yolo3 debug start

CMakeLists.txt

@@ -77,6 +77,9 @@ target_link_libraries(test_yolo_224 tkDNN)
 
 add_executable(test_yolo_berkeley tests/yolo_berkeley/yolo_berkeley.cpp)
 target_link_libraries(test_yolo_berkeley tkDNN)
+
+add_executable(test_yolo3_berkeley tests/yolo3_berkeley/yolo3_berkeley.cpp)
+target_link_libraries(test_yolo3_berkeley tkDNN)
 ################################################################################
 
 

src/utils.cpp

@@ -70,7 +70,7 @@ void printDeviceVector(int size, dnnType* vec_d, bool device)
 int checkResult(int size, dnnType *data_d, dnnType *correct_d, bool device) {
 
     dnnType *data_h, *correct_h;
-    const float eps = 0.001f;
+    const float eps = 0.02f;
 
     if(device) {
         data_h = new dnnType[size];

tests/yolo3_berkeley/yolo3_berkeley.cfg

@@ -0,0 +1,785 @@
+[net]
+# Testing
+batch=1
+subdivisions=1
+# Training
+#batch=32
+#subdivisions=8
+width=544
+height=320
+channels=3
+momentum=0.9
+decay=0.0005
+angle=0
+saturation = 1.5
+exposure = 1.5
+hue=.1
+
+learning_rate=0.001
+burn_in=1000
+max_batches = 50200
+policy=steps
+steps=40000,45000
+scales=.1,.1
+
+
+
+[convolutional]
+batch_normalize=1
+filters=32
+size=3
+stride=1
+pad=1
+activation=leaky
+
+# Downsample
+
+[convolutional]
+batch_normalize=1
+filters=64
+size=3
+stride=2
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=32
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=64
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[shortcut]
+from=-3
+activation=linear
+
+# Downsample
+
+[convolutional]
+batch_normalize=1
+filters=128
+size=3
+stride=2
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=64
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=128
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[shortcut]
+from=-3
+activation=linear
+
+[convolutional]
+batch_normalize=1
+filters=64
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=128
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[shortcut]
+from=-3
+activation=linear
+
+# Downsample
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=2
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=128
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[shortcut]
+from=-3
+activation=linear
+
+[convolutional]
+batch_normalize=1
+filters=128
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[shortcut]
+from=-3
+activation=linear
+
+[convolutional]
+batch_normalize=1
+filters=128
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[shortcut]
+from=-3
+activation=linear
+
+[convolutional]
+batch_normalize=1
+filters=128
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[shortcut]
+from=-3
+activation=linear
+
+
+[convolutional]
+batch_normalize=1
+filters=128
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[shortcut]
+from=-3
+activation=linear
+
+[convolutional]
+batch_normalize=1
+filters=128
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[shortcut]
+from=-3
+activation=linear
+
+[convolutional]
+batch_normalize=1
+filters=128
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[shortcut]
+from=-3
+activation=linear
+
+[convolutional]
+batch_normalize=1
+filters=128
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[shortcut]
+from=-3
+activation=linear
+
+# Downsample
+
+[convolutional]
+batch_normalize=1
+filters=512
+size=3
+stride=2
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=512
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[shortcut]
+from=-3
+activation=linear
+
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=512
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[shortcut]
+from=-3
+activation=linear
+
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=512
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[shortcut]
+from=-3
+activation=linear
+
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=512
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[shortcut]
+from=-3
+activation=linear
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=512
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[shortcut]
+from=-3
+activation=linear
+
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=512
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[shortcut]
+from=-3
+activation=linear
+
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=512
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[shortcut]
+from=-3
+activation=linear
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=512
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[shortcut]
+from=-3
+activation=linear
+
+# Downsample
+
+[convolutional]
+batch_normalize=1
+filters=1024
+size=3
+stride=2
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=512
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=1024
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[shortcut]
+from=-3
+activation=linear
+
+[convolutional]
+batch_normalize=1
+filters=512
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=1024
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[shortcut]
+from=-3
+activation=linear
+
+[convolutional]
+batch_normalize=1
+filters=512
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=1024
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[shortcut]
+from=-3
+activation=linear
+
+[convolutional]
+batch_normalize=1
+filters=512
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=1024
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[shortcut]
+from=-3
+activation=linear
+
+######################
+
+[convolutional]
+batch_normalize=1
+filters=512
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+size=3
+stride=1
+pad=1
+filters=1024
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=512
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+size=3
+stride=1
+pad=1
+filters=1024
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=512
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+size=3
+stride=1
+pad=1
+filters=1024
+activation=leaky
+
+[convolutional]
+size=1
+stride=1
+pad=1
+filters=45
+activation=linear
+
+[yolo]
+mask = 6,7,8
+anchors = 8.2087,8.5515, 18.4134,20.3391, 40.2194,29.2990, 31.6137,69.2240, 69.8497,48.3838, 108.8817,76.6316, 96.5753,145.5743, 165.9182,117.4493, 215.7497,198.4648
+classes=10
+num=9
+jitter=.3
+ignore_thresh = .5
+truth_thresh = 1
+random=0
+
+[route]
+layers = -4
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[upsample]
+stride=2
+
+[route]
+layers = -1, 61
+
+
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+size=3
+stride=1
+pad=1
+filters=512
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+size=3
+stride=1
+pad=1
+filters=512
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+size=3
+stride=1
+pad=1
+filters=512
+activation=leaky
+
+[convolutional]
+size=1
+stride=1
+pad=1
+filters=45
+activation=linear
+
+[yolo]
+mask = 3,4,5
+anchors = 8.2087,8.5515, 18.4134,20.3391, 40.2194,29.2990, 31.6137,69.2240, 69.8497,48.3838, 108.8817,76.6316, 96.5753,145.5743, 165.9182,117.4493, 215.7497,198.4648
+classes=10
+num=9
+jitter=.3
+ignore_thresh = .5
+truth_thresh = 1
+random=0
+
+[route]
+layers = -4
+
+[convolutional]
+batch_normalize=1
+filters=128
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[upsample]
+stride=2
+
+[route]
+layers = -1, 36
+
+
+
+[convolutional]
+batch_normalize=1
+filters=128
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+size=3
+stride=1
+pad=1
+filters=256
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=128
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+size=3
+stride=1
+pad=1
+filters=256
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=128
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+size=3
+stride=1
+pad=1
+filters=256
+activation=leaky
+
+[convolutional]
+size=1
+stride=1
+pad=1
+filters=45
+activation=linear
+
+[yolo]
+mask = 0,1,2
+anchors = 8.2087,8.5515, 18.4134,20.3391, 40.2194,29.2990, 31.6137,69.2240, 69.8497,48.3838, 108.8817,76.6316, 96.5753,145.5743, 165.9182,117.4493, 215.7497,198.4648
+classes=10
+num=9
+jitter=.3
+ignore_thresh = .5
+truth_thresh = 1
+random=0
+

