Hugging Face's logo

OpenVINO
/
bge-reranker-base-fp16-ov

OpenVINO
English
xlm-roberta
Model card Files
xet
Community
openvino-ci commited on
Commit
13894d5
·
verified ·
1 Parent(s): ce0f5aa

Upload folder using huggingface_hub

Browse files
Files changed (13) hide show
  1. .gitattributes +1 -0
  2. README.md +77 -0
  3. config.json +34 -0
  4. openvino_detokenizer.bin +3 -0
  5. openvino_detokenizer.xml +324 -0
  6. openvino_model.bin +3 -0
  7. openvino_model.xml +0 -0
  8. openvino_tokenizer.bin +3 -0
  9. openvino_tokenizer.xml +923 -0
  10. sentencepiece.bpe.model +3 -0
  11. special_tokens_map.json +51 -0
  12. tokenizer.json +3 -0
  13. tokenizer_config.json +56 -0
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
README.md ADDED
@@ -0,0 +1,77 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: mit
3
+ language:
4
+ - en
5
+ license_link: https://github.com/FlagOpen/FlagEmbedding/blob/master/LICENSE
6
+ base_model:
7
+ - BAAI/bge-base-en-v1.5
8
+ ---
9
+ # bge-base-en-v1.5-fp16-ov
10
+ * Model creator: [BAAI](https://huggingface.co/BAAI)
11
+ * Original model: [bge-base-en-v1.5](https://huggingface.co/BAAI/bge-base-en-v1.5)
12
+
13
+ ## Description
14
+ This is [bge-base-en-v1.5](https://huggingface.co/BAAI/bge-base-en-v1.5) model converted to the [OpenVINO™ IR](https://docs.openvino.ai/2025/documentation/openvino-ir-format.html) (Intermediate Representation) format with weights compressed to FP16.
15
+
16
+ **Disclaimer**: Model is provided as a preview and may be update in the future.
17
+
18
+ ## Compatibility
19
+
20
+ The provided OpenVINO™ IR model is compatible with:
21
+
22
+ * OpenVINO version 2025.1.0 and higher
23
+ * Optimum Intel 1.24.0 and higher
24
+
25
+
26
+ ## Running Model Inference with [Optimum Intel](https://huggingface.co/docs/optimum/intel/index)
27
+
28
+ 1. Install packages required for using [Optimum Intel](https://huggingface.co/docs/optimum/intel/index) integration with the OpenVINO backend:
29
+
30
+ ```
31
+ pip install "langchain-community>=0.2.15" optimum[openvino] huggingface_hub
32
+ ```
33
+
34
+ 2. Run model inference:
35
+
36
+ ```
37
+ from langchain_community.embeddings import OpenVINOBgeEmbeddings
38
+
39
+
40
+ embedding_model_name = 'OpenVINO/bge-base-en-v1.5-fp16-ov'
41
+ embedding_model_kwargs = {"device": "CPU", "compile": False}
42
+ encode_kwargs = {
43
+ "mean_pooling": False,
44
+ "normalize_embeddings": True,
45
+ "batch_size": 4,
46
+ }
47
+
48
+ embedding = OpenVINOBgeEmbeddings(
49
+ model_name_or_path=embedding_model_name,
50
+ model_kwargs=embedding_model_kwargs,
51
+ encode_kwargs=encode_kwargs,
52
+ )
53
+
54
+ embedding.ov_model.compile()
55
+
56
+ text = "This is a test document."
57
+ embedding_result = embedding.embed_query(text)
58
+ embedding_result[:3]
59
+ ```
60
+
61
+ For more examples and possible optimizations, refer to the [Inference with Optimum Intel](https://docs.openvino.ai/2025/openvino-workflow-generative/inference-with-optimum-intel.html).
62
+
63
+ You can find more detailed usage examples in OpenVINO Notebooks:
64
+
65
+ - [RAG text generation](https://openvinotoolkit.github.io/openvino_notebooks/?search=RAG+system)
66
+
67
+ ## Limitations
68
+
69
+ Check the original [model card](https://huggingface.co/BAAI/bge-base-en-v1.5) for limitations.
