add quant files
Browse files- config.json +1 -1
- onnx/build_delay_pattern_mask_quantized.onnx +3 -0
- onnx/decoder_model_merged_quantized.onnx +3 -0
- onnx/decoder_model_quantized.onnx +3 -0
- onnx/decoder_with_past_model_quantized.onnx +3 -0
- onnx/encodec_decode_quantized.onnx +3 -0
- onnx/text_encoder_quantized.onnx +3 -0
- ort_config.json +35 -0
config.json
CHANGED
@@ -1,5 +1,5 @@
|
|
1 |
{
|
2 |
-
"_name_or_path": "
|
3 |
"architectures": [
|
4 |
"MusicgenForConditionalGeneration"
|
5 |
],
|
|
|
1 |
{
|
2 |
+
"_name_or_path": "onnx",
|
3 |
"architectures": [
|
4 |
"MusicgenForConditionalGeneration"
|
5 |
],
|
onnx/build_delay_pattern_mask_quantized.onnx
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:18baab9eb99f1fc11fefce7b53315cf4e2e3eaa3e2368685f218546d3351c64a
|
3 |
+
size 52083
|
onnx/decoder_model_merged_quantized.onnx
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f4851bc74867aee7b861fddeb3e1667c3c101ed74428d02acd4af0a7565ef121
|
3 |
+
size 174070818
|
onnx/decoder_model_quantized.onnx
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ed83623cb82df2bdd8d38ce42385baa0a579050a2216ae70e43fc71f67e1e48f
|
3 |
+
size 173881253
|
onnx/decoder_with_past_model_quantized.onnx
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:60067fdfe9e288dae23e2ddaa2625d501542e7f174531a4977b0f12f0618c4d5
|
3 |
+
size 155484971
|
onnx/encodec_decode_quantized.onnx
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e68b0bacd527f46f408a28d6baa5a049eb46b1fb60746d2a52bd3be2ddf907b8
|
3 |
+
size 59766870
|
onnx/text_encoder_quantized.onnx
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e42330a9cff0be8014b9aecd120b26a8c04692bc1fee95a829983ab2cedca5ec
|
3 |
+
size 110069861
|
ort_config.json
ADDED
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"one_external_file": true,
|
3 |
+
"opset": null,
|
4 |
+
"optimization": {},
|
5 |
+
"optimum_version": "1.19.0",
|
6 |
+
"quantization": {
|
7 |
+
"activations_dtype": "QUInt8",
|
8 |
+
"activations_symmetric": false,
|
9 |
+
"format": "QOperator",
|
10 |
+
"is_static": false,
|
11 |
+
"mode": "IntegerOps",
|
12 |
+
"nodes_to_exclude": [],
|
13 |
+
"nodes_to_quantize": [],
|
14 |
+
"operators_to_quantize": [
|
15 |
+
"Conv",
|
16 |
+
"MatMul",
|
17 |
+
"Attention",
|
18 |
+
"LSTM",
|
19 |
+
"Gather",
|
20 |
+
"Transpose",
|
21 |
+
"EmbedLayerNormalization"
|
22 |
+
],
|
23 |
+
"per_channel": false,
|
24 |
+
"qdq_add_pair_to_weight": false,
|
25 |
+
"qdq_dedicated_pair": false,
|
26 |
+
"qdq_op_type_per_channel_support_to_axis": {
|
27 |
+
"MatMul": 1
|
28 |
+
},
|
29 |
+
"reduce_range": false,
|
30 |
+
"weights_dtype": "QInt8",
|
31 |
+
"weights_symmetric": true
|
32 |
+
},
|
33 |
+
"transformers_version": "4.39.3",
|
34 |
+
"use_external_data_format": false
|
35 |
+
}
|