Welcome to Evaluate model!

Here are some instrctions for training the network.

Requirements: software

1. Tensorflow 1.3 or above.
2. opencv3

Preparation for training

update soon....

Pre-processing

update soon....

Files after pre-processing

After pre-processing, you will have following files.

mean_pixelrgb.npy, std_pixelrgb.npy : the mean and the standard deviation of each pixel on images in the entire dataset.

training set:

sf_filename.npy : A list of colorized depth image path. (shuffle)

sf_vector.npy : A list of grasp parameters. (shuffle)

sf_label.npy : A list of label. (shuffle)

validation set

sf_vfilename.npy : A list of colorized depth image path. (shuffle)

sf_vvector.npy : A list of grasp parameters. (shuffle)

sf_vlabel.npy : A list of label. (shuffle)

Above files are the input of the network. Please see example floder to see the right format of above files.

Train

It`s easy to train the network if you have above files. Just run following command:

python train.py

It will read above files and train the network for 40 epochs. The hyperparameters are easily to change in the train.py

Using Tensorboard to see the process

Update soon.........