nncase v2.8.3

New features

• qemu system mode: #1201
• support layernorm channel first(C#): #1204

Bug fixes and performance improve

• Feature/add xsgetn for char array buffer: #1198
• fix reduce sum opset conditio: #1203
• Remove unavailable nuget source(nuget.cnblogs.com): #1207

nncase v2.8.2

New features

• Feature/update docs: #1174
• Feature/egraph extract constrains: #1175
• Feature/update docs: #1176

Bug fixes and performance improve

• fix fold binary: #1182
• Fix macos-latest doesn't support python 3.7 issue: #1194

nncase v2.8.1

New features

• Support multiple kmodels for accuracy_test: #1162
• Supprot npy for pytest: #1164

Bug fixes and performance improve

• Fix/dump_manager: #1163
• Fix Melgan: #1168
• Fix the missing new line issue of dynamic shape: #1170
• Fix in_ci env judge with bool type: #1171

nncase v2.8.0

New features

• Feature/gem5: #1157

Bug fixes and performance improve

• GNNE-1714:Fix/search time egraph: #1155
• Fix Not Equal : #1156
• GNNE-1714:Fix/some opt: #1158
• Fix Bucket: #1159

nncase v2.7.0

New features

• Add accuracy test: #1140
• Refactor Quantize: #1144
• Feature/k800 qemu: #1145
• GNNE-1980 Xpu Sram Codegen: #1148
• GNNE-1714:adjust for processor counter: #1150

Bug fixes and performance improve

• Fix/retinaface similarity bugs: #1141
• Fix memory leak: Avoid disposing python stream in finalizer thread: #1142
• Fix ordering of PytestCalibrationDatasetProvider: #1143
• Fix Reshape: #1147
• Fix Studio: #1151, #1152
• Add clear device mem pool util: #1153

nncase v2.6.0

New features

• Feature/cpu: #1019
• add onnx-op-support-md file: #1081
• Feature/egraph fakepass: #1111
• Feature/audio: #1118
• Feature/remove apply: #1124
• GNNE-1974: Feature/non uniform: #1128
• add CheckedShape for DistributedType: #1133
• add rvv_memcpy && opt profile: #1135
• Nncase studio: #1125, #1136, #1137

Bug fixes and performance improve

• Fix runtime: #1115
• Refactor infer report for k230_benchmark_test: #1119
• Fix/melgan: #1126
• fix LN evaluator: #1131
• Fix metadata: #1132
• GNNE-1714:Fix/regression bugs: #1116, #1117, #1121, #1123, #1127, #1130

nncase v2.5.1

New features

Bug fixes and performance improve

• Fix GetItem: #1114

nncase v2.5.0

New features

• add mse: #1110

Bug fixes and performance improve

• fix op's datatype: #1068
• fix timeout issue of read_until in pyserial: in #1101
• Fix/quant by json: #1106
• Fix/cos nan: #1090
• fix new linked section: #1108
• fix onnximport's dt: #1103
• Upgrade vulkan version from 1.2.182.0 to 1.3.268.0: in #1112

nncase v2.4.0

New features

• GNNE-1904 Support coordinate_mode and nearest_mode selection for nearest resize_image: #1065
• Feature/json: #1074
• Feature/add colab link: #1075
• TestRunner support dynamic shape: #1076, #1089, #1095, #1096
• github CI cancle: #1078, #1093
• Support large kmodel: #1079
• ncnn importer: #332

Bug fixes and performance improve

• fix kernel test: #1069, #1072, #1073, #1077, #1082, #1084, #1085
• GNNE-1891 ShapeBucket Optimize: #1088
• Fix Enable Nop: #1097
• fix euclidean calc: #1094
• Fix BucketPad: #1098
• fix nncase version suffix: #1099, #1100

nncase v2.3.0

New features

• Broadcast marker: #1033
• Feature/add squeeze binary shape: #1059
• fix python interface: #1040 #1057 #1063 #1066 #1067 #1071
• remove date in version: #1038
• Support one-dim input in matmul op: #1051
• Add inference report for both simulator and evb: #1061

Bug fixes and performance improve

• runtime kernel test fix: #1024 #1028 #1029 #1030 #1031 #1032 #1034 #1035 #1036 #1043 #1044 #1046
• Improve test runner: #1037 #1041 #1047
• Fix Dynamic TTS: #1039
• fix SpaceToBatchTest: #1053
• Fix 3D BatchToSpace: #1058
• Fix Tuple Compare: #1050