7 - Add optimization pass: ReplaceMulAddWithDepthwiseConvPass, SubstitutePackToReshape, RemoveRedundantTranspose, ShuffleWeightTo16x1Float32Pass
8 - Add quantization for InstanceNorm.
9 - Fix bug of `one-import-bcq` command for `--v1`, `--v2` arguments.
10 - Fix FuseBCQPass to work with inter-subgraphs in the model file and minor BCQ related optimizations.
14 ### Runtime backend operation supports more operations and types
18 - DepthToSpace: float, uint8, int8
21 - ArgMin: float, uint8, int8
24 - ArgMin: float, uint8, int8
26 ### nnpackage defines configuration file
28 - Allow users to set configuration variable via conf file. For more information, See [nnpackage spec](../../../nnpackage/spec)