70
+
71
+ ## Legal information
72
+
73
+ The original model is distributed under [MIT](https://github.com/FlagOpen/FlagEmbedding/blob/master/LICENSE) license. More details can be found in [bge-base-en-v1.5](https://huggingface.co/BAAI/bge-base-en-v1.5).
74
+
75
+ ## Disclaimer
76
+
77
+ Intel is committed to respecting human rights and avoiding causing or contributing to adverse impacts on human rights. See [Intel’s Global Human Rights Principles](https://www.intel.com/content/dam/www/central-libraries/us/en/documents/policy-human-rights.pdf). Intel’s products and software are intended only to be used in applications that do not cause or contribute to adverse impacts on human rights.
config.json ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_attn_implementation_autoset": true,
3
+ "_name_or_path": "BAAI/bge-reranker-base",
4
+ "architectures": [
5
+ "XLMRobertaForSequenceClassification"
6
+ ],
7
+ "attention_probs_dropout_prob": 0.1,
8
+ "bos_token_id": 0,
9
+ "classifier_dropout": null,
10
+ "eos_token_id": 2,
11
+ "hidden_act": "gelu",
12
+ "hidden_dropout_prob": 0.1,
13
+ "hidden_size": 768,
14
+ "id2label": {
15
+ "0": "LABEL_0"
16
+ },
17
+ "initializer_range": 0.02,
18
+ "intermediate_size": 3072,
19
+ "label2id": {
20
+ "LABEL_0": 0
21
+ },
22
+ "layer_norm_eps": 1e-05,
23
+ "max_position_embeddings": 514,
24
+ "model_type": "xlm-roberta",
25
+ "num_attention_heads": 12,
26
+ "num_hidden_layers": 12,
27
+ "output_past": true,
28
+ "pad_token_id": 1,
29
+ "position_embedding_type": "absolute",
30
+ "transformers_version": "4.48.3",
31
+ "type_vocab_size": 1,
32
+ "use_cache": true,
33
+ "vocab_size": 250002
34
+ }
openvino_detokenizer.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a016273ee13b5debc5467f3215863b70ef698453ee3343087658fd09a85eec50
3
+ size 4272179
openvino_detokenizer.xml ADDED
@@ -0,0 +1,324 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ <?xml version="1.0"?>
2
+ <net name="detokenizer" version="11">
3
+ <layers>
4
+ <layer id="0" name="Parameter_10206" type="Parameter" version="opset1">
5
+ <data shape="?,?" element_type="i64" />
6
+ <output>
7
+ <port id="0" precision="I64" names="Parameter_10206">
8
+ <dim>-1</dim>
9
+ <dim>-1</dim>
10
+ </port>
11
+ </output>
12
+ </layer>
13
+ <layer id="1" name="Convert_10386" type="Convert" version="opset1">
14
+ <data destination_type="i32" />
15
+ <input>
16
+ <port id="0" precision="I64">
17
+ <dim>-1</dim>
18
+ <dim>-1</dim>
19
+ </port>
20
+ </input>
21
+ <output>
22
+ <port id="1" precision="I32">
23
+ <dim>-1</dim>
24
+ <dim>-1</dim>
25
+ </port>
26
+ </output>
27
+ </layer>
28
+ <layer id="2" name="Constant_10208" type="Const" version="opset1">
29
+ <data element_type="i32" shape="250002" offset="0" size="1000008" />
30
+ <output>
31
+ <port id="0" precision="I32">
32
+ <dim>250002</dim>
33
+ </port>
34
+ </output>
35
+ </layer>
36
+ <layer id="3" name="Constant_10210" type="Const" version="opset1">
37
+ <data element_type="i32" shape="250002" offset="1000008" size="1000008" />
38
+ <output>
39
+ <port id="0" precision="I32">
40
+ <dim>250002</dim>
41
+ </port>
42
+ </output>
43
+ </layer>
44
+ <layer id="4" name="Constant_10212" type="Const" version="opset1">
