ONNXっぽいやつができたのでコミット

config.json

@@ -0,0 +1,27 @@
+{
+  "_attn_implementation_autoset": true,
+  "_name_or_path": "cl-nagoya/ruri-small",
+  "activation": "gelu",
+  "architectures": [
+    "DistilBertModel"
+  ],
+  "attention_dropout": 0.1,
+  "dim": 768,
+  "dropout": 0.1,
+  "export_model_type": "transformer",
+  "hidden_dim": 3072,
+  "initializer_range": 0.02,
+  "max_position_embeddings": 512,
+  "model_type": "distilbert",
+  "n_heads": 12,
+  "n_layers": 6,
+  "output_hidden_states": true,
+  "pad_token_id": 0,
+  "qa_dropout": 0.1,
+  "seq_classif_dropout": 0.2,
+  "sinusoidal_pos_embds": true,
+  "tie_weights_": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.46.3",
+  "vocab_size": 32768
+}

onnx/model.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dc3a2fc698d0ec5aa7ab45f0b10517aab972dfede0243a31b9430659609334da
+size 272490076

onnx/model_bnb4.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a0600443d7191cf5c7aa7a22f78a80397e16ab1264b7d03fcf862e79b3b02705
+size 126514174

onnx/model_fp16.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:492b59f1a59751f77559d0f569dc9572b61ef97e50b737f73005524199b9b154
+size 136314565

onnx/model_int8.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:38a2813abf640e6d0aa9ce004edc380fdebe87c84258494f8e2f6412d79ac9e6
+size 68668245

onnx/model_q4.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b7bc29cca42120fc9a226c6137a3f030abd5698c313e685f285f23af8c8c551e
+size 129168118

onnx/model_q4f16.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b55a7be79de330304ba2e0b37fa832ca7e411d91f3beb722fdfc65ad81d5320a
+size 75273104

onnx/model_quantized.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:38a2813abf640e6d0aa9ce004edc380fdebe87c84258494f8e2f6412d79ac9e6
+size 68668245

onnx/model_uint8.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:15688908de5bc58d3d0e9c7e3ab2c8504d6b2e05ff96df8411b1fac147122311
+size 68668245

quantize_config.json

@@ -0,0 +1,18 @@
+{
+    "modes": [
+        "fp16",
+        "q8",
+        "int8",
+        "uint8",
+        "q4",
+        "q4f16",
+        "bnb4"
+    ],
+    "per_channel": true,
+    "reduce_range": true,
+    "block_size": null,
+    "is_symmetric": true,
+    "accuracy_level": null,
+    "quant_type": 1,
+    "op_block_list": null
+}