policy class (with both actor and critic) for SAC.
:param observation_space: Observation space
:param action_space: Action space
:param lr_schedule: Learning rate schedule (could be constant)
:param net_arch: The specification of the policy and value networks.
:param activation_fn: Activation function
:param use_sde: Whether to use State Dependent Exploration or not
:param log_std_init: Initial value for the log standard deviation
:param sde_net_arch: Network architecture for extracting features
when using gSDE. If None, the latent features from the policy will be used.
Pass an empty list to use the states as features.
:param use_expln: Use ``expln()`` function instead of ``exp()`` when using gSDE to ensure
a positive standard deviation (cf paper). It allows to keep variance
above zero and prevent it from growing too fast. In practice, ``exp()`` is usually enough.
:param clip_mean: Clip the mean output when using gSDE to avoid numerical instability.
:param features_extractor_class: Features extractor to use.
:param features_extractor_kwargs: Keyword arguments
to pass to the features extractor.
:param normalize_images: Whether to normalize images or not,
dividing by 255.0 (True by default)
:param optimizer_class: The optimizer to use,
``th.optim.Adam`` by default
:param optimizer_kwargs: Additional keyword arguments,
excluding the learning rate, to pass to the optimizer
:param n_critics: Number of critic networks to create.
:param share_features_extractor: Whether to share or not the features extractor
between the actor and the critic (this saves computation time)
"""
def __init__(
self,
observation_space: gym.spaces.Space,
action_space: gym.spaces.Space,
lr_schedule: Schedule,
net_arch: Optional[Union[List[int], Dict[str, List[int]]]] = None,
activation_fn: Type[nn.Module] = nn.ReLU,
use_sde: bool = False,
log_std_init: float = -3,
sde_net_arch: Optional[List[int]] = None,
use_expln: bool = False,
clip_mean: float = 2.0,
features_extractor_class: Type[BaseFeaturesExtractor] = FlattenExtractor,
features_extractor_kwargs: Optional[Dict[str, Any]] = None,
normalize_images: bool = True,
optimizer_class: Type[th.optim.Optimizer] = th.optim.Adam,
optimizer_kwargs: Optional[Dict[str, Any]] = None,
n_critics: int = 2,
share_features_extractor: bool = True,
):
super().__init__(
observation_space,
action_space,
features_extractor_class,
features_extractor_kwargs,
optimizer_class=optimizer_class,
optimizer_kwargs=optimizer_kwargs,
squash_output=True,
)
if net_arch is None:
if features_extractor_class == NatureCNN:
net_arch = []
else:
net_arch = [256, 256]
actor_arch, critic_arch = get_actor_critic_arch(net_arch)
self.net_arch = net_arch
self.activation_fn = activation_fn
self.net_args = {
"observation_space": self.observation_space,
"action_space": self.action_space,
"net_arch": actor_arch,
"activation_fn": self.activation_fn,
"normalize_images": normalize_images,
}
self.actor_kwargs = self.net_args.copy()
if sde_net_arch is not None:
warnings.warn("sde_net_arch is deprecated and will be removed in SB3 v2.4.0.", DeprecationWarning)
sde_kwargs = {
"use_sde": use_sde,
"log_std_init": log_std_init,
"use_expln": use_expln,
"clip_mean": clip_mean,
}
self.actor_kwargs.update(sde_kwargs)
self.critic_kwargs = self.net_args.copy()
self.critic_kwargs.update(
{
"n_critics": n_critics,
"net_arch": critic_arch,
"share_features_extractor": share_features_extractor,
}
)
self.actor, self.actor_target = None, None
self.critic, self.critic_target = None, None
self.share_features_extractor = share_features_extractor
self._build(lr_schedule)
def _build(self, lr_schedule: Schedule) -> None:
self.actor = self.make_actor()
self.actor.optimizer = self.optimizer_class(self.actor.parameters(), lr=lr_schedule(1), **self.optimizer_kwargs)
if self.share_features_extractor:
self.critic = self.make_critic(features_extractor=self.actor.features_extractor)
# Do not optimize the shared features extractor with the critic loss
# otherwise, there are gradient computation issues
critic_parameters = [param for name, param in self.critic.named_parameters() if "features_extractor" not in name]
