Skip to content

Commit e273bf7

Browse files
author
Eman Copty
committed
Changes for 1.11 release
1 parent 429f928 commit e273bf7

File tree

5 files changed

+69
-16
lines changed

5 files changed

+69
-16
lines changed

Makefile

+6-1
Original file line numberDiff line numberDiff line change
@@ -21,6 +21,10 @@ all: install examples
2121
opencv:
2222
./install-opencv.sh
2323

24+
.PHONY: uninstallopencv
25+
uninstallopencv:
26+
./uninstall-opencv.sh
27+
2428
.PHONY: prereqs
2529
prereqs:
2630
@sed -i 's/\r//' ncsdk.conf
@@ -35,9 +39,10 @@ prereqs:
3539
@chmod +x install.sh
3640
@chmod +x uninstall.sh
3741
@chmod +x install-opencv.sh
42+
@chmod +x uninstall-opencv.sh
3843

3944
.PHONY: install
40-
install: prereqs
45+
install: prereqs uninstallopencv
4146
@echo "\nmake install starting."
4247
./install.sh
4348

docs/release_notes.md

+24-11
Original file line numberDiff line numberDiff line change
@@ -1,17 +1,21 @@
11
============================================================
22
# Movidius Neural Compute SDK Release Notes
3-
# V1.10.01 2017-11-15
3+
# V1.11.00 2017-12-06
44
============================================================
55

66
As of V1.09.00, SDK has been refactored and contains many new features and structural changes. It is recommended you read the documentation to familiarize with the new features and contents. Please see v1.09.00 release notes, using github tag https://github.com/movidius/ncsdk/tree/v1.09.00.06
77

88
## SDK Notes:
99
### New features:
10+
1. Tensorflow 1.4 supported. Tensorflow 1.4 is automatically installed on Ubuntu.
11+
2. SSD supported in Caffe. ssd-caffe is automatically installed with this version. If a previous version of the SDK is installed, Berkeley Caffe is preserved and the symlink at /opt/movidius/caffe is moved to ssd caffe.
1012
#### Networks:
11-
1. VGG 16 for both Caffe and Tensorflow
12-
2. Resnet 18 for Caffe
13-
3. Resnet 50 for Caffe
14-
13+
1. SSD Mobilenet v1 for caffe
14+
2. TinyYolo v2 for Tensorflow via Darkflow transformation of original TinyYolo v2
15+
#### Layers:
16+
1. L2 Normalization
17+
2. Input
18+
1519
## API Notes:
1620
1. No change
1721

@@ -27,8 +31,9 @@ Support for the following networks has been tested.
2731
6. Alexnet
2832
7. TinyYolo v1
2933
8. VGG 16
30-
9. Resnet 18
31-
10. Resnet 50
34+
9. Resnet 50
35+
10. SSD Mobilenet v1
36+
3237

3338
### Tensorflow r1.3
3439
1. inception-v1
@@ -54,6 +59,7 @@ Support for the following networks has been tested.
5459
- MobileNet_v1_0.25_192
5560
- MobileNet_v1_0.25_160
5661
- MobileNet_v1_0.25_128
62+
8. TinyYolo v2 via Darkflow tranformation
5763

5864
## Firmware Features:
5965
1. Convolutions
@@ -80,13 +86,16 @@ Support for the following networks has been tested.
8086
17. Crop
8187
18. ELU
8288
19. Batch Normalization
89+
20. L2 Normalization
90+
21. Input Layer
8391

