programmersilvanus/ner-quantization

1{2  "one_external_file": true,3  "opset": null,4  "optimization": {},5  "quantization": {6    "activations_dtype": "QUInt8",7    "activations_symmetric": false,8    "format": "QOperator",9    "is_static": false,10    "mode": "IntegerOps",11    "nodes_to_exclude": [],12    "nodes_to_quantize": [],13    "operators_to_quantize": [14      "Conv",15      "MatMul",16      "Attention",17      "LSTM",18      "Gather",19      "Transpose",20      "EmbedLayerNormalization"21    ],22    "per_channel": true,23    "qdq_add_pair_to_weight": false,24    "qdq_dedicated_pair": false,25    "qdq_op_type_per_channel_support_to_axis": {26      "MatMul": 127    },28    "reduce_range": false,29    "weights_dtype": "QInt8",30    "weights_symmetric": true31  },32  "use_external_data_format": false33}34