Impulse wars 4.0 bindings

.gitignore

@@ -162,11 +162,11 @@ raylib*/
 box2d*/
 
 # Temp Impulse Wars files
-pufferlib/ocean/impulse_wars/*-debug/
-pufferlib/ocean/impulse_wars/*-release/
-pufferlib/ocean/impulse_wars/debug-*/
-pufferlib/ocean/impulse_wars/release-*/
-pufferlib/ocean/impulse_wars/benchmark/
+ocean/impulse_wars/*-debug/
+ocean/impulse_wars/*-release/
+ocean/impulse_wars/debug-*/
+ocean/impulse_wars/release-*/
+ocean/impulse_wars/benchmark/
 
 # Data
 resources/drive/data/*

build.sh

@@ -50,20 +50,23 @@ if [ "$ENV" = "all" ]; then
     exit 0
 fi
 
+STANDALONE_LDFLAGS=(-fuse-ld=lld)
+SHARED_LDFLAGS=(-fuse-ld=lld)
+
 # Linux/mac
 PLATFORM="$(uname -s)"
 if [ "$PLATFORM" = "Linux" ]; then
     RAYLIB_NAME='raylib-5.5_linux_amd64'
     OMP_LIB=-lomp5
     SANITIZE_FLAGS=(-fsanitize=address,undefined,bounds,pointer-overflow,leak -fno-omit-frame-pointer)
-    STANDALONE_LDFLAGS=(-lGL)
-    SHARED_LDFLAGS=(-Bsymbolic-functions)
+    STANDALONE_LDFLAGS+=(-lGL)
+    SHARED_LDFLAGS+=(-Bsymbolic-functions)
 else
     RAYLIB_NAME='raylib-5.5_macos'
     OMP_LIB=-lomp
     SANITIZE_FLAGS=()
-    STANDALONE_LDFLAGS=(-framework Cocoa -framework IOKit -framework CoreVideo -framework OpenGL)
-    SHARED_LDFLAGS=(-framework Cocoa -framework OpenGL -framework IOKit -undefined dynamic_lookup)
+    STANDALONE_LDFLAGS+=(-framework Cocoa -framework IOKit -framework CoreVideo -framework OpenGL)
+    SHARED_LDFLAGS+=(-framework Cocoa -framework OpenGL -framework IOKit -undefined dynamic_lookup)
 fi
 
 CLANG_WARN=(
@@ -75,6 +78,7 @@ CLANG_WARN=(
     -Wno-incompatible-pointer-types-discards-qualifiers
     -Wno-error=array-parameter
 )
+CLANG_OPT=()
 
 download() {
     local name=$1 url=$2
@@ -108,14 +112,24 @@ elif [ "$ENV" = "trailer" ]; then
     OUTPUT_NAME="trailer/trailer"
 elif [ "$ENV" = "impulse_wars" ]; then
     SRC_DIR="ocean/$ENV"
-    if [ "$MODE" = "web" ]; then BOX2D_NAME='box2d-web'
-    elif [ "$PLATFORM" = "Linux" ]; then BOX2D_NAME='box2d-linux-amd64'
-    else BOX2D_NAME='box2d-macos-arm64'
+    if [ "$MODE" = "web" ]; then
+        BOX2D_NAME='box2d-web'
+    elif [ "$PLATFORM" = "Linux" ]; then
+        BOX2D_NAME='box2d-linux-amd64'
+    else
+        BOX2D_NAME='box2d-macos-arm64'
     fi
+
     BOX2D_URL="https://github.com/capnspacehook/box2d/releases/latest/download"
     download "$BOX2D_NAME" "$BOX2D_URL/$BOX2D_NAME.tar.gz"
     INCLUDES+=(-I./$BOX2D_NAME/include -I./$BOX2D_NAME/src)
-    LINK_ARCHIVES+=("./$BOX2D_NAME/libbox2d.a")
+
+    if [ -z "$DEBUG" ]; then
+        CLANG_OPT+=(-flto -fno-math-errno -march=native)
+        LINK_ARCHIVES+=("./$BOX2D_NAME/libbox2d.a")
+    else
+        LINK_ARCHIVES+=("./$BOX2D_NAME/libbox2dd.a")
+    fi
 elif [ -d "ocean/$ENV" ]; then
     SRC_DIR="ocean/$ENV"
 else
@@ -126,13 +140,13 @@ OUTPUT_NAME=${OUTPUT_NAME:-$ENV}
 
 # Standalone environment build
 if [ -n "$DEBUG" ] || [ "$MODE" = "local" ]; then
-    CLANG_OPT=(-g -O0 "${CLANG_WARN[@]}" "${SANITIZE_FLAGS[@]}")
+    CLANG_OPT+=(-g -O0 "${CLANG_WARN[@]}" "${SANITIZE_FLAGS[@]}")
     NVCC_OPT="-O0 -g"
     LINK_OPT="-g"
 else
-    CLANG_OPT=(-O2 -DNDEBUG "${CLANG_WARN[@]}")
-    NVCC_OPT="-O2 --threads 0"
-    LINK_OPT="-O2"
+    CLANG_OPT+=(-O3 -DNDEBUG "${CLANG_WARN[@]}")
+    NVCC_OPT="-O3 --threads 0"
+    LINK_OPT="-O3"
 fi
 if [ "$MODE" = "local" ] || [ "$MODE" = "fast" ]; then
     FLAGS=(
@@ -242,6 +256,7 @@ echo "Compiling static library for $ENV..."
 ${CC:-clang} -c "${CLANG_OPT[@]}" $EXTRA_CFLAGS \
     -I. -Isrc -I$SRC_DIR -Ivendor \
     -I./$RAYLIB_NAME/include -I$CUDA_HOME/include \
+    "${INCLUDES[@]}" \
     -DPLATFORM_DESKTOP \
     -fno-semantic-interposition -fvisibility=hidden \
     -fPIC -fopenmp \
@@ -273,7 +288,7 @@ if [ -z "$MODE" ]; then
 