8492

8593
## Bug Fixes:
86-
Fixed: Installation does not succeed as root user
87-
Fixed: Support for larger convolutions, enabling VGG 16 support
88-
Fixed: Issues blocking Resnet 18 and Resnet 50
89-
94+
1. Fixed: Most recent version of Inception_v4 run fails with FusedBatchNorm not supported error
95+
2. Fixed: Installer deletes installation script, making review of installation process difficult
96+
3. Fixed: Inception Resnet V2 failing with Stage Details Not Supported: FusedBatchNorm error
97+
4. Fixed: Input layer can't be linked to multiple hidden layers
98+
9099
## Errata:
91100
1. Python 2.7 is fully supported for making user applications, but only the helloworld_py example runs as-is in both python 2.7 and 3.5 due to dependencies on modules.
92101
2. SDK tools for tensorflow on Rasbpian Stretch are not supported for this release, due to lack of an integrated tensorflow installer for Rasbpian in the SDK. TF examples are provided with pre-compiled graph files to allow them to run on Rasperry Pi, however the compile, profile, and check functions will not be available on Raspberry Pi, and 'make examples' will generate failures for the tensorflow examples on Raspberry Pi.
@@ -98,3 +107,7 @@ Support for the following networks has been tested.
98107
8. Convolution may fail to find a solution for very large inputs.
99108
9. Depth convolution is tested for 3x3 kernels.
100109
10. A TanH layer’s “top” & “bottom” blobs must have different names.  This is different from a ReLU layer, whose “top” & “bottom” should be named the same as its previous layer.
110+
11. On upgrade from previous versions of SDK, the installer will detect if openCV 3.3.0 was installed, for example from http://github.com/movidius/ncappzoo/apps/stream_ty_gn/install-opencv-from_source.sh. For this release, the installer will prompt to uninstall this specific version of openCV. This is required for ssd-caffe to run correctly. After 1.11 installation is complete, openCV 3.3.0 can be re-installed and the ssd-caffe will continue to function.
111+
12. Resnet 18 is showing unexpected behavior for this release, and is being investigated. To use Resnet 18, please use version 1.10.01 of SDK.
112+
13. Although mvNCCheck shows per-pixel error for some metrics for mobilenet_v1_224, classification results are not impacted.
113+
14. Initial validation has been done on SSD Mobilenet v1 and TinyYolo v2 but more thorough evaluation is underway.

install.sh

+2-3
Original file line numberDiff line numberDiff line change
@@ -11,10 +11,10 @@ then
1111
cd /tmp
1212
else
1313
cd /tmp
14-
wget --no-cache http://ncs-forum-uploads.s3.amazonaws.com/ncsdk/ncsdk_01_10_01/ncsdk_redirector.txt
14+
wget --no-cache http://ncs-forum-uploads.s3.amazonaws.com/ncsdk/ncsdk_01_11/ncsdk_redirector.txt
1515
fi
1616

17-
download_filename=NCSDK-1.10.01.tar.gz
17+
download_filename=NCSDK-1.11.tar.gz
1818

1919
# redirector is the url from redirector text file
2020
redirector=$(<ncsdk_redirector.txt)
@@ -58,7 +58,6 @@ cp /tmp/ncsdk.conf .
5858
#sudo cp ./uninstall-ncsdk.sh ../ncsdk/
5959

6060
# cleanup
61-
sudo rm ./install-ncsdk.sh
6261
cd ..
6362
sudo rm ncsdk.tar.gz
6463
#sudo rm -r NCSDK

ncsdk.conf

+1-1
Original file line numberDiff line numberDiff line change
@@ -2,7 +2,7 @@ MAKE_PROCS=1
22
SETUPDIR=/opt/movidius
33
VERBOSE=yes
44
SYSTEM_INSTALL=yes
5-
INTEL_CAFFE=no
5+
CAFFE_FLAVOR=ssd
66
CAFFE_USE_CUDA=no
77
INSTALL_TENSORFLOW=yes
88
INSTALL_TOOLKIT=yes

uninstall-opencv.sh

+36
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,36 @@
1+
#! /bin/bash
2+
3+
#test for python already installed for opencv
4+
python3 -c "import cv2" > /dev/null 2>&1
5+
if [ $? -eq 0 ] ;
6+
then
7+
echo ""
8+
echo "************************ Please confirm *******************************"
9+
echo " NCSDK 1.11 requires that previous installations of openCV"
10+
echo " be uninstalled before proceeding with NCSDK installation."
11+
echo " Note that if you installed opencv via pip3 or from source into the"
12+
echo " home directory, it will be uninstalled."
13+
read -p " Continue uninstalling OpenCV (y/n) ? " CONTINUE
14+
if [[ "$CONTINUE" == "y" || "$CONTINUE" == "Y" ]]; then
15+
echo "";
16+
echo "OpenCV already setup for python";
17+
echo "";
18+
echo "Uninstalling opencv pip installation";
19+
sudo pip3 uninstall opencv-contrib-python
20+
sudo pip3 uninstall opencv-python
21+
22+
echo "Looking for opencv source installation";
23+
if [ -d "$HOME/opencv-3.3.0" ]; then
24+
echo "opencv-3.3.0 directory exists"
25+
if [ -e "$HOME/opencv-3.3.0/build/Makefile" ]; then
26+
echo "opencv-3.3.0 Makefile exists, attempting to uninstall opencv-3.3.0"
27+
cd "$HOME/opencv-3.3.0/build"
28+
sudo make uninstall &> /dev/null
29+
echo "done."
30+
fi
31+
fi
32+
else
33+
echo " Not removing opencv, quitting."
34+
exit 1
35+
fi
36+
fi

0 commit comments

Comments
 (0)