new models 🐍 🐍 🐍

Makefile

@@ -1,7 +1,7 @@
-GPU=0
-CUDNN=0
-OPENCV=0
-OPENMP=0
+GPU=1
+CUDNN=1
+OPENCV=1
+OPENMP=1
 DEBUG=0
 
 ARCH= -gencode arch=compute_30,code=sm_30 \

cfg/alexnet.cfg

@@ -1,5 +1,9 @@
 [net]
-batch=128
+# Training
+# batch=128
+# subdivisions=1
+# Testing
+batch=1
 subdivisions=1
 height=227
 width=227

cfg/coco.data

@@ -1,7 +1,7 @@
 classes= 80
 train  = /home/pjreddie/data/coco/trainvalno5k.txt
-#valid  = coco_testdev
-valid = data/coco_val_5k.list
+valid  = coco_testdev
+#valid = data/coco_val_5k.list
 names = data/coco.names
 backup = /home/pjreddie/backup/
 eval=coco

cfg/darknet.cfg

@@ -1,21 +1,30 @@
 [net]
-# Train
- batch=128
- subdivisions=1
-# Test
-#batch=1
-#subdivisions=1
+# Training
+# batch=128
+# subdivisions=1
+# Testing
+batch=1
+subdivisions=1
 height=256
 width=256
+min_crop=128
+max_crop=448
 channels=3
 momentum=0.9
 decay=0.0005
-max_crop=320
 
+burn_in=1000
 learning_rate=0.1
 policy=poly
 power=4
-max_batches=1600000
+max_batches=800000
+
+angle=7
+hue=.1
+saturation=.75
+exposure=.75
+aspect=.75
+
 
 [convolutional]
 batch_normalize=1
@@ -97,14 +106,14 @@ stride=1
 pad=1
 activation=leaky
 
+[avgpool]
+
 [convolutional]
 filters=1000
 size=1
 stride=1
 pad=1
-activation=leaky
-
-[avgpool]
+activation=linear
 
 [softmax]
 groups=1

cfg/gru.cfg

@@ -1,10 +1,10 @@
 [net]
-subdivisions=1
-batch = 256
 inputs=256
 momentum=0.9
 decay=0.0
-time_steps=128
+subdivisions=1
+batch = 1
+time_steps=1
 learning_rate=.002
 adam=1
 
@@ -13,13 +13,13 @@ power=4
 max_batches=1000000
 
 [gru]
-output = 1024
+output = 256
 
 [gru]
-output = 1024
+output = 256
 
 [gru]
-output = 1024
+output = 256
 
 [connected]
 output=256
@@ -27,4 +27,3 @@ activation=linear
 
 [softmax]
 
-

cfg/resnet101.cfg

@@ -0,0 +1,990 @@
+[net]
+# Training
+# batch=128
+# subdivisions=2
+
+# Testing
+batch=1
+subdivisions=1
+
+height=256
+width=256
+channels=3
+min_crop=128
+max_crop=448
+
+burn_in=1000
+learning_rate=0.1
+policy=poly
+power=4
+max_batches=800000
+momentum=0.9
+decay=0.0005
+
+angle=7
+hue=.1
+saturation=.75
+exposure=.75
+aspect=.75
+
+
+
+[convolutional]
+batch_normalize=1
+filters=64
+size=7
+stride=2
+pad=1
+activation=leaky
+
+[maxpool]
+size=2
+stride=2
+
+[convolutional]
+batch_normalize=1
+filters=64
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=64
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=64
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=64
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=64
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=64
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=128
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=128
+size=3
+stride=2
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=512
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=128
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=128
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=512
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=128
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=128
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=512
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=128
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=128
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=512
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+
+# Conv 4
+[convolutional]
+batch_normalize=1
+filters=256
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=2
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=1024
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=1024
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=1024
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=1024
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=1024
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=1024
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=1024
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=1024
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=1024
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=1024
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=1024
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=1024
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=1024
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=1024
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=1024
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=1024
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=1024
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=1024
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=1024
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=1024
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=1024
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=1024
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=1024
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+#Conv 5
+[convolutional]
+batch_normalize=1
+filters=512
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=512
+size=3
+stride=2
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=2048
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=512
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=512
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=2048
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=512
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=512
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=2048
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+
+
+
+
+
+[convolutional]
+filters=1000
+size=1
+stride=1
+pad=1
+activation=linear
+
+[avgpool]
+
+[softmax]
+groups=1
+
+[cost]
+type=sse
+

