* update model save load

* first add

* update model save and load

* update train.py

* update LaneNet model saving and loading

* adapt slim to paddle-1.8

* update distillation save and load

* update nas model save and load

* update model load op

* update utils.py

* update load_model_utils.py

* update model saving and loading

* update copyright

* update palette.py

* update *_batch_size_like ops

* remove the preprocess and postprocess of model exporting

* modify label tool

* add lovasz loss

* lovasz

* model builder

* test

* lovasz

* update lovasz

* update lovasz

* add sh yaml

* add sh yaml

* add sh yaml

* no per_image and device_guard

* no per_image and device_guard

* Update lovasz_loss.md

* Update lovasz_loss.md

* Update lovasz_loss.md

* Update lovasz_loss.md

* Update lovasz_loss.md

* elementwise_mul replace matmul

* save inference model

* fix details

* add export model in quantization

* remove useless code

* remove useless code

* change fluid.layers.data to fluid.data

* release change log

* update readme

* update readme

* upload fast-scnn image

* update readme

* update readme

* replace fluid.layers.data

* update api fluid.data and fluid.io.DataLoader
Co-authored-by: NLielinJiang <jianglielin@baidu.com>
Co-authored-by: NLielinJiang <50691816+LielinJiang@users.noreply.github.com>

* add quantization demo train.py

* add quantizatiom demo

* fix readme

* fix details

* fix details

* add last_conv name_scope

* fix config file in eval.sj

* remove train.sh eval.sh

* add fast scnn

* fix readme

* update model name, add inference time

* fix typos

* remove importlib

* remove importlib

* update solver.py and model_builder.py

* update solver.py and model_builder.py

* update solver.py

* add hrnet tutorial

* add preprocess and argmax to export process

* add comments

* add hrnet

* update model_builder.py

* update loss_select.md

* init deepglobe_road_extraction.yaml

* update download_mini_deepglobe_road_extraction.py

* update model_builder.py

* fp16_err

* add fp16 and multiple gpus train

* add fp16 and multiple gpus train

* add fp16 and multiple gpus train

* add fp16 and multiple gpus train

* add fp16 and multiple gpus train

* add fp16 and multiple gpus train

* add fp16 and multiple gpus train

* update solver_group.md