Upload folder using huggingface_hub

onnx/model_quantized.onnx

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1a3bda057b541217345984a0574ca39c22eba4ef02daf1656cc1033ab1fc4f20
-size 24417500
+oid sha256:089213f71783e3a0a41eee1db4af6a1d7bfc5272c35076879e0fd9fea323248a
+size 24625229

ort_config.json

@@ -11,21 +11,40 @@
     "nodes_to_exclude": [],
     "nodes_to_quantize": [],
     "operators_to_quantize": [
+      "Concat",
+      "Constant",
+      "Resize",
       "Conv",
+      "Div",
+      "Unsqueeze",
+      "Shape",
+      "Gemm",
       "MatMul",
-      "Attention",
-      "LSTM",
-      "Gather",
       "Transpose",
-      "EmbedLayerNormalization"
+      "Pow",
+      "Softmax",
+      "Slice",
+      "Where",
+      "Expand",
+      "ReduceMean",
+      "Mul",
+      "Add",
+      "Sqrt",
+      "Gather",
+      "Sub",
+      "ConstantOfShape",
+      "Reshape",
+      "Equal",
+      "Identity",
+      "Erf"
     ],
-    "per_channel": false,
+    "per_channel": true,
     "qdq_add_pair_to_weight": false,
     "qdq_dedicated_pair": false,
     "qdq_op_type_per_channel_support_to_axis": {
       "MatMul": 1
     },
-    "reduce_range": false,
+    "reduce_range": true,
     "weights_dtype": "QUInt8",
     "weights_symmetric": true
   },