Skip to content

NCSDK v2.08.01
Compare
Choose a tag to compare
@nealvis nealvis released this 08 Oct 22:28
· 17 commits to master since this release
v2.08.01.02
0aaa3bd

============================================================

Movidius Neural Compute SDK Release Notes

V2.08.01 2018-10-04

============================================================

This release has been refactored and is now licensed under Apache 2.0 license.

As of V2.05.00, 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.

SDK Notes:

New features:

  1. Tensorflow 1.09 supported. Tensorflow 1.09 is automatically installed on Ubuntu.
  2. The compiler has been refactored for best performance however some networks may still see slight performance degradation.

Networks:

  1. No significant changes.

Layers:

  1. No change.

API Notes:

  1. Apps written with NCAPI v1 are not compatible with this release and need to be migrated to NCAPI v2, refer to Migrating Applications from NCAPI v1 to NCAPI v2 for information about migrating apps to the new API.

Network Notes:

Support for the following networks has been tested.

Caffe

  1. GoogleNet V1
  2. SqueezeNet V1.1
  3. LeNet
  4. CaffeNet
  5. VGG (Sousmith VGG_A)
  6. Alexnet
  7. TinyYolo v1
  8. VGG 16 (Configuration D)
  9. Resnet 18
  10. Resnet 50
  11. Mobilenet v1
  12. SSD Mobilenet v1

Tensorflow r1.09

  1. inception-v1
  2. inception-v2
  3. inception-v3
  4. inception-v4
  5. Inception ResNet v2
  6. VGG 16 (Configuration D)
  7. Mobilenet_V1_1.0 variants:
    • MobileNet_v1_1.0_224
    • MobileNet_v1_1.0_192
    • MobileNet_v1_1.0_160
    • MobileNet_v1_1.0_128
    • MobileNet_v1_0.75_224
    • MobileNet_v1_0.75_192
    • MobileNet_v1_0.75_160
    • MobileNet_v1_0.75_128
    • MobileNet_v1_0.5_224
    • MobileNet_v1_0.5_192
    • MobileNet_v1_0.5_160
    • MobileNet_v1_0.5_128
    • MobileNet_v1_0.25_224
    • MobileNet_v1_0.25_192
    • MobileNet_v1_0.25_160
    • MobileNet_v1_0.25_128
  8. TinyYolo v2 via Darkflow tranformation
  9. facenet based on inception-resnet-v1

Firmware Features:

  1. Convolutions
  • The following convolution cases have been extensively tested (for stride s): 1x1s1, 3x3s1, 5x5s1, 7x7s1, 7x7s2, 7x7s4, 1x3, 3x1, 1x7, 7x1
  • Group convolution
  • Depth Convolution
  • Dilated convolution
  1. Max Pooling Radix NxM with Stride S
  2. Average Pooling: Radix NxM with Stride S, Global average pooling
  3. Local Response Normalization
  4. Relu, Relu-X, Prelu (see erratum #9)
  5. Softmax
  6. Sigmoid
  7. Tanh (see erratum #9)
  8. Deconvolution
  9. Slice
  10. Scale
  11. ElmWise unit : supported operations - sum, prod, max
  12. Fully Connected Layers (limited support -- see erratum #8)
  13. Reshape
  14. Flatten
  15. Power
  16. Crop
  17. ELU
  18. Batch Normalization
  19. L2 Normalization
  20. Input Layer

Bug Fixes:

  1. Fixed: Various profiler issues.

Errata:

  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.
  2. Depth-wise convolution may not be supported if channel multiplier > 1.
  3. If working behind proxy, proper proxy settings must be applied for the installer to succeed.
  4. Although improved, the installer is known to take a long time on Raspberry Pi. Date/time must be correct for SDK installation to succeed on Raspberry Pi.
  5. The Docker Non-privileged mode of operation as described in the documentation has an issue with multiple NCS devices.
  6. Inception V1 obtained values are invalid for mvNCCheck.
  7. Convolution may fail to find a solution for very large inputs.
  8. Depth-wise convolution is tested for 3x3 kernels.
  9. 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.
  10. 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 installation is complete, openCV 3.3.0 can be re-installed and the ssd-caffe will continue to function.
  11. The MTCNN network in the app zoo is showing unexpected behavior for this release, and is being investigated. To use MTCNN, please use version 1.12.01 of SDK.
  12. Although mvNCCheck shows per-pixel error for some metrics for mobilenet_v1_224, classification results are not impacted.
  13. Only Ubuntu 16.04 LTS is supported as a host OS for this release. Ubuntu 18.04 is being evaluated.
  14. Group Deconvolution with "group" parameter != 1 is not supported on the new parser.
  15. For this release, use of Myriad devices connected to some specific hubs can fail. If you encounter errors, please try direct connect to PC port, or try a different hub.
  16. Loading multiple graphs into one Myriad device may show stability issues with this release.
Assets 2
Loading