cfg/resnet18.cfg

@@ -0,0 +1,228 @@
+[net]
+# Training
+# batch=128
+# subdivisions=1
+
+# Testing
+batch=1
+subdivisions=1
+
+height=256
+width=256
+channels=3
+min_crop=128
+max_crop=448
+
+burn_in=1000
+learning_rate=0.1
+policy=poly
+power=4
+max_batches=800000
+momentum=0.9
+decay=0.0005
+
+angle=7
+hue=.1
+saturation=.75
+exposure=.75
+aspect=.75
+
+
+[convolutional]
+batch_normalize=1
+filters=64
+size=7
+stride=2
+pad=1
+activation=leaky
+
+[maxpool]
+size=2
+stride=2
+
+
+# Residual Block
+[convolutional]
+batch_normalize=1
+filters=64
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=64
+size=3
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+activation=leaky
+from=-3
+
+# Residual Block
+[convolutional]
+batch_normalize=1
+filters=64
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=64
+size=3
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+activation=leaky
+from=-3
+
+# Strided Residual Block
+[convolutional]
+batch_normalize=1
+filters=128
+size=3
+stride=2
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=128
+size=3
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+activation=leaky
+from=-3
+
+# Residual Block
+[convolutional]
+batch_normalize=1
+filters=128
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=128
+size=3
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+activation=leaky
+from=-3
+
+
+# Strided Residual Block
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=2
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+activation=leaky
+from=-3
+
+# Residual Block
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+activation=leaky
+from=-3
+
+
+# Strided Residual Block
+[convolutional]
+batch_normalize=1
+filters=512
+size=3
+stride=2
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=512
+size=3
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+activation=leaky
+from=-3
+
+# Residual Block
+[convolutional]
+batch_normalize=1
+filters=512
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=512
+size=3
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+activation=leaky
+from=-3
+
+
+
+
+[avgpool]
+
+[convolutional]
+filters=1000
+size=1
+stride=1
+pad=1
+activation=linear
+
+[softmax]
+groups=1
+

cfg/resnet34.cfg

@@ -0,0 +1,392 @@
+[net]
+# Training
+# batch=128
+# subdivisions=2
+
+# Testing
+batch=1
+subdivisions=1
+
+height=256
+width=256
+channels=3
+min_crop=128
+max_crop=448
+
+burn_in=1000
+learning_rate=0.1
+policy=poly
+power=4
+max_batches=800000
+momentum=0.9
+decay=0.0005
+
+angle=7
+hue=.1
+saturation=.75
+exposure=.75
+aspect=.75
+
+
+[convolutional]
+batch_normalize=1
+filters=64
+size=7
+stride=2
+pad=1
+activation=leaky
+
+[maxpool]
+size=2
+stride=2
+
+# Residual Block
+[convolutional]
+batch_normalize=1
+filters=64
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=64
+size=3
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+activation=leaky
+from=-3
+
+# Residual Block
+[convolutional]
+batch_normalize=1
+filters=64
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=64
+size=3
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+activation=leaky
+from=-3
+
+# Residual Block
+[convolutional]
+batch_normalize=1
+filters=64
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=64
+size=3
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+activation=leaky
+from=-3
+
+# Strided Residual Block
+[convolutional]
+batch_normalize=1
+filters=128
+size=3
+stride=2
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=128
+size=3
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+activation=leaky
+from=-3
+
+# Residual Block
+[convolutional]
+batch_normalize=1
+filters=128
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=128
+size=3
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+activation=leaky
+from=-3
+
+# Residual Block
+[convolutional]
+batch_normalize=1
+filters=128
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=128
+size=3
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+activation=leaky
+from=-3
+
+# Residual Block
+[convolutional]
+batch_normalize=1
+filters=128
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=128
+size=3
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+activation=leaky
+from=-3
+
+# Strided Residual Block
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=2
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+activation=leaky
+from=-3
+
+# Residual Block
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+activation=leaky
+from=-3
+
+# Residual Block
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+activation=leaky
+from=-3
+
+# Residual Block
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+activation=leaky
+from=-3
+
+# Residual Block
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+activation=leaky
+from=-3
+
+# Residual Block
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+activation=leaky
+from=-3
+
+# Residual Block
+[convolutional]
+batch_normalize=1
+filters=512
+size=3
+stride=2
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=512
+size=3
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+activation=leaky
+from=-3
+
+# Residual Block
+[convolutional]
+batch_normalize=1
+filters=512
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=512
+size=3
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+activation=leaky
+from=-3
+
+# Residual Block
+[convolutional]
+batch_normalize=1
+filters=512
+size=3
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=512
+size=3
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+activation=leaky
+from=-3
+
+
+
+[avgpool]
+
+[convolutional]
+filters=1000
+size=1
+stride=1
+pad=1
+activation=linear
+
+[softmax]
+groups=1
+

