File size: 830 Bytes
afed0f4
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
{
  "one_external_file": true,
  "opset": null,
  "optimization": {},
  "optimum_version": "1.16.2",
  "quantization": {
    "activations_dtype": "QUInt8",
    "activations_symmetric": false,
    "format": "QOperator",
    "is_static": false,
    "mode": "IntegerOps",
    "nodes_to_exclude": [],
    "nodes_to_quantize": [],
    "operators_to_quantize": [
      "Conv",
      "MatMul",
      "Attention",
      "LSTM",
      "Gather",
      "Transpose",
      "EmbedLayerNormalization"
    ],
    "per_channel": false,
    "qdq_add_pair_to_weight": false,
    "qdq_dedicated_pair": false,
    "qdq_op_type_per_channel_support_to_axis": {
      "MatMul": 1
    },
    "reduce_range": false,
    "weights_dtype": "QInt8",
    "weights_symmetric": true
  },
  "transformers_version": "4.37.2",
  "use_external_data_format": false
}