Upload folder using huggingface_hub

config.json

@@ -23,7 +23,7 @@
   "pooler_size_per_head": 128,
   "pooler_type": "first_token_transform",
   "position_embedding_type": "absolute",
-  "transformers_version": "4.29.2",
+  "transformers_version": "4.33.0.dev0",
   "type_vocab_size": 2,
   "use_cache": true,
   "vocab_size": 105879

generation_config.json

@@ -1,5 +1,5 @@
 {
   "_from_model_config": true,
   "pad_token_id": 0,
-  "transformers_version": "4.29.2"
+  "transformers_version": "4.33.0.dev0"
 }

onnx/model.onnx

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a8bb56c1c0b8468cddfab60c87b57b43eb43b6ccc3bb41e251dfef7c6c7c7b03
-size 995386085
+oid sha256:c8b0ea2376868d5393163b87fedc6da0a83b2a2378922cfead2d80049de8501d
+size 670125922

onnx/model_quantized.onnx

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fc95dcdf605709e469c2858eb2243a4827ca39c90d9b45f8ebec22f6667d952b
-size 250869303
+oid sha256:a6dcae35e1de2d31e7afb13a2a9b4b43d2c9a63f9a438ea5f75d66d359036da7
+size 169024183

quantize_config.json

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