else:
# Create a separate features extractor for the critic
# this requires more memory and computation
self.critic = self.make_critic(features_extractor=None)
critic_parameters = self.critic.parameters()
# Critic target should not share the features extractor with critic
self.critic_target = self.make_critic(features_extractor=None)
self.critic_target.load_state_dict(self.critic.state_dict())
self.critic.optimizer = self.optimizer_class(critic_parameters, lr=lr_schedule(1), **self.optimizer_kwargs)
# Target networks should always be in eval mode
self.critic_target.set_training_mode(False)
def _get_constructor_parameters(self) -> Dict[str, Any]:
data = super()._get_constructor_parameters()
data.update(
dict(
net_arch=self.net_arch,
activation_fn=self.net_args["activation_fn"],
use_sde=self.actor_kwargs["use_sde"],
log_std_init=self.actor_kwargs["log_std_init"],
use_expln=self.actor_kwargs["use_expln"],
clip_mean=self.actor_kwargs["clip_mean"],
n_critics=self.critic_kwargs["n_critics"],
lr_schedule=self._dummy_schedule, # dummy lr schedule, not needed for loading policy alone
optimizer_class=self.optimizer_class,
optimizer_kwargs=self.optimizer_kwargs,
features_extractor_class=self.features_extractor_class,
features_extractor_kwargs=self.features_extractor_kwargs,
)
)
return data
def reset_noise(self, batch_size: int = 1) -> None:
"""
Sample new weights for the exploration matrix, when using gSDE.
:param batch_size:
"""
self.actor.reset_noise(batch_size=batch_size)
def make_actor(self, features_extractor: Optional[BaseFeaturesExtractor] = None) -> Actor:
actor_kwargs = self._update_features_extractor(self.actor_kwargs, features_extractor)
return Actor(**actor_kwargs).to(self.device)
def make_critic(self, features_extractor: Optional[BaseFeaturesExtractor] = None) -> ContinuousCritic:
critic_kwargs = self._update_features_extractor(self.critic_kwargs, features_extractor)
return ContinuousCritic(**critic_kwargs).to(self.device)
def forward(self, obs: th.Tensor, deterministic: bool = False) -> th.Tensor:
return self._predict(obs, deterministic=deterministic)
def _predict(self, observation: th.Tensor, deterministic: bool = False) -> th.Tensor:
return self.actor(observation, deterministic)
def set_training_mode(self, mode: bool) -> None:
"""
Put the policy in either training or evaluation mode.
This affects certain modules, such as batch normalisation and dropout.
:param mode: if true, set to training mode, else set to evaluation mode
"""
self.actor.set_training_mode(mode)
self.critic.set_training_mode(mode)
self.training = mode
MlpPolicy = SACPolicy
class CnnPolicy(SACPolicy):
"""
Policy class (with both actor and critic) for SAC.
:param observation_space: Observation space
:param action_space: Action space
:param lr_schedule: Learning rate schedule (could be constant)
:param net_arch: The specification of the policy and value networks.
:param activation_fn: Activation function
:param use_sde: Whether to use State Dependent Exploration or not
:param log_std_init: Initial value for the log standard deviation
:param sde_net_arch: Network architecture for extracting features
when using gSDE. If None, the latent features from the policy will be used.
Pass an empty list to use the states as features.
:param use_expln: Use ``expln()`` function instead of ``exp()`` when using gSDE to ensure
a positive standard deviation (cf paper). It allows to keep variance
above zero and prevent it from growing too fast. In practice, ``exp()`` is usually enough.
:param clip_mean: Clip the mean output when using gSDE to avoid numerical instability.
:param features_extractor_class: Features extractor to use.
:param normalize_images: Whether to normalize images or not,
dividing by 255.0 (True by default)
:param optimizer_class: The optimizer to use,
``th.optim.Adam`` by default
:param optimizer_kwargs: Additional keyword arguments,
excluding the learning rate, to pass to the optimizer
:param n_critics: Number of critic networks to create.
