Hugging Face's logo

itdainb
/
vietnamese-embedding-onnx

Sentence Similarity
sentence-transformers
ONNX
Vietnamese
roberta
semantic-search
cosine-similarity
Eval Results
Inference Endpoints
Model card Files Community
itdainb commited on
Commit
ac08411
·
verified ·
1 Parent(s): a9588a7

Added optimized and quantized version

Browse files
Files changed (24) hide show
  1. dynamic_q8/added_tokens.json +3 -0
  2. dynamic_q8/bpe.codes +0 -0
  3. dynamic_q8/config.json +29 -0
  4. dynamic_q8/model_quantized.onnx +3 -0
  5. dynamic_q8/ort_config.json +35 -0
  6. dynamic_q8/special_tokens_map.json +51 -0
  7. dynamic_q8/tokenizer_config.json +54 -0
  8. dynamic_q8/vocab.txt +0 -0
  9. optimized/added_tokens.json +3 -0
  10. optimized/bpe.codes +0 -0
  11. optimized/config.json +29 -0
  12. optimized/model_optimized.onnx +3 -0
  13. optimized/ort_config.json +39 -0
  14. optimized/special_tokens_map.json +51 -0
  15. optimized/tokenizer_config.json +54 -0
  16. optimized/vocab.txt +0 -0
  17. static_q8/added_tokens.json +3 -0
  18. static_q8/bpe.codes +0 -0
  19. static_q8/config.json +29 -0
  20. static_q8/model_quantized.onnx +3 -0
  21. static_q8/ort_config.json +47 -0
  22. static_q8/special_tokens_map.json +51 -0
  23. static_q8/tokenizer_config.json +54 -0
  24. static_q8/vocab.txt +0 -0
dynamic_q8/added_tokens.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "<mask>": 64000
3
+ }
dynamic_q8/bpe.codes ADDED
The diff for this file is too large to render. See raw diff
 
dynamic_q8/config.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "onnx_v2",
3
+ "architectures": [
4
+ "RobertaModel"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": 0,
8
+ "classifier_dropout": null,
9
+ "eos_token_id": 2,
10
+ "gradient_checkpointing": false,
11
+ "hidden_act": "gelu",
12
+ "hidden_dropout_prob": 0.1,
13
+ "hidden_size": 768,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 3072,
16
+ "layer_norm_eps": 1e-05,
17
+ "max_position_embeddings": 258,
18
+ "model_type": "roberta",
19
+ "num_attention_heads": 12,
20
+ "num_hidden_layers": 12,
21
+ "pad_token_id": 1,
22
+ "position_embedding_type": "absolute",
23
+ "tokenizer_class": "PhobertTokenizer",
24
+ "torch_dtype": "float32",
25
+ "transformers_version": "4.40.1",
26
+ "type_vocab_size": 1,
27
+ "use_cache": true,
28
+ "vocab_size": 64001
29
+ }
dynamic_q8/model_quantized.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ee0ced372506cd045769c94c764efe4bba5a29494c3ccbc1a50699627b82bda7
3
+ size 135223909
dynamic_q8/ort_config.json ADDED
@@ -0,0 +1,35 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "one_external_file": true,
3
+ "opset": null,
4
+ "optimization": {},
5
+ "optimum_version": "1.19.1",
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": "4.40.1",
34
+ "use_external_data_format": false
35
+ }
dynamic_q8/special_tokens_map.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "cls_token": {
10
+ "content": "<s>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "eos_token": {
17
+ "content": "</s>",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "mask_token": {
24
+ "content": "<mask>",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "pad_token": {
31
+ "content": "<pad>",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ },
37
+ "sep_token": {
38
+ "content": "</s>",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false
43
+ },
44
+ "unk_token": {
45
+ "content": "<unk>",
46
+ "lstrip": false,
47
+ "normalized": false,
48
+ "rstrip": false,
49
+ "single_word": false
50
+ }
51
+ }
dynamic_q8/tokenizer_config.json ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<pad>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "<unk>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "64000": {
36
+ "content": "<mask>",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "bos_token": "<s>",
45
+ "clean_up_tokenization_spaces": true,
46
+ "cls_token": "<s>",
47
+ "eos_token": "</s>",
48
+ "mask_token": "<mask>",
49
+ "model_max_length": 512,
50
+ "pad_token": "<pad>",
51
+ "sep_token": "</s>",
52
+ "tokenizer_class": "PhobertTokenizer",
53
+ "unk_token": "<unk>"
54
+ }
dynamic_q8/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
optimized/added_tokens.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "<mask>": 64000
3
+ }
optimized/bpe.codes ADDED
The diff for this file is too large to render. See raw diff
 