     LINK_CMD=(
         ${CXX:-g++} -shared -fPIC -fopenmp
-        build/bindings.o "$STATIC_LIB" "$RAYLIB_A"
+        build/bindings.o "$STATIC_LIB" "${LINK_ARCHIVES[@]}"
         -L$CUDA_HOME/lib64 $CUDNN_LFLAG $NCCL_LFLAG
         "${WHEEL_RPATH_FLAGS[@]}"
         -lcudart -lnccl -lnvidia-ml -lcublas -lcusolver -lcurand -lcudnn
@@ -298,7 +313,7 @@ elif [ "$MODE" = "cpu" ]; then
         src/bindings_cpu.cpp -o build/bindings_cpu.o
     LINK_CMD=(
         ${CXX:-g++} -shared -fPIC -fopenmp
-        build/bindings_cpu.o "$STATIC_LIB" "$RAYLIB_A"
+        build/bindings_cpu.o "$STATIC_LIB" "${LINK_ARCHIVES[@]}"
         -lm -lpthread $OMP_LIB $LINK_OPT
         "${SHARED_LDFLAGS[@]}"
         -o "$OUTPUT"
@@ -317,7 +332,7 @@ elif [ "$MODE" = "profile" ]; then
         $PRECISION \
         -Xcompiler=-fopenmp \
         tests/profile_kernels.cu vendor/ini.c \
-        "$STATIC_LIB" "$RAYLIB_A" \
+        "$STATIC_LIB" "${LINK_ARCHIVES[@]}" \
         -lnccl -lnvidia-ml -lcublas -lcurand -lcudnn \
         -lGL -lm -lpthread $OMP_LIB \
         -o profile

config/impulse_wars.ini

@@ -1,143 +1,25 @@
 [base]
 env_name = impulse_wars
 
-max_suggestion_cost = 10_800
-
-[policy]
-hidden_size = 512
-cnn_channels = 64
-
-# These must match what's set in env below
-continuous = False
-num_drones = 2
-is_training = True
-
-[vec]
-num_envs = 4
-#num_workers = 4
-#batch_size = 4
-
 [env]
-num_envs = 1024
 num_drones = 2
 num_agents = 1
 enable_teams = False
-sitting_duck = False
+sitting_duck = True
 continuous = False
 is_training = True
 
+reward_win = 2.0
+reward_self_kill = -1.0
+reward_enemy_death = 1.0
+reward_enemy_kill = 1.0
+reward_death = 0.0
+reward_energy_emptied = -0.75
+reward_weapon_pickup = 0.5
+reward_shield_break = 0.5
+reward_shot_hit_coef = 0.005
+reward_explosion_hit_coef = 0.005
+
 [train]
 total_timesteps = 1_000_000_000
 checkpoint_interval = 250
-
-learning_rate = 0.005
-
-compile = False
-compile_mode = reduce-overhead
-compile_fullgraph = False
-
-
-[sweep]
-downsample = 10
-max_cost = 900
-
-[sweep.env.num_envs]
-distribution = uniform_pow2
-min = 1
-max = 1024
-mean = 128
-scale = auto
-
-# reward parameters
-[sweep.env.reward_win]
-distribution = uniform
-min = 0.0
-mean = 2.0
-max = 5.0
-scale = auto
-
-[sweep.env.reward_self_kill]
-distribution = uniform
-min = -3.0
-mean = -1.0
-max = 0.0
-scale = auto
-
-[sweep.env.reward_enemy_death]
-distribution = uniform
-min = 0.0
-mean = 1.0
-max = 3.0
-scale = auto
-
-[sweep.env.reward_kill]
-distribution = uniform
-min = 0.0
-mean = 1.0
-max = 3.0
-scale = auto
-
-[sweep.env.reward_death]
-distribution = uniform
-min = -1.0
-mean = -0.25
-max = 0.0
-scale = auto
-
-[sweep.env.reward_energy_emptied]
-distribution = uniform
-min = -2.0
-mean = -0.75
-max = 0.0
-scale = auto
-
-[sweep.env.reward_weapon_pickup]
-distribution = uniform
-min = 0.0
-mean = 0.5
-max = 3.0
-scale = auto
-
-[sweep.env.reward_shield_break]
-distribution = uniform
-min = 0.0
-mean = 0.5
-max = 3.0
-scale = auto
-
-[sweep.env.reward_shot_hit_coef]
-distribution = log_normal
-min = 0.0005
-mean = 0.005
-max = 0.05
-scale = auto
-
-[sweep.env.reward_explosion_hit_coef]
-distribution = log_normal
-min = 0.0005
-mean = 0.005
-max = 0.05
-scale = auto
-
-# hyperparameters
-[sweep.train.total_timesteps]
-distribution = log_normal
-min = 250_000_000
-max = 1_500_000_000
-mean = 500_000_000
-scale = time
-
-[sweep.train.batch_size]
-distribution = uniform_pow2
-min = 65_536
-max = 1_048_576
-mean = 262_144
-scale = auto
-
-[sweep.train.horizon]
-distribution = uniform_pow2
-min = 64
-max = 256
-mean = 128
-scale = auto
-