Merge pull request !2666 from jinyaohui/backend

Merge pull request !2682 from vlne-v1/I1LJMR-quant-the-top-level-add-op-prefix_name-check-error-r0.5

Merge pull request !2665 from Margaret_wangrui/r0.5

Merge pull request !2647 from yanghaoran/r0.5

Merge pull request !2649 from ghzl/fix-group-params-type-r0.5

Merge pull request !2650 from yanghaitao/yht_sampling_interval_r0.5

Merge pull request !2663 from laiyongqiang/gpu_opt_r05

Merge pull request !2657 from Chong/r5

Merge pull request !2645 from caojian05/ms_r0.5

Merge pull request !2637 from zyli2020/add_wide_deep_standalone_training_script

Merge pull request !2627 from zhoufeng/modify-long-description-format-r0.5

Merge pull request !2546 from liuwenhao/r0.5

Merge pull request !2515 from 李鸿章/no_step_zero

Merge pull request !2539 from geekun/yjk_r0.5

Merge pull request !2538 from xianwz/r0.5

Merge pull request !2617 from ougongchang/fix_collect_freq_r0.5

Signed-off-by: Nzhoufeng <zhoufeng54@huawei.com>

Merge pull request !2596 from zhoufeng/assign-order-before-jump-r0.5

Merge pull request !2613 from yanghaitao/yht_fastrcnn_0.5

Before, we only decide whether to collect data by current step,
it will not work well in dataset sink mode, so we check to see
if it's a dataset sink mode, and decide whether to collect data.

!2469 add pretrain for lstm & vgg16 and remove lstm/vgg16/googlenet from directory 'mindspore/model_zoo'
Merge pull request !2469 from caojian05/ms_master_dev

Merge pull request !2577 from Chong/r5

Merge pull request !2581 from TronZhang/add_test_lamb_r0.5

Merge pull request !2597 from DeshiChen/0624_akg_batmatmul_r0.5

input
Signed-off-by: Nzhoufeng <zhoufeng54@huawei.com>