optimized/config.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "vietnamese-embedding\\config.json",
3
+ "architectures": [
4
+ "RobertaModel"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": 0,
8
+ "classifier_dropout": null,
9
+ "eos_token_id": 2,
10
+ "gradient_checkpointing": false,
11
+ "hidden_act": "gelu",
12
+ "hidden_dropout_prob": 0.1,
13
+ "hidden_size": 768,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 3072,
16
+ "layer_norm_eps": 1e-05,
17
+ "max_position_embeddings": 258,
18
+ "model_type": "roberta",
19
+ "num_attention_heads": 12,
20
+ "num_hidden_layers": 12,
21
+ "pad_token_id": 1,
22
+ "position_embedding_type": "absolute",
23
+ "tokenizer_class": "PhobertTokenizer",
24
+ "torch_dtype": "float32",
25
+ "transformers_version": "4.40.1",
26
+ "type_vocab_size": 1,
27
+ "use_cache": true,
28
+ "vocab_size": 64001
29
+ }
optimized/model_optimized.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a7505b37bcde94d195fbedaa3197f0fb7a0407664d55c840aff1e24eccb1b76d
3
+ size 537664772
optimized/ort_config.json ADDED
@@ -0,0 +1,39 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "one_external_file": true,
3
+ "opset": null,
4
+ "optimization": {
5
+ "disable_attention": null,
6
+ "disable_attention_fusion": false,
7
+ "disable_bias_gelu": null,
8
+ "disable_bias_gelu_fusion": false,
9
+ "disable_bias_skip_layer_norm": null,
10
+ "disable_bias_skip_layer_norm_fusion": false,
11
+ "disable_embed_layer_norm": true,
12
+ "disable_embed_layer_norm_fusion": true,
13
+ "disable_gelu": null,
14
+ "disable_gelu_fusion": false,
15
+ "disable_group_norm_fusion": true,
16
+ "disable_layer_norm": null,
17
+ "disable_layer_norm_fusion": false,
18
+ "disable_packed_kv": true,
19
+ "disable_rotary_embeddings": false,
20
+ "disable_shape_inference": false,
21
+ "disable_skip_layer_norm": null,
22
+ "disable_skip_layer_norm_fusion": false,
23
+ "enable_gelu_approximation": true,
24
+ "enable_gemm_fast_gelu_fusion": false,
25
+ "enable_transformers_specific_optimizations": true,
26
+ "fp16": false,
27
+ "no_attention_mask": false,
28
+ "optimization_level": 2,
29
+ "optimize_for_gpu": true,
30
+ "optimize_with_onnxruntime_only": null,
31
+ "use_mask_index": false,
32
+ "use_multi_head_attention": false,
33
+ "use_raw_attention_mask": false
34
+ },
35
+ "optimum_version": "1.19.1",
36
+ "quantization": {},
37
+ "transformers_version": "4.40.1",
38
+ "use_external_data_format": false
39
+ }
optimized/special_tokens_map.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "cls_token": {
10
+ "content": "<s>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "eos_token": {
17
+ "content": "</s>",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "mask_token": {
24
+ "content": "<mask>",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "pad_token": {
31
+ "content": "<pad>",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ },
37
+ "sep_token": {
38
+ "content": "</s>",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false
43
+ },
44
+ "unk_token": {
45
+ "content": "<unk>",
46
+ "lstrip": false,
47
+ "normalized": false,
48
+ "rstrip": false,
49
+ "single_word": false
50
+ }
51
+ }
optimized/tokenizer_config.json ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<pad>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "<unk>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "64000": {
36
+ "content": "<mask>",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "bos_token": "<s>",
45
+ "clean_up_tokenization_spaces": true,
46
+ "cls_token": "<s>",
47
+ "eos_token": "</s>",
48
+ "mask_token": "<mask>",
49
+ "model_max_length": 512,
50
+ "pad_token": "<pad>",
51
+ "sep_token": "</s>",
52
+ "tokenizer_class": "PhobertTokenizer",
53
+ "unk_token": "<unk>"
54
+ }
optimized/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
static_q8/added_tokens.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "<mask>": 64000
3
+ }
static_q8/bpe.codes ADDED
The diff for this file is too large to render. See raw diff
 
