Follow the neon installation procedure before proceeding.
If neon is installed into a
virtualenv, make sure that it is activated before running the commands below.
Also, the commands below use the GPU backend by default so add
-b cpu if you are running on a system without
a compatible GPU.
To measure the performance of the trained file (see above) using the Pascal VOC test data set, activate the neon virtual env (if applicable) and run the following command from the neon repo root directory:
python examples/fast-rcnn/run_validation.py --model_file frcn_vgg.p
The trained model weights can be downloaded from AWS using the following link: [trained Fast-RCNN model weights][S3_WEIGHTS_FILE]. [S3_WEIGHTS_FILE]: https://s3-us-west-1.amazonaws.com/nervana-modelzoo/Fast_RCNN/frcn_vgg.p
The weights provded here have been verified to work with neon version 1.4.0 The weights file may not work with other versions of neon.
This set of weights was trained using the train.py script for 20 epochs and reaches a mAP of 0.56 on the vaidation data set.