tests/yolo3_berkeley/yolo3_berkeley.cpp

@@ -0,0 +1,52 @@
+#include<iostream>
+#include "tkdnn.h"
+
+const char *input_bin  = "../tests/yolo3_berkeley/layers/input.bin";
+const char *c0_bin     = "../tests/yolo3_berkeley/layers/c0.bin";
+const char *c1_bin     = "../tests/yolo3_berkeley/layers/c1.bin";
+const char *c2_bin     = "../tests/yolo3_berkeley/layers/c2.bin";
+const char *c3_bin     = "../tests/yolo3_berkeley/layers/c3.bin";
+const char *output_bin = "../tests/yolo3_berkeley/debug/layer3_out.bin";
+
+int main() {
+
+    // Network layout
+    tk::dnn::dataDim_t dim(1, 3, 320, 544, 1);
+    tk::dnn::Network net(dim);
+
+    tk::dnn::Conv2d     c0 (&net, 32, 3, 3, 1, 1, 1, 1,   c0_bin, true);
+    tk::dnn::Activation a0 (&net, tk::dnn::ACTIVATION_LEAKY);
+    tk::dnn::Conv2d     c1 (&net, 64, 3, 3, 2, 2, 1, 1,   c1_bin, true);
+    tk::dnn::Activation a1 (&net, tk::dnn::ACTIVATION_LEAKY);
+    tk::dnn::Conv2d     c2 (&net, 32, 1, 1, 1, 1, 0, 0,   c2_bin, true);
+    tk::dnn::Activation a2 (&net, tk::dnn::ACTIVATION_LEAKY);
+    tk::dnn::Conv2d     c3 (&net, 64, 3, 3, 1, 1, 1, 1,   c3_bin, true);
+    tk::dnn::Activation a3 (&net, tk::dnn::ACTIVATION_LEAKY);
+
+    // Load input
+    dnnType *data;
+    dnnType *input_h;
+    readBinaryFile(input_bin, dim.tot(), &input_h, &data);
+    
+    //print network model
+    net.print();
+
+
+    dnnType *out_data;  // cudnn output
+
+    tk::dnn::dataDim_t dim1 = dim; //input dim
+    printCenteredTitle(" CUDNN inference ", '=', 30); {
+        dim1.print();
+        TIMER_START
+        out_data = net.infer(dim1, data);    
+        TIMER_STOP
+        dim1.print();   
+    }
+
+    printCenteredTitle(" CHECK RESULTS ", '=', 30);
+    dnnType *out, *out_h;
+    int out_dim = net.getOutputDim().tot();
+    readBinaryFile(output_bin, out_dim, &out_h, &out);
+    std::cout<<"CUDNN vs correct"; checkResult(out_dim, out_data, out);
+    return 0;
+}