cfg/resnet50.cfg

@@ -9,16 +9,17 @@ subdivisions=1
 
 height=256
 width=256
-max_crop=448
 channels=3
-momentum=0.9
-decay=0.0005
+min_crop=128
+max_crop=448
 
 burn_in=1000
 learning_rate=0.1
 policy=poly
 power=4
-max_batches=1600000
+max_batches=800000
+momentum=0.9
+decay=0.0005
 
 angle=7
 hue=.1
@@ -26,6 +27,7 @@ saturation=.75
 exposure=.75
 aspect=.75
 
+
 [convolutional]
 batch_normalize=1
 filters=64
@@ -493,6 +495,7 @@ activation=leaky
 
 
 
+[avgpool]
 
 [convolutional]
 filters=1000
@@ -501,8 +504,6 @@ stride=1
 pad=1
 activation=linear
 
-[avgpool]
-
 [softmax]
 groups=1
 

cfg/resnext50.cfg

@@ -0,0 +1,523 @@
+[net]
+# Training
+# batch=128
+# subdivisions=4
+
+# Testing
+batch=1
+subdivisions=1
+
+height=256
+width=256
+channels=3
+min_crop=128
+max_crop=448
+
+burn_in=1000
+learning_rate=0.1
+policy=poly
+power=4
+max_batches=800000
+momentum=0.9
+decay=0.0005
+
+angle=7
+hue=.1
+saturation=.75
+exposure=.75
+aspect=.75
+
+
+
+[convolutional]
+batch_normalize=1
+filters=64
+size=7
+stride=2
+pad=1
+activation=leaky
+
+[maxpool]
+size=2
+stride=2
+
+[convolutional]
+batch_normalize=1
+filters=128
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=128
+size=3
+groups=32
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=128
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=128
+size=3
+groups=32
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=128
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=128
+size=3
+groups=32
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+groups=32
+stride=2
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=512
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+groups=32
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=512
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+groups=32
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=512
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=256
+size=3
+groups=32
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=512
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+
+# Conv 4
+[convolutional]
+batch_normalize=1
+filters=512
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=512
+size=3
+groups=32
+stride=2
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=1024
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=512
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=512
+size=3
+groups=32
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=1024
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=512
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=512
+size=3
+groups=32
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=1024
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=512
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=512
+size=3
+groups=32
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=1024
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=512
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=512
+size=3
+groups=32
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=1024
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=512
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=512
+size=3
+groups=32
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=1024
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+#Conv 5
+[convolutional]
+batch_normalize=1
+filters=1024
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=1024
+size=3
+groups=32
+stride=2
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=2048
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=1024
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=1024
+size=3
+groups=32
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=2048
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=1024
+size=1
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=1024
+size=3
+groups=32
+stride=1
+pad=1
+activation=leaky
+
+[convolutional]
+batch_normalize=1
+filters=2048
+size=1
+stride=1
+pad=1
+activation=linear
+
+[shortcut]
+from=-4
+activation=leaky
+
+[avgpool]
+
+[convolutional]
+filters=1000
+size=1
+stride=1
+pad=1
+activation=linear
+
+[softmax]
+groups=1
+
+

cfg/yolov2.cfg

@@ -5,8 +5,8 @@ subdivisions=1
 # Training
 # batch=64
 # subdivisions=8
-width=416
-height=416
+width=608
+height=608
 channels=3
 momentum=0.9
 decay=0.0005