45
+ <data element_type="u8" shape="2272087" offset="2000016" size="2272087" />
46
+ <output>
47
+ <port id="0" precision="U8">
48
+ <dim>2272087</dim>
49
+ </port>
50
+ </output>
51
+ </layer>
52
+ <layer id="5" name="Slice_10217" type="Const" version="opset1">
53
+ <data element_type="i32" shape="5" offset="4272103" size="20" />
54
+ <output>
55
+ <port id="0" precision="I32">
56
+ <dim>5</dim>
57
+ </port>
58
+ </output>
59
+ </layer>
60
+ <layer id="6" name="VocabDecoder_10219" type="VocabDecoder" version="extension">
61
+ <data skip_tokens="" />
62
+ <input>
63
+ <port id="0" precision="I32">
64
+ <dim>-1</dim>
65
+ <dim>-1</dim>
66
+ </port>
67
+ <port id="1" precision="I32">
68
+ <dim>250002</dim>
69
+ </port>
70
+ <port id="2" precision="I32">
71
+ <dim>250002</dim>
72
+ </port>
73
+ <port id="3" precision="U8">
74
+ <dim>2272087</dim>
75
+ </port>
76
+ <port id="4" precision="I32">
77
+ <dim>5</dim>
78
+ </port>
79
+ </input>
80
+ <output>
81
+ <port id="5" precision="I32">
82
+ <dim>-1</dim>
83
+ </port>
84
+ <port id="6" precision="I32">
85
+ <dim>-1</dim>
86
+ </port>
87
+ <port id="7" precision="I32">
88
+ <dim>-1</dim>
89
+ </port>
90
+ <port id="8" precision="I32">
91
+ <dim>-1</dim>
92
+ </port>
93
+ <port id="9" precision="U8">
94
+ <dim>-1</dim>
95
+ </port>
96
+ </output>
97
+ </layer>
98
+ <layer id="7" name="FuzeRagged_10220" type="FuzeRagged" version="extension">
99
+ <input>
100
+ <port id="0" precision="I32">
101
+ <dim>-1</dim>
102
+ </port>
103
+ <port id="1" precision="I32">
104
+ <dim>-1</dim>
105
+ </port>
106
+ <port id="2" precision="I32">
107
+ <dim>-1</dim>
108
+ </port>
109
+ <port id="3" precision="I32">
110
+ <dim>-1</dim>
111
+ </port>
112
+ </input>
113
+ <output>
114
+ <port id="4" precision="I32">
115
+ <dim>-1</dim>
116
+ </port>
117
+ <port id="5" precision="I32">
118
+ <dim>-1</dim>
119
+ </port>
120
+ </output>
121
+ </layer>
122
+ <layer id="8" name="Constant_10222" type="Const" version="opset1">
123
+ <data element_type="u8" shape="2" offset="4272123" size="2" />
124
+ <output>
125
+ <port id="0" precision="U8">
126
+ <dim>2</dim>
127
+ </port>
128
+ </output>
129
+ </layer>
130
+ <layer id="9" name="Constant_10224" type="Const" version="opset1">
131
+ <data element_type="u8" shape="0" offset="4272125" size="1" />
132
+ <output>
133
+ <port id="0" precision="U8">
134
+ <dim>0</dim>
135
+ </port>
136
+ </output>
137
+ </layer>
138
+ <layer id="10" name="RegexNormalization_10225" type="RegexNormalization" version="extension">
139
+ <data global_replace="true" />
140
+ <input>
141
+ <port id="0" precision="I32">
142
+ <dim>-1</dim>
143
+ </port>
144
+ <port id="1" precision="I32">
145
+ <dim>-1</dim>
146
+ </port>
147
+ <port id="2" precision="U8">
148
+ <dim>-1</dim>
149
+ </port>
150
+ <port id="3" precision="U8">
151
+ <dim>2</dim>
152
+ </port>
153
+ <port id="4" precision="U8">
154
+ <dim>0</dim>
155
+ </port>
156
+ </input>
157
+ <output>
158
+ <port id="5" precision="I32">
159
+ <dim>-1</dim>
160
+ </port>
161
+ <port id="6" precision="I32">
162
+ <dim>-1</dim>
163
+ </port>
164
+ <port id="7" precision="U8">
165
+ <dim>-1</dim>
166
+ </port>
