Upload ONNX weights (#2)

- Upload ONNX weights (7499423f540588db41167cfe3ba003bc377ca9a5)


Co-authored-by: Joshua <Xenova@users.noreply.huggingface.co>

onnx/model.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6a69af93de0a4c349d263c398534d31b25c375fc67859049d3c08615c761257c
+size 1336854282

onnx/model_quantized.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:21ac80429bb5d96f6879ff1283692a502dc65aa8993d89ea3ae95a1745149479
+size 336983163

onnx/quantize_config.json

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