TensorFlow

TFLearn

Config

tflearn.init_graph(seed=8888, num_cores=4, gpu_memory_fraction=0.3)

Train

try:
    model = tflearn.DNN(network, tensorboard_verbose=0,
    checkpoint_path='/tmp/tflearn/checkpoint_path',
    best_checkpoint_path='/tmp/tflearn/best_checkpoint_path')
    model.fit({'input': X}, {'target': Y}, n_epoch=20,
               validation_set=({'input': testX}, {'target': testY}),
               snapshot_step=100, show_metric=True, run_id='yzbx_convnet_mnist')
except KeyboardInterrupt:
    print('save model '+'.'*60);
    model.save('/tmp/tflearn/convnet_mnist_ctrl_c_model')

TF-Slim

https://github.com/tensorflow/models/tree/master/slim

http://warmspringwinds.github.io/tensorflow/tf-slim/2016/10/30/image-classification-and-segmentation-using-tensorflow-and-tf-slim/

example

https://github.com/balancap/SSD-Tensorflow

"""datasets: voc/imagenet"""
from datasets import dataset_factory
"""preprocessing: vgg/res-net"""
from preprocessing import preprocessing_factory
"""deployment: gpu/cpu config"""
from deployment import model_deploy
"""nets: ssd/yolo"""
from nets import nets_factory

Keras

fast-rcnn: https://github.com/yhenon/keras-frcnn