ZhennanQin commented on a change in pull request #13697: [MKLDNN] Enable signed
int8 support for convolution.
URL: https://github.com/apache/incubator-mxnet/pull/13697#discussion_r244627308
##########
File path: include/mxnet/c_api.h
##########
@@ -1556,13 +1556,12 @@ MXNET_DLL int MXSymbolInferType(SymbolHandle sym,
* \param num_offline number of parameters that are quantized offline
* \param offline_params array of c strings representing the names of params
quantized offline
* \param quantized_dtype the quantized destination type for input data.
- * \param calib_quantize whether calibrate quantize op with offline
calibration data.
*/
MXNET_DLL int MXQuantizeSymbol(SymbolHandle sym_handle, SymbolHandle
*ret_sym_handle,
const mx_uint num_excluded_symbols,
const char **excluded_symbols,
const mx_uint num_offline, const char
**offline_params,
- const char *quantized_dtype, const bool
calib_quantize);
+ const char *quantized_dtype);
Review comment:
Yes, it's only for c_api. `calib_quantize` is newly added from mkldnn uint8
quantization. And we think its not necessary for now.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
