fix(pu): fix smz compile_args and num_simulations bug in world_model (#…

…297)

* fix(pu): fix smz compile_args

* fix(pu): fix num_simulations bug in world_model

lzero/config/meta.py

@@ -7,7 +7,7 @@
 __TITLE__ = "LightZero"
 
 #: Version of this project.
-__VERSION__ = "0.0.3"
+__VERSION__ = "0.1.0"
 
 #: Short description of the project, will be included in ``setup.py``.
 __DESCRIPTION__ = 'A lightweight and efficient MCTS/AlphaZero/MuZero algorithm toolkits.'

lzero/entry/__init__.py

@@ -9,3 +9,4 @@
 from .train_rezero import train_rezero
 from .train_unizero import train_unizero
 from .train_unizero_segment import train_unizero_segment
+from .utils import *

lzero/mcts/ctree/ctree_sampled_muzero/lib/cnode.cpp

@@ -610,7 +610,7 @@ namespace tree
         }
     }
 
-    std::vector<std::vector<float> > CNode::get_trajectory()
+    std::vector<std::vector<float>> CNode::get_trajectory()
     {
         /*
         Overview:
@@ -629,7 +629,7 @@ namespace tree
             best_action = node->best_action;
         }
 
-        std::vector<std::vector<float> > traj_return;
+        std::vector<std::vector<float>> traj_return;
         for (int i = 0; i < traj.size(); ++i)
         {
             traj_return.push_back(traj[i].value);
@@ -676,7 +676,7 @@ namespace tree
         this->num_of_sampled_actions = 20;
     }
 
-    CRoots::CRoots(int root_num, std::vector<std::vector<float> > legal_actions_list, int action_space_size, int num_of_sampled_actions, bool continuous_action_space)
+    CRoots::CRoots(int root_num, std::vector<std::vector<float>> legal_actions_list, int action_space_size, int num_of_sampled_actions, bool continuous_action_space)
     {
         /*
         Overview:
@@ -728,7 +728,7 @@ namespace tree
 
     CRoots::~CRoots() {}
 
-    void CRoots::prepare(float root_noise_weight, const std::vector<std::vector<float> > &noises, const std::vector<float> &rewards, const std::vector<std::vector<float> > &policies, std::vector<int> &to_play_batch)
+    void CRoots::prepare(float root_noise_weight, const std::vector<std::vector<float>> &noises, const std::vector<float> &rewards, const std::vector<std::vector<float>> &policies, std::vector<int> &to_play_batch)
     {
         /*
         Overview:

lzero/mcts/ctree/ctree_sampled_muzero/lib/cnode.h

@@ -55,8 +55,8 @@ namespace tree
         CNode(float prior, std::vector<CAction> &legal_actions, int action_space_size, int num_of_sampled_actions, bool continuous_action_space);
         ~CNode();
 
-        // 辅助采样函数
-        std::pair<std::vector<std::vector<float>>, std::vector<float>> sample_actions(
+        // Auxiliary sampling function
+        std::pair<std::vector<std::vector<float> >, std::vector<float> > sample_actions(
             const std::vector<float>& mu,
             const std::vector<float>& sigma,
             int num_samples,

lzero/model/__init__.py

@@ -1 +1,3 @@
 from .image_transform import Intensity, RandomCrop, ImageTransforms
+from .utils import *
+from .common import *

lzero/model/unizero_model.py

@@ -134,8 +134,7 @@ def __init__(
                 self.encoder_hook = FeatureAndGradientHook()
                 self.encoder_hook.setup_hooks(self.representation_network)
 
-            self.tokenizer = Tokenizer(with_lpips=True, encoder=self.representation_network,
-                                       decoder_network=self.decoder_network)
+            self.tokenizer = Tokenizer(encoder=self.representation_network, decoder_network=self.decoder_network)
             self.world_model = WorldModel(config=world_model_cfg, tokenizer=self.tokenizer)
             print(f'{sum(p.numel() for p in self.world_model.parameters())} parameters in agent.world_model')
             print(f'{sum(p.numel() for p in self.world_model.parameters()) - sum(p.numel() for p in self.tokenizer.decoder_network.parameters()) - sum(p.numel() for p in self.tokenizer.lpips.parameters())} parameters in agent.world_model - (decoder_network and lpips)')

lzero/model/unizero_world_models/tokenizer.py

@@ -9,8 +9,6 @@
 from einops import rearrange
 from torch.nn import functional as F
 