:param share_features_extractor: Whether to share or not the features extractor
between the actor and the critic (this saves computation time)
"""
def __init__(
self,
observation_space: gym.spaces.Space,
action_space: gym.spaces.Space,
lr_schedule: Schedule,
net_arch: Optional[Union[List[int], Dict[str, List[int]]]] = None,
activation_fn: Type[nn.Module] = nn.ReLU,
use_sde: bool = False,
log_std_init: float = -3,
sde_net_arch: Optional[List[int]] = None,
use_expln: bool = False,
clip_mean: float = 2.0,
features_extractor_class: Type[BaseFeaturesExtractor] = NatureCNN,
features_extractor_kwargs: Optional[Dict[str, Any]] = None,
normalize_images: bool = True,
optimizer_class: Type[th.optim.Optimizer] = th.optim.Adam,
optimizer_kwargs: Optional[Dict[str, Any]] = None,
n_critics: int = 2,
share_features_extractor: bool = True,
):
super().__init__(
observation_space,
action_space,
lr_schedule,
net_arch,
activation_fn,
use_sde,
log_std_init,
sde_net_arch,
use_expln,
clip_mean,
features_extractor_class,
features_extractor_kwargs,
normalize_images,
optimizer_class,
optimizer_kwargs,
n_critics,
share_features_extractor,
)
class MultiInputPolicy(SACPolicy):
"""
Policy class (with both actor and critic) for SAC.
:param observation_space: Observation space
:param action_space: Action space
:param lr_schedule: Learning rate schedule (could be constant)
:param net_arch: The specification of the policy and value networks.
:param activation_fn: Activation function
:param use_sde: Whether to use State Dependent Exploration or not
:param log_std_init: Initial value for the log standard deviation
:param sde_net_arch: Network architecture for extracting features
when using gSDE. If None, the latent features from the policy will be used.
Pass an empty list to use the states as features.
:param use_expln: Use ``expln()`` function instead of ``exp()`` when using gSDE to ensure
a positive standard deviation (cf paper). It allows to keep variance
above zero and prevent it from growing too fast. In practice, ``exp()`` is usually enough.
:param clip_mean: Clip the mean output when using gSDE to avoid numerical instability.
:param features_extractor_class: Features extractor to use.
:param normalize_images: Whether to normalize images or not,
dividing by 255.0 (True by default)
:param optimizer_class: The optimizer to use,
``th.optim.Adam`` by default
:param optimizer_kwargs: Additional keyword arguments,
excluding the learning rate, to pass to the optimizer
:param n_critics: Number of critic networks to create.
:param share_features_extractor: Whether to share or not the features extractor
between the actor and the critic (this saves computation time)
"""
def __init__(
self,
observation_space: gym.spaces.Space,
action_space: gym.spaces.Space,
lr_schedule: Schedule,
net_arch: Optional[Union[List[int], Dict[str, List[int]]]] = None,
activation_fn: Type[nn.Module] = nn.ReLU,
use_sde: bool = False,
log_std_init: float = -3,
sde_net_arch: Optional[List[int]] = None,
use_expln: bool = False,
clip_mean: float = 2.0,
features_extractor_class: Type[BaseFeaturesExtractor] = CombinedExtractor,
features_extractor_kwargs: Optional[Dict[str, Any]] = None,
normalize_images: bool = True,
optimizer_class: Type[th.optim.Optimizer] = th.optim.Adam,
optimizer_kwargs: Optional[Dict[str, Any]] = None,
n_critics: int = 2,
share_features_extractor: bool = True,
):
super().__init__(
observation_space,
action_space,
lr_schedule,
net_arch,
activation_fn,
use_sde,
log_std_init,
sde_net_arch,
use_expln,
clip_mean,
features_extractor_class,
features_extractor_kwargs,
normalize_images,
optimizer_class,
optimizer_kwargs,
n_critics,
share_features_extractor,
)

324
trl_sac/trl_sac.py
View File

@ -1,324 +0,0 @@
from typing import Any, Dict, List, Optional, Tuple, Type, Union
import gym
import numpy as np
import torch as th
from torch.nn import functional as F
from stable_baselines3.common.buffers import ReplayBuffer
from stable_baselines3.common.noise import ActionNoise
from stable_baselines3.common.off_policy_algorithm import OffPolicyAlgorithm
from stable_baselines3.common.policies import BasePolicy
from stable_baselines3.common.type_aliases import GymEnv, MaybeCallback, Schedule
from stable_baselines3.common.utils import polyak_update
from stable_baselines3.sac.policies import CnnPolicy, MlpPolicy, MultiInputPolicy, SACPolicy
class TRL_SAC(OffPolicyAlgorithm):
"""
Trust Region Layers (TRL) based on SAC (Soft Actor Critic)
This implementation is almost a 1:1-copy of the sb3-code for SAC.
