add int8 models

model.int8.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9ada9127ca5b82320385ac12340eb8b05dee64fd45cf8cf593ec693826ec2fd7
+size 223385835

quantize-model.py

@@ -0,0 +1,23 @@
+#!/usr/bin/env python3
+
+import onnx
+from onnxruntime.quantization import QuantType, quantize_dynamic
+
+
+def main():
+    onnx_model = onnx.load("model.onnx")
+    nodes = [n.name for n in onnx_model.graph.node]
+    nodes_to_exclude = [m for m in nodes if "output" in m]
+    print(nodes_to_exclude)
+    quantize_dynamic(
+        model_input="model.onnx",
+        model_output="model.int8.onnx",
+        op_types_to_quantize=["MatMul"],
+        per_channel=True,
+        weight_type=QuantType.QUInt8,
+        nodes_to_exclude=nodes_to_exclude,
+    )
+
+
+if __name__ == "__main__":
+    main()