Fix error msg of pythontokenizer

From: @xiefangqi
Reviewed-by: 
Signed-off-by:

From: @luoyang42
Reviewed-by: 
Signed-off-by:

From: @ziruiwu
Reviewed-by: 
Signed-off-by:

From: @alexyuyue
Reviewed-by: @robingrosman,@nsyca
Signed-off-by: @robingrosman

Signed-off-by: alex-yuyue <yue.yu1@huawei.com>

Adjust memory usage to not increase as parallelism increases.  it will
stay at same level it would be with 4 parallelism

From: @ezphlow
Reviewed-by: 
Signed-off-by:

what if we just do nothing

trying wrapper approach

Fix yolo

From: @alexyuyue
Reviewed-by: @robingrosman,@nsyca
Signed-off-by: @robingrosman

Signed-off-by: alex-yuyue <yue.yu1@huawei.com>

From: @hfarahat
Reviewed-by: 
Signed-off-by:

From: @ezphlow
Reviewed-by: 
Signed-off-by:

Fixed compile error

Fixed more testcases

And support multiple tensor as input in call method

Signed-off-by: alex-yuyue <yue.yu1@huawei.com>

From: @lixiachen
Reviewed-by: 
Signed-off-by:

From: @luoyang42
Reviewed-by: 
Signed-off-by:

fix sampler

From: @ezphlow
Reviewed-by: @robingrosman,@nsyca,@robingrosman
Signed-off-by: @robingrosman

Added validator checks

Added builtin sampler check

Added proper input check for samplers

From: @lixiachen
Reviewed-by: 
Signed-off-by:

update copyright

add examples to CacheClient docstring

From: @alexyuyue
Reviewed-by: 
Signed-off-by:

From: @xiefangqi
Reviewed-by: @liucunwei,@heleiwang
Signed-off-by: @liucunwei

Signed-off-by: alex-yuyue <yue.yu1@huawei.com>

From: @hfarahat
Reviewed-by: 
Signed-off-by:

From: @heleiwang
Reviewed-by: @liucunwei
Signed-off-by: @liucunwei