Hi ,
I have used 5 files that i have used in DS 4.02 in DS5.0 :
- deepstream_app_config_yoloV3_tiny
[source0]
enable=1
#Type - 1=CameraV4L2 2=URI 3=MultiURI
type=1
#uri=file://…/…/samples/streams/sample_1080p_h264.mp4
uri=file:out.mp4
num-sources=1
gpu-id=0
(0): memtype_device - Memory type Device
(1): memtype_pinned - Memory type Host Pinned
(2): memtype_unified - Memory type Unified
cudadec-memtype=0
camera-width=1920
camera-height=1080
camera-fps-n=30
camera-v4l2-dev-node=0
camera-fps-d=1
#num-extra-surfaces=10
[sink0]
enable=1
#Type - 1=FakeSink 2=EglSink 3=File 4=RTSPStreaming
type=4
#1=mp4 2=mkv
container=2
#1=h264 2=h265 3=mpeg4
gpu-id=0
nvbuf-memory-type=0
#only mpeg4 is supported right now.
codec=1
sync=0
bitrate=4000
#set below properties in case of RTSPStreaming
rtsp-port=8554
udp-port=5400
-
config_infer_primary_yoloV3_tiny.txt:
[property]
gpu-id=0
net-scale-factor=1
#0=RGB, 1=BGR
model-color-format=0
custom-network-config=yolov3-tiny-obj-custom.cfg
#model-file=yolov3-tiny-obj_98000.weights -with out calc anchors
model-file=yolov3-tiny-obj_83000.weights
model-engine-file=model_b1_fp16.engine
labelfile-path=labels_5.txt
0=FP32, 1=INT8, 2=FP16 mode
network-mode=2
num-detected-classes=5
gie-unique-id=1
is-classifier=0
maintain-aspect-ratio=1
parse-bbox-func-name=NvDsInferParseCustomYoloV3Tiny
custom-lib-path=nvdsinfer_custom_impl_Yolo/libnvdsinfer_custom_impl_Yolo.so
5 classes …
-
yolov3-tiny-obj-custom.cfg :
[net]
Testing
batch=1
subdivisions=1
Training
#batch=64
#subdivisions=8
width=416
#width=800
height=416
#height=800
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 = 500200
policy=steps
steps=400000,450000
scales=.1,.1
[convolutional]
batch_normalize=1
filters=16
size=3
stride=1
pad=1
activation=leaky
[maxpool]
size=2
stride=2
[convolutional]
batch_normalize=1
filters=32
size=3
stride=1
pad=1
activation=leaky
[maxpool]
size=2
stride=2
[convolutional]
batch_normalize=1
filters=64
size=3
stride=1
pad=1
activation=leaky
[maxpool]
size=2
stride=2
[convolutional]
batch_normalize=1
filters=128
size=3
stride=1
pad=1
activation=leaky
[maxpool]
size=2
stride=2
[convolutional]
batch_normalize=1
filters=256
size=3
stride=1
pad=1
activation=leaky
[maxpool]
size=2
stride=2
[convolutional]
batch_normalize=1
filters=512
size=3
stride=1
pad=1
activation=leaky
[maxpool]
size=2
stride=1
[convolutional]
batch_normalize=1
filters=1024
size=3
stride=1
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
[convolutional]
size=1
stride=1
pad=1
filters=30
activation=linear
[yolo]
mask = 3,4,5
#anchors = 20, 85, 41, 43, 27, 68, 46, 75, 47,108, 70,116
anchors= 45,75,47,141,105,63,38,226,79,123,101,202
#anchors= 45,75,47,141,105,63,38,226,79,123,101,202
classes=5
num=6
jitter=.3
ignore_thresh = .7
truth_thresh = 1
random=1
[route]
layers = -4
[convolutional]
batch_normalize=1
filters=128
size=1
stride=1
pad=1
activation=leaky
[upsample]
stride=2
[route]
layers = -1, 8
[convolutional]
batch_normalize=1
filters=256
size=3
stride=1
pad=1
activation=leaky
[convolutional]
size=1
stride=1
pad=1
filters=30
activation=linear
[yolo]
mask = 0,1,2
#anchors = 20, 85, 41, 43, 27, 68, 46, 75, 47,108, 70,116
anchors= 45,75,47,141,105,63,38,226,79,123,101,202
classes=5
num=6
jitter=.3
ignore_thresh = .7
truth_thresh = 1
random=1
-
yolov3-tiny-obj_105000.weights ( custom tinyYoloV3 weights file)