-from lzero.model.unizero_world_models.lpips import LPIPS
-
 
 class LossWithIntermediateLosses:
     def __init__(self, **kwargs):
@@ -47,7 +45,12 @@ def __init__(self, encoder=None, decoder_network=None, with_lpips: bool = False)
             with_lpips (bool, optional): Whether to use LPIPS for perceptual loss. Defaults to False.
         """
         super().__init__()
-        self.lpips = LPIPS().eval() if with_lpips else None
+        if with_lpips:
+            from lzero.model.unizero_world_models.lpips import LPIPS
+            self.lpips = LPIPS().eval()
+        else:
+            self.lpips = None
+
         self.encoder = encoder
         self.decoder_network = decoder_network
 

lzero/model/unizero_world_models/world_model.py

@@ -114,7 +114,7 @@ def __init__(self, config: TransformerConfig, tokenizer) -> None:
         # TODO: check the size of the shared pool
         # for self.kv_cache_recurrent_infer
         # If needed, recurrent_infer should store the results of the one MCTS search.
-        self.num_simulations = self.config.num_simulations
+        self.num_simulations = getattr(self.config, 'num_simulations', 50)
         self.shared_pool_size = int(self.num_simulations*self.env_num)
         self.shared_pool_recur_infer = [None] * self.shared_pool_size
         self.shared_pool_index = 0

lzero/policy/utils.py

@@ -198,7 +198,6 @@ def forward(self, input):
         return F.layer_norm(input, self.weight.shape, self.weight, self.bias, 1e-5)
 
 
-
 # modified from https://github.com/karpathy/nanoGPT/blob/master/model.py#L263
 def configure_optimizers_nanogpt(model, weight_decay, learning_rate, betas, device_type):
     # start with all of the candidate parameters

lzero/reward_model/__init__.py

@@ -0,0 +1 @@
+from .rnd_reward_model import *

requirements.txt

@@ -11,3 +11,4 @@ pytest
 pooltool-billiards>=0.3.1
 line_profiler
 xxhash
+einops

setup.py

@@ -11,11 +11,12 @@
 # limitations under the License.
 import os
 import re
+import sys
 from distutils.core import setup
 
 import numpy as np
+from Cython.Build import cythonize
 from setuptools import find_packages, Extension
-from Cython.Build import cythonize  # this line should be after 'from setuptools import find_packages'
 
 here = os.path.abspath(os.path.dirname(__file__))
 
@@ -32,6 +33,19 @@ def _load_req(file: str):
     for item in [_REQ_PATTERN.fullmatch(reqpath) for reqpath in os.listdir()] if item
 }
 
+# Set C++11 compile parameters according to the operating system
+extra_compile_args = []
+extra_link_args = []
+
+if sys.platform == 'win32':
+    # Use the VS compiler on Windows platform
+    extra_compile_args = ["/std:c++11"]
+    extra_link_args = ["/std:c++11"]
+else:
+    # Linux/macOS Platform
+    extra_compile_args = ["-std=c++11"]
+    extra_link_args = ["-std=c++11"]
+
 
 def find_pyx(path=None):
     path = path or os.path.join(here, 'lzero')
@@ -60,8 +74,8 @@ def find_cython_extensions(path=None):
             extname, [item],
             include_dirs=[np.get_include()],
             language="c++",
-            # extra_compile_args=["/std:c++latest"],  # only for Windows
-            # extra_link_args=["/std:c++latest"],  # only for Windows
+            extra_compile_args=extra_compile_args,
+            extra_link_args=extra_link_args,
         ))
 
     return extensions