167
+ </output>
168
+ </layer>
169
+ <layer id="11" name="UTF8Validate_10226" type="UTF8Validate" version="extension">
170
+ <data replace_mode="true" />
171
+ <input>
172
+ <port id="0" precision="I32">
173
+ <dim>-1</dim>
174
+ </port>
175
+ <port id="1" precision="I32">
176
+ <dim>-1</dim>
177
+ </port>
178
+ <port id="2" precision="U8">
179
+ <dim>-1</dim>
180
+ </port>
181
+ </input>
182
+ <output>
183
+ <port id="3" precision="I32">
184
+ <dim>-1</dim>
185
+ </port>
186
+ <port id="4" precision="I32">
187
+ <dim>-1</dim>
188
+ </port>
189
+ <port id="5" precision="U8">
190
+ <dim>-1</dim>
191
+ </port>
192
+ </output>
193
+ </layer>
194
+ <layer id="12" name="Constant_10228" type="Const" version="opset1">
195
+ <data element_type="u8" shape="51" offset="4272126" size="51" />
196
+ <output>
197
+ <port id="0" precision="U8">
198
+ <dim>51</dim>
199
+ </port>
200
+ </output>
201
+ </layer>
202
+ <layer id="13" name="Constant_10230" type="Const" version="opset1">
203
+ <data element_type="u8" shape="2" offset="4272177" size="2" />
204
+ <output>
205
+ <port id="0" precision="U8">
206
+ <dim>2</dim>
207
+ </port>
208
+ </output>
209
+ </layer>
210
+ <layer id="14" name="RegexNormalization_10231" type="RegexNormalization" version="extension">
211
+ <data global_replace="true" />
212
+ <input>
213
+ <port id="0" precision="I32">
214
+ <dim>-1</dim>
215
+ </port>
216
+ <port id="1" precision="I32">
217
+ <dim>-1</dim>
218
+ </port>
219
+ <port id="2" precision="U8">
220
+ <dim>-1</dim>
221
+ </port>
222
+ <port id="3" precision="U8">
223
+ <dim>51</dim>
224
+ </port>
225
+ <port id="4" precision="U8">
226
+ <dim>2</dim>
227
+ </port>
228
+ </input>
229
+ <output>
230
+ <port id="5" precision="I32">
231
+ <dim>-1</dim>
232
+ </port>
233
+ <port id="6" precision="I32">
234
+ <dim>-1</dim>
235
+ </port>
236
+ <port id="7" precision="U8">
237
+ <dim>-1</dim>
238
+ </port>
239
+ </output>
240
+ </layer>
241
+ <layer id="15" name="StringTensorPack_10232" type="StringTensorPack" version="opset15">
242
+ <input>
243
+ <port id="0" precision="I32">
244
+ <dim>-1</dim>
245
+ </port>
246
+ <port id="1" precision="I32">
247
+ <dim>-1</dim>
248
+ </port>
249
+ <port id="2" precision="U8">
250
+ <dim>-1</dim>
251
+ </port>
252
+ </input>
253
+ <output>
254
+ <port id="3" precision="STRING" names="Result_10233,string_output">
255
+ <dim>-1</dim>
256
+ </port>
257
+ </output>
258
+ </layer>
259
+ <layer id="16" name="Result_10233" type="Result" version="opset1" output_names="Result_10233,string_output">
260
+ <input>
261
+ <port id="0" precision="STRING">
262
+ <dim>-1</dim>
263
+ </port>
264
+ </input>
265
+ </layer>
266
+ </layers>
267
+ <edges>
268
+ <edge from-layer="0" from-port="0" to-layer="1" to-port="0" />
269
+ <edge from-layer="1" from-port="1" to-layer="6" to-port="0" />
270
+ <edge from-layer="2" from-port="0" to-layer="6" to-port="1" />
271
+ <edge from-layer="3" from-port="0" to-layer="6" to-port="2" />
272
+ <edge from-layer="4" from-port="0" to-layer="6" to-port="3" />
273
+ <edge from-layer="5" from-port="0" to-layer="6" to-port="4" />
274
+ <edge from-layer="6" from-port="5" to-layer="7" to-port="0" />
275
