Skip to content

Caffe Models for DesignWare EV Processors, R-2021.03
Compare
Choose a tag to compare
@dmitrygolovkin dmitrygolovkin released this 12 Apr 11:51
· 6 commits to master since this release
R-2021_03
49f324d

This is release R-2021.03 of the Synopsys-caffe-models, a set of Caffe Deep Learning Models adapted for use with DesignWare EV Processors.

These models must be used together with Synopsys-Caffe R-2021.03 and the MetaWare EV Development Toolkit R-2021.03 from Synopsys.

Supported Models

Alexnet, DAN, Denoiser, Densenet, Deeplab, Facedetect v1/v2, FaceNet, Faster_rcnn_resnet101, FCN, Googlenet. ICNet,
Inception_Resnet v1/v2, Inception v1/v2/v3/v4, LeNet, Mobilenet, Mobilenet_ssd, mtcnn v1, OpenPose, PSPNet,
Resnet 50/101/152, Resnet50_ssd, ResNext 50/101/152, RetinaNet, SegNet, ShuffleNet v1/v2, SRGAN, SqueezeNet, SRCNN, SSD,
PSPNet, UNet, VDCR, VGG16, Yolo v1/v1_tiny/v2_coco/v2_voc/v2_tiny/v3/v3_tiny

Images

  • imagenet_mean - mean images for different image sizes
  • imagenet_test_images - simple set of test images
  • images - different image data sub-sets

Changes vs 2020.12

New models

  • Yolo-V2 Tiny

New zipped models and downloader

In the archived_models folder, we placed zipped models and a Python downloader script. Please see details in:
archived_models/README.md

Helper tools

git_sparse_download.sh(bat) - old git based downloader
archived_models/download_unpack_cnn_models.py - new Python based downloader

Assets 2
Loading