Only minor changes have been made to implement Differential Trust Region Layers
Description from original SAC implementation:
Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor,
This implementation borrows code from original implementation (https://github.com/haarnoja/sac)
from OpenAI Spinning Up (https://github.com/openai/spinningup), from the softlearning repo
(https://github.com/rail-berkeley/softlearning/)
and from Stable Baselines (https://github.com/hill-a/stable-baselines)
Paper: https://arxiv.org/abs/1801.01290
Introduction to SAC: https://spinningup.openai.com/en/latest/algorithms/sac.html
Note: we use double q target and not value target as discussed
in https://github.com/hill-a/stable-baselines/issues/270
:param policy: The policy model to use (MlpPolicy, CnnPolicy, ...)
:param env: The environment to learn from (if registered in Gym, can be str)
:param learning_rate: learning rate for adam optimizer,
the same learning rate will be used for all networks (Q-Values, Actor and Value function)
it can be a function of the current progress remaining (from 1 to 0)
:param buffer_size: size of the replay buffer
:param learning_starts: how many steps of the model to collect transitions for before learning starts
:param batch_size: Minibatch size for each gradient update
:param tau: the soft update coefficient ("Polyak update", between 0 and 1)
:param gamma: the discount factor
:param train_freq: Update the model every ``train_freq`` steps. Alternatively pass a tuple of frequency and unit
like ``(5, "step")`` or ``(2, "episode")``.
:param gradient_steps: How many gradient steps to do after each rollout (see ``train_freq``)
Set to ``-1`` means to do as many gradient steps as steps done in the environment
during the rollout.
:param action_noise: the action noise type (None by default), this can help
for hard exploration problem. Cf common.noise for the different action noise type.
:param replay_buffer_class: Replay buffer class to use (for instance ``HerReplayBuffer``).
If ``None``, it will be automatically selected.
:param replay_buffer_kwargs: Keyword arguments to pass to the replay buffer on creation.
:param optimize_memory_usage: Enable a memory efficient variant of the replay buffer
at a cost of more complexity.
See https://github.com/DLR-RM/stable-baselines3/issues/37#issuecomment-637501195
:param ent_coef: Entropy regularization coefficient. (Equivalent to
inverse of reward scale in the original SAC paper.) Controlling exploration/exploitation trade-off.
Set it to 'auto' to learn it automatically (and 'auto_0.1' for using 0.1 as initial value)
:param target_update_interval: update the target network every ``target_network_update_freq``
gradient steps.
:param target_entropy: target entropy when learning ``ent_coef`` (``ent_coef = 'auto'``)
:param use_sde: Whether to use generalized State Dependent Exploration (gSDE)
instead of action noise exploration (default: False)
:param sde_sample_freq: Sample a new noise matrix every n steps when using gSDE
Default: -1 (only sample at the beginning of the rollout)
:param use_sde_at_warmup: Whether to use gSDE instead of uniform sampling
during the warm up phase (before learning starts)
:param create_eval_env: Whether to create a second environment that will be
used for evaluating the agent periodically. (Only available when passing string for the environment)
:param policy_kwargs: additional arguments to be passed to the policy on creation
:param verbose: the verbosity level: 0 no output, 1 info, 2 debug
:param seed: Seed for the pseudo random generators
:param device: Device (cpu, cuda, ...) on which the code should be run.
Setting it to auto, the code will be run on the GPU if possible.