cfg/yolov3-tiny.cfg

@@ -172,7 +172,7 @@ filters=255
 activation=linear
 
 [yolo]
-mask = 1,2,3
+mask = 0,1,2
 anchors = 10,14,  23,27,  37,58,  81,82,  135,169,  344,319
 classes=80
 num=6

cfg/yolov3.cfg

@@ -1,12 +1,12 @@
 [net]
 # Testing
-batch=1
-subdivisions=1
+# batch=1
+# subdivisions=1
 # Training
-# batch=64
-# subdivisions=16
-width=416
-height=416
+batch=64
+subdivisions=16
+width=608
+height=608
 channels=3
 momentum=0.9
 decay=0.0005

examples/classifier.c

@@ -396,6 +396,7 @@ void validate_classifier_single(char *datacfg, char *filename, char *weightfile)
         }
         image im = load_image_color(paths[i], 0, 0);
         image crop = center_crop_image(im, net->w, net->h);
+        //grayscale_image_3c(crop);
         //show_image(im, "orig");
         //show_image(crop, "cropped");
         //cvWaitKey(0);

examples/instance-segmenter.c

@@ -2,6 +2,7 @@
 #include <sys/time.h>
 #include <assert.h>
 
+void normalize_image2(image p);
 void train_isegmenter(char *datacfg, char *cfgfile, char *weightfile, int *gpus, int ngpus, int clear, int display)
 {
     int i;
@@ -26,6 +27,10 @@ void train_isegmenter(char *datacfg, char *cfgfile, char *weightfile, int *gpus,
     network *net = nets[0];
     image pred = get_network_image(net);
 
+    image embed = pred;
+    embed.c = 3;
+    embed.data += embed.w*embed.h*80;
+
     int div = net->w/pred.w;
     assert(pred.w * div == net->w);
     assert(pred.h * div == net->h);
@@ -98,6 +103,11 @@ void train_isegmenter(char *datacfg, char *cfgfile, char *weightfile, int *gpus,
             pred.c = 80;
             image mask = mask_to_rgb(tr);
             image prmask = mask_to_rgb(pred);
+            image ecopy = copy_image(embed);
+            normalize_image2(ecopy);
+            show_image(ecopy, "embed", 1);
+            free_image(ecopy);
+
             show_image(im, "input", 1);
             show_image(prmask, "pred", 1);
             show_image(mask, "truth", 100);

src/data.c

@@ -127,6 +127,7 @@ matrix load_image_augment_paths(char **paths, int n, int min, int max, int size,
         show_image(crop, "crop");
         cvWaitKey(0);
         */
+        //grayscale_image_3c(crop);
         free_image(im);
         X.vals[i] = crop.data;
         X.cols = crop.h*crop.w*crop.c;

src/iseg_layer.c

@@ -109,9 +109,8 @@ void forward_iseg_layer(const layer l, network net)
         }
 
 
-        memset(l.counts, 0, 90*sizeof(float));
+        memset(l.counts, 0, 90*sizeof(int));
         for(i = 0; i < 90; ++i){
-            l.counts[i] = 0;
             fill_cpu(ids, 0, l.sums[i], 1);
             
             int c = net.truth[b*l.truths + i*(l.w*l.h+1)];
@@ -153,7 +152,7 @@ void forward_iseg_layer(const layer l, network net)
             scal_cpu(ids, 1.f/l.counts[i], l.sums[i], 1);
             if(b == 0 && net.gpu_index == 0){
                 printf("%4d, %6.3f, ", l.counts[i], mse[i]);
-                for(j = 0; j < ids/4; ++j){
+                for(j = 0; j < ids; ++j){
                     printf("%6.3f,", l.sums[i][j]);
                 }
                 printf("\n");
@@ -180,6 +179,13 @@ void forward_iseg_layer(const layer l, network net)
                 }
             }
         }
+
+        for(i = 0; i < ids; ++i){
+            for(k = 0; k < l.w*l.h; ++k){
+                int index = b*l.outputs + (i+l.classes)*l.w*l.h + k;
+                l.delta[index] *= .01;
+            }
+        }
     }
 
     *(l.cost) = pow(mag_array(l.delta, l.outputs * l.batch), 2);