When `collect_tensor_freq` is specified as `None`,
the `collect_tensor_freq` would be auto calculated.

The previous behavior is to collect at most 50 steps,
now changing to 20

Merge pull request !3511 from linqingke/new_ops

Merge pull request !3513 from ZPaC/master-add-ps-test-cases

Merge pull request !3425 from fangzehua/avgpool

Merge pull request !3444 from zhouyuanshen/master

Merge pull request !3497 from yuchaojie/transformer_st

Merge pull request !3349 from liubuyu/master

Merge pull request !3490 from yangyongjie/master

Merge pull request !3495 from chenweifeng/strided_slice

Merge pull request !3324 from fangzehua/reduce_any

Merge pull request !3457 from chenzhongming/master

Merge pull request !3461 from Xiaoda/12-add-annotations-for-embeddinglookup

Merge pull request !3460 from ms_yan/api_format

Merge pull request !3450 from fary86/fix_large_for_loop_execute_error

Merge pull request !3494 from guoqi/master

Merge pull request !3510 from chenfei_mindspore/avoid-insert-assign-from-condition-to-true-branch

Merge pull request !3485 from Kang/optimization

Merge pull request !3459 from fangzehua/fix-argmin

Merge pull request !3464 from kisnwang/fix-cpu-nonop-fp16

Merge pull request !3496 from YuJianfeng/internal_output

Merge pull request !3469 from panfengfeng/fix_getdataset_size_error_II

Merge pull request !3284 from riemann_penn/eager_mode_enable_sparse

Merge pull request !3465 from changzherui/mod_monitor

Merge pull request !3452 from nhussain/refactor_name_function

Merge pull request !2983 from h.farahat/tensor_class

Merge pull request !3395 from Alexey_Shevlyakov/remove_windows_define

Merge pull request !3474 from liuxiao93/fix-conv1d

Merge pull request !3491 from ZPaC/delete-param-name-hard-code

Merge pull request !3487 from Simson/push-to-opensource

Merge pull request !3447 from EricZ/crop_rbswap

Merge pull request !3502 from chengang/graceful_shutdown_ps

Merge pull request !3486 from chenhaozhe/fix_bert_performance_test_case