:param _init_setup_model: Whether or not to build the network at the creation of the instance
"""
policy_aliases: Dict[str, Type[BasePolicy]] = {
"MlpPolicy": MlpPolicy,
"CnnPolicy": CnnPolicy,
"MultiInputPolicy": MultiInputPolicy,
}
def __init__(
self,
policy: Union[str, Type[SACPolicy]],
env: Union[GymEnv, str],
learning_rate: Union[float, Schedule] = 3e-4,
buffer_size: int = 1_000_000, # 1e6
learning_starts: int = 100,
batch_size: int = 256,
tau: float = 0.005,
gamma: float = 0.99,
train_freq: Union[int, Tuple[int, str]] = 1,
gradient_steps: int = 1,
action_noise: Optional[ActionNoise] = None,
replay_buffer_class: Optional[ReplayBuffer] = None,
replay_buffer_kwargs: Optional[Dict[str, Any]] = None,
optimize_memory_usage: bool = False,
ent_coef: Union[str, float] = "auto",
target_update_interval: int = 1,
target_entropy: Union[str, float] = "auto",
use_sde: bool = False,
sde_sample_freq: int = -1,
use_sde_at_warmup: bool = False,
tensorboard_log: Optional[str] = None,
create_eval_env: bool = False,
policy_kwargs: Optional[Dict[str, Any]] = None,
verbose: int = 0,
seed: Optional[int] = None,
device: Union[th.device, str] = "auto",
_init_setup_model: bool = True,
):
super().__init__(
policy,
env,
learning_rate,
buffer_size,
learning_starts,
batch_size,
tau,
gamma,
train_freq,
gradient_steps,
action_noise,
replay_buffer_class=replay_buffer_class,
replay_buffer_kwargs=replay_buffer_kwargs,
policy_kwargs=policy_kwargs,
tensorboard_log=tensorboard_log,
verbose=verbose,
device=device,
create_eval_env=create_eval_env,
seed=seed,
use_sde=use_sde,
sde_sample_freq=sde_sample_freq,
use_sde_at_warmup=use_sde_at_warmup,
optimize_memory_usage=optimize_memory_usage,
supported_action_spaces=(gym.spaces.Box),
support_multi_env=True,
)
self.target_entropy = target_entropy
self.log_ent_coef = None # type: Optional[th.Tensor]
# Entropy coefficient / Entropy temperature
# Inverse of the reward scale
self.ent_coef = ent_coef
self.target_update_interval = target_update_interval
self.ent_coef_optimizer = None
if _init_setup_model:
self._setup_model()
def _setup_model(self) -> None:
super()._setup_model()
self._create_aliases()
# Target entropy is used when learning the entropy coefficient
if self.target_entropy == "auto":
# automatically set target entropy if needed
self.target_entropy = -np.prod(self.env.action_space.shape).astype(np.float32)
else:
# Force conversion
# this will also throw an error for unexpected string
self.target_entropy = float(self.target_entropy)
# The entropy coefficient or entropy can be learned automatically
# see Automating Entropy Adjustment for Maximum Entropy RL section
# of https://arxiv.org/abs/1812.05905
if isinstance(self.ent_coef, str) and self.ent_coef.startswith("auto"):
# Default initial value of ent_coef when learned
init_value = 1.0
if "_" in self.ent_coef:
init_value = float(self.ent_coef.split("_")[1])
assert init_value > 0.0, "The initial value of ent_coef must be greater than 0"
# Note: we optimize the log of the entropy coeff which is slightly different from the paper
# as discussed in https://github.com/rail-berkeley/softlearning/issues/37
self.log_ent_coef = th.log(th.ones(1, device=self.device) * init_value).requires_grad_(True)
self.ent_coef_optimizer = th.optim.Adam([self.log_ent_coef], lr=self.lr_schedule(1))
else:
# Force conversion to float
# this will throw an error if a malformed string (different from 'auto')
# is passed
self.ent_coef_tensor = th.tensor(float(self.ent_coef)).to(self.device)
def _create_aliases(self) -> None:
self.actor = self.policy.actor
self.critic = self.policy.critic
self.critic_target = self.policy.critic_target
def train(self, gradient_steps: int, batch_size: int = 64) -> None:
# Switch to train mode (this affects batch norm / dropout)
self.policy.set_training_mode(True)
# Update optimizers learning rate
optimizers = [self.actor.optimizer, self.critic.optimizer]
if self.ent_coef_optimizer is not None:
optimizers += [self.ent_coef_optimizer]
# Update learning rate according to lr schedule
self._update_learning_rate(optimizers)
ent_coef_losses, ent_coefs = [], []
actor_losses, critic_losses = [], []
