Upload folder using huggingface_hub

config.json

@@ -18,7 +18,7 @@
   "num_hidden_layers": 12,
   "pad_token_id": 0,
   "position_embedding_type": "absolute",
-  "transformers_version": "4.29.2",
+  "transformers_version": "4.33.0.dev0",
   "type_vocab_size": 2,
   "use_cache": true,
   "vocab_size": 30522

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 sha256:22d991d9df8d3b9e852da76c2a9db9a1400de8a3d7aa04147107885698c4a37e
-size 532091245
+oid sha256:92b66126e73e6667ef701c253d0a6be8e6709bb6a03436de226b4f64ff3cbf35
+size 438327788

onnx/model_quantized.onnx

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:961b4247a9ead0f2b2aa494ffe0d0a7c11f88ccdd124e3cca3cea26d9fbd0397
-size 134442738
+oid sha256:f2c1ce78f723dfa260bf743c56596ddd5097e45081cdffa24820aab900855ed0
+size 110848579

quantize_config.json

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