some updates

CMakeLists.txt

@@ -1,11 +1,12 @@
 CMAKE_MINIMUM_REQUIRED(VERSION 3.5)
 
-PROJECT(RM-AutoAim)
+PROJECT(AutoAim)
 SET(CMAKE_CXX_STANDARD 11)
 SET(CMAKE_BUILD_TYPE RELEASE)
 SET(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -O3")
 SET(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -DPATH=\"\\\"${PROJECT_SOURCE_DIR}\\\"\"")
 SET(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -D${CMAKE_SYSTEM_NAME}")
+
 # Todo
 ## 使用编译期固定选项，以略微提升性能。
 #SET(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -DFIX_OPTIONS")
@@ -21,17 +22,28 @@ IF(CCACHE_FOUND)
     MESSAGE("< Use ccache for compiler >")
 ENDIF()
 
+FIND_PACKAGE(Eigen3 REQUIRED)
 FIND_PACKAGE(OpenCV 3 REQUIRED)
 FIND_PACKAGE(Threads)
 
-INCLUDE_DIRECTORIES("D:/Program\ Files/eigen3.3.7")
+INCLUDE_DIRECTORIES(${EIGEN3_INCLUDE_DIR})
 INCLUDE_DIRECTORIES(${PROJECT_SOURCE_DIR}/energy/include)
 INCLUDE_DIRECTORIES(${PROJECT_SOURCE_DIR}/armor/include)
 INCLUDE_DIRECTORIES(${PROJECT_SOURCE_DIR}/others/include)
 
 FILE(GLOB_RECURSE sourcefiles "others/src/*.cpp" "energy/src/*cpp" "armor/src/*.cpp")
-ADD_EXECUTABLE(${BIN_NAME} main.cpp ${sourcefiles} )
+ADD_EXECUTABLE(${BIN_NAME} main.cpp ${sourcefiles} others/include/additions/additions.h)
 
 TARGET_LINK_LIBRARIES(${BIN_NAME} ${CMAKE_THREAD_LIBS_INIT})
 TARGET_LINK_LIBRARIES(${BIN_NAME} ${OpenCV_LIBS})
+IF (CMAKE_SYSTEM_NAME MATCHES "Linux")
+    MESSAGE(STATUS "current platform: Linux ")
+    TARGET_LINK_LIBRARIES(${BIN_NAME} "${PROJECT_SOURCE_DIR}/others/libMVSDK.so")
+ELSEIF (CMAKE_SYSTEM_NAME MATCHES "Windows")
+    MESSAGE(STATUS "current platform: Windows")
     TARGET_LINK_LIBRARIES(${BIN_NAME} "${PROJECT_SOURCE_DIR}/others/MVCAMSDK_X64.lib")
+ELSE ()
+    MESSAGE(STATUS "Unsupport platform: ${CMAKE_SYSTEM_NAME}")
+ENDIF()
+
+ADD_CUSTOM_TARGET(bind-monitor COMMAND "${PROJECT_SOURCE_DIR}/tools/bind-monitor.sh" "${PROJECT_SOURCE_DIR}" "${CMAKE_BINARY_DIR}")

tools/TrainCNN/backward.py

@@ -75,7 +75,7 @@ def train(dataset, show_bar=False):
     learning_rate = tf.train.exponential_decay(
         LEARNING_RATE_BASE,
         global_step,
-        len(dataset.train_samples) / BATCH,
+        len(dataset.train_samples) / BATCH / 5,
         LEARNING_RATE_DECAY,
         staircase=False)
     train_step = tf.train.AdamOptimizer(learning_rate).minimize(loss, global_step=global_step)

tools/TrainCNN/forward.py

@@ -71,14 +71,16 @@ def forward(x, regularizer=None):
     pool_shape = pool2.get_shape().as_list()
     node = pool_shape[1] * pool_shape[2] * pool_shape[3]
     reshaped = tf.reshape(pool2, [-1, node])
+#    reshaped = tf.nn.dropout(reshaped, 0.1)
 
-    fc1_w = tf.nn.dropout(get_weight([node, FC1_OUTPUT_NODES], regularizer), 0.1)
+    fc1_w = get_weight([node, FC1_OUTPUT_NODES], regularizer)
     fc1_b = get_bias([FC1_OUTPUT_NODES])
     fc1   = tf.nn.relu(tf.matmul(reshaped, fc1_w) + fc1_b)
+#    fc1   = tf.nn.dropout(fc1, 0.2)
     vars.extend([fc1_w, fc1_b])
     nodes.extend([fc1])
 
-    fc2_w = tf.nn.dropout(get_weight([FC1_OUTPUT_NODES, FC2_OUTPUT_NODES], regularizer), 0.1)
+    fc2_w = get_weight([FC1_OUTPUT_NODES, FC2_OUTPUT_NODES], regularizer)
     fc2_b = get_bias([FC2_OUTPUT_NODES])
     fc2   = tf.nn.softmax(tf.matmul(fc1, fc2_w) + fc2_b)
     vars.extend([fc2_w, fc2_b])

tools/TrainCNN/generate.py

@@ -41,6 +41,13 @@ class DataSet:
 
     def generate_data_sets(self, folder):
         sets = []
+        mini = 1000000
+        for i in range(OUTPUT_NODES):
+            dir = "%s/%d" % (folder, i)
+            files = os.listdir(dir)
+            if mini > len(files):
+                mini = len(files)
+
         for i in range(OUTPUT_NODES):
             dir = "%s/%d" % (folder, i)
             files = os.listdir(dir)