for gradient_step in range(gradient_steps):
# Sample replay buffer
replay_data = self.replay_buffer.sample(batch_size, env=self._vec_normalize_env)
# We need to sample because `log_std` may have changed between two gradient steps
if self.use_sde:
self.actor.reset_noise()
# Action by the current actor for the sampled state
actions_pi, log_prob = self.actor.action_log_prob(replay_data.observations)
log_prob = log_prob.reshape(-1, 1)
ent_coef_loss = None
if self.ent_coef_optimizer is not None:
# Important: detach the variable from the graph
# so we don't change it with other losses
# see https://github.com/rail-berkeley/softlearning/issues/60
ent_coef = th.exp(self.log_ent_coef.detach())
ent_coef_loss = -(self.log_ent_coef * (log_prob + self.target_entropy).detach()).mean()
ent_coef_losses.append(ent_coef_loss.item())
else:
ent_coef = self.ent_coef_tensor
ent_coefs.append(ent_coef.item())
# Optimize entropy coefficient, also called
# entropy temperature or alpha in the paper
if ent_coef_loss is not None:
self.ent_coef_optimizer.zero_grad()
ent_coef_loss.backward()
self.ent_coef_optimizer.step()
with th.no_grad():
# Select action according to policy
next_actions, next_log_prob = self.actor.action_log_prob(replay_data.next_observations)
# Compute the next Q values: min over all critics targets
next_q_values = th.cat(self.critic_target(replay_data.next_observations, next_actions), dim=1)
next_q_values, _ = th.min(next_q_values, dim=1, keepdim=True)
# add entropy term
next_q_values = next_q_values - ent_coef * next_log_prob.reshape(-1, 1)
# td error + entropy term
target_q_values = replay_data.rewards + (1 - replay_data.dones) * self.gamma * next_q_values
# Get current Q-values estimates for each critic network
# using action from the replay buffer
current_q_values = self.critic(replay_data.observations, replay_data.actions)
# Compute critic loss
critic_loss = 0.5 * sum(F.mse_loss(current_q, target_q_values) for current_q in current_q_values)
critic_losses.append(critic_loss.item())
# Optimize the critic
self.critic.optimizer.zero_grad()
critic_loss.backward()
self.critic.optimizer.step()
# Compute actor loss
# Alternative: actor_loss = th.mean(log_prob - qf1_pi)
# Mean over all critic networks
q_values_pi = th.cat(self.critic(replay_data.observations, actions_pi), dim=1)
min_qf_pi, _ = th.min(q_values_pi, dim=1, keepdim=True)
actor_loss = (ent_coef * log_prob - min_qf_pi).mean()
actor_losses.append(actor_loss.item())
# Optimize the actor
self.actor.optimizer.zero_grad()
actor_loss.backward()
self.actor.optimizer.step()
# Update target networks
if gradient_step % self.target_update_interval == 0:
polyak_update(self.critic.parameters(), self.critic_target.parameters(), self.tau)
self._n_updates += gradient_steps
self.logger.record("train/n_updates", self._n_updates, exclude="tensorboard")
self.logger.record("train/ent_coef", np.mean(ent_coefs))
self.logger.record("train/actor_loss", np.mean(actor_losses))
self.logger.record("train/critic_loss", np.mean(critic_losses))
if len(ent_coef_losses) > 0:
self.logger.record("train/ent_coef_loss", np.mean(ent_coef_losses))
def learn(
self,
total_timesteps: int,
callback: MaybeCallback = None,
log_interval: int = 4,
eval_env: Optional[GymEnv] = None,
eval_freq: int = -1,
n_eval_episodes: int = 5,
tb_log_name: str = "SAC",
eval_log_path: Optional[str] = None,
reset_num_timesteps: bool = True,
) -> OffPolicyAlgorithm:
return super().learn(
total_timesteps=total_timesteps,
callback=callback,
log_interval=log_interval,
eval_env=eval_env,
eval_freq=eval_freq,
n_eval_episodes=n_eval_episodes,
tb_log_name=tb_log_name,
eval_log_path=eval_log_path,
reset_num_timesteps=reset_num_timesteps,
)
def _excluded_save_params(self) -> List[str]:
return super()._excluded_save_params() + ["actor", "critic", "critic_target"]
def _get_torch_save_params(self) -> Tuple[List[str], List[str]]:
state_dicts = ["policy", "actor.optimizer", "critic.optimizer"]
if self.ent_coef_optimizer is not None:
saved_pytorch_variables = ["log_ent_coef"]
state_dicts.append("ent_coef_optimizer")
else:
saved_pytorch_variables = ["ent_coef_tensor"]
return state_dicts, saved_pytorch_variables