Upload ONNX weights

onnx/model.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4e0cbababab52b0f2b49fcc66fdc13505eb73f87d0d8bf8691954152ab54b74c
+size 1298376458

onnx/model_quantized.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5e91bb6e537ec96aa9cd621e2d5295cfdf91f935f179bdeb3a343c5bb159b963
+size 327363708

onnx/quantize_config.json

@@ -0,0 +1,30 @@
+{
+    "per_channel": true,
+    "reduce_range": true,
+    "per_model_config": {
+        "model": {
+            "op_types": [
+                "Reshape",
+                "Sub",
+                "Cast",
+                "Gather",
+                "Unsqueeze",
+                "Sqrt",
+                "Pow",
+                "MatMul",
+                "Erf",
+                "Div",
+                "Constant",
+                "Add",
+                "Shape",
+                "Slice",
+                "Concat",
+                "ReduceMean",
+                "Transpose",
+                "Softmax",
+                "Mul"
+            ],
+            "weight_type": "QInt8"
+        }
+    }
+}