static_q8/config.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "onnx_v2",
3
+ "architectures": [
4
+ "RobertaModel"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": 0,
8
+ "classifier_dropout": null,
9
+ "eos_token_id": 2,
10
+ "gradient_checkpointing": false,
11
+ "hidden_act": "gelu",
12
+ "hidden_dropout_prob": 0.1,
13
+ "hidden_size": 768,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 3072,
16
+ "layer_norm_eps": 1e-05,
17
+ "max_position_embeddings": 258,
18
+ "model_type": "roberta",
19
+ "num_attention_heads": 12,
20
+ "num_hidden_layers": 12,
21
+ "pad_token_id": 1,
22
+ "position_embedding_type": "absolute",
23
+ "tokenizer_class": "PhobertTokenizer",
24
+ "torch_dtype": "float32",
25
+ "transformers_version": "4.40.1",
26
+ "type_vocab_size": 1,
27
+ "use_cache": true,
28
+ "vocab_size": 64001
29
+ }
static_q8/model_quantized.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c72afa92c9bb7523d5937dd5e37e8040a60ab5aa5350c1bf32f754c7eb431405
3
+ size 135308653
static_q8/ort_config.json ADDED
@@ -0,0 +1,47 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "one_external_file": true,
3
+ "opset": null,
4
+ "optimization": {},
5
+ "optimum_version": "1.19.1",
6
+ "quantization": {
7
+ "activations_dtype": "QUInt8",
8
+ "activations_symmetric": false,
9
+ "format": "QDQ",
10
+ "is_static": true,
11
+ "mode": "QLinearOps",
12
+ "nodes_to_exclude": [],
13
+ "nodes_to_quantize": [],
14
+ "operators_to_quantize": [
15
+ "Conv",
16
+ "ConvTranspose",
17
+ "Gemm",
18
+ "Clip",
19
+ "Relu",
20
+ "Reshape",
21
+ "Transpose",
22
+ "Squeeze",
23
+ "Unsqueeze",
24
+ "Resize",
25
+ "MaxPool",
26
+ "AveragePool",
27
+ "MatMul",
28
+ "Split",
29
+ "Gather",
30
+ "Softmax",
31
+ "Where",
32
+ "InstanceNormalization",
33
+ "LayerNormalization"
34
+ ],
35
+ "per_channel": false,
36
+ "qdq_add_pair_to_weight": false,
37
+ "qdq_dedicated_pair": false,
38
+ "qdq_op_type_per_channel_support_to_axis": {
39
+ "MatMul": 1
40
+ },
41
+ "reduce_range": false,
42
+ "weights_dtype": "QInt8",
43
+ "weights_symmetric": true
44
+ },
45
+ "transformers_version": "4.40.1",
46
+ "use_external_data_format": false
47
+ }
static_q8/special_tokens_map.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "cls_token": {
10
+ "content": "<s>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "eos_token": {
17
+ "content": "</s>",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "mask_token": {
24
+ "content": "<mask>",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "pad_token": {
31
+ "content": "<pad>",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ },
37
+ "sep_token": {
38
+ "content": "</s>",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false
43
+ },
44
+ "unk_token": {
45
+ "content": "<unk>",
46
+ "lstrip": false,
47
+ "normalized": false,
48
+ "rstrip": false,
49
+ "single_word": false
50
+ }
51
+ }
static_q8/tokenizer_config.json ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<pad>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "<unk>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "64000": {
36
+ "content": "<mask>",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "bos_token": "<s>",
45
+ "clean_up_tokenization_spaces": true,
46
+ "cls_token": "<s>",
47
+ "eos_token": "</s>",
48
+ "mask_token": "<mask>",
49
+ "model_max_length": 512,
50
+ "pad_token": "<pad>",
51
+ "sep_token": "</s>",
52
+ "tokenizer_class": "PhobertTokenizer",
53
+ "unk_token": "<unk>"
54
+ }
static_q8/vocab.txt ADDED
The diff for this file is too large to render. See raw diff