I started to use tensorflow official models implementation and need ResNet50 parameters for resnet implementation. It requires a lot of parameters, but I don't know the proper way to do it, cause I'm also going to restore from checkpoints.
Here is what you need.
from official.resnet.resnet_model import Model
resnet = Model(resnet_size=50, num_classes=1001, kernel_size=7, num_filters=64, conv_stride=2,
first_pool_size=3, first_pool_stride=2, block_sizes=[3, 4, 6, 3], block_strides=[1, 2, 2, 2],
bottleneck=True, resnet_version=1, data_format='channels_last')
model = resnet(input_tensor, training=is_training_placeholder)Remember to add your tensorflow-models path in PYTHONPATH env variable.