+ <edge from-layer="6" from-port="6" to-layer="7" to-port="1" />
276
+ <edge from-layer="6" from-port="7" to-layer="7" to-port="2" />
277
+ <edge from-layer="6" from-port="8" to-layer="7" to-port="3" />
278
+ <edge from-layer="6" from-port="9" to-layer="10" to-port="2" />
279
+ <edge from-layer="7" from-port="4" to-layer="10" to-port="0" />
280
+ <edge from-layer="7" from-port="5" to-layer="10" to-port="1" />
281
+ <edge from-layer="8" from-port="0" to-layer="10" to-port="3" />
282
+ <edge from-layer="9" from-port="0" to-layer="10" to-port="4" />
283
+ <edge from-layer="10" from-port="5" to-layer="11" to-port="0" />
284
+ <edge from-layer="10" from-port="6" to-layer="11" to-port="1" />
285
+ <edge from-layer="10" from-port="7" to-layer="11" to-port="2" />
286
+ <edge from-layer="11" from-port="3" to-layer="14" to-port="0" />
287
+ <edge from-layer="11" from-port="4" to-layer="14" to-port="1" />
288
+ <edge from-layer="11" from-port="5" to-layer="14" to-port="2" />
289
+ <edge from-layer="12" from-port="0" to-layer="14" to-port="3" />
290
+ <edge from-layer="13" from-port="0" to-layer="14" to-port="4" />
291
+ <edge from-layer="14" from-port="5" to-layer="15" to-port="0" />
292
+ <edge from-layer="14" from-port="6" to-layer="15" to-port="1" />
293
+ <edge from-layer="14" from-port="7" to-layer="15" to-port="2" />
294
+ <edge from-layer="15" from-port="3" to-layer="16" to-port="0" />
295
+ </edges>
296
+ <rt_info>
297
+ <add_attention_mask value="True" />
298
+ <add_prefix_space />
299
+ <add_special_tokens value="True" />
300
+ <bos_token_id value="0" />
301
+ <clean_up_tokenization_spaces />
302
+ <detokenizer_input_type value="i64" />
303
+ <eos_token_id value="2" />
304
+ <handle_special_tokens_with_re />
305
+ <max_length />
306
+ <number_of_inputs value="1" />
307
+ <openvino_tokenizers_version value="2025.1.0.0-523-710ddf14de8" />
308
+ <openvino_version value="2025.1.0-18503-6fec06580ab-releases/2025/1" />
309
+ <original_post_processor_template value="{&quot;type&quot;: &quot;TemplateProcessing&quot;, &quot;single&quot;: [{&quot;SpecialToken&quot;: {&quot;id&quot;: &quot;&lt;s>&quot;, &quot;type_id&quot;: 0}}, {&quot;Sequence&quot;: {&quot;id&quot;: &quot;A&quot;, &quot;type_id&quot;: 0}}, {&quot;SpecialToken&quot;: {&quot;id&quot;: &quot;&lt;/s>&quot;, &quot;type_id&quot;: 0}}], &quot;pair&quot;: [{&quot;SpecialToken&quot;: {&quot;id&quot;: &quot;&lt;s>&quot;, &quot;type_id&quot;: 0}}, {&quot;Sequence&quot;: {&quot;id&quot;: &quot;A&quot;, &quot;type_id&quot;: 0}}, {&quot;SpecialToken&quot;: {&quot;id&quot;: &quot;&lt;/s>&quot;, &quot;type_id&quot;: 0}}, {&quot;SpecialToken&quot;: {&quot;id&quot;: &quot;&lt;/s>&quot;, &quot;type_id&quot;: 0}}, {&quot;Sequence&quot;: {&quot;id&quot;: &quot;B&quot;, &quot;type_id&quot;: 0}}, {&quot;SpecialToken&quot;: {&quot;id&quot;: &quot;&lt;/s>&quot;, &quot;type_id&quot;: 0}}], &quot;special_tokens&quot;: {&quot;&lt;/s>&quot;: {&quot;id&quot;: &quot;&lt;/s>&quot;, &quot;ids&quot;: [2], &quot;tokens&quot;: [&quot;&lt;/s>&quot;]}, &quot;&lt;s>&quot;: {&quot;id&quot;: &quot;&lt;s>&quot;, &quot;ids&quot;: [0], &quot;tokens&quot;: [&quot;&lt;s>&quot;]}}}" />
310
+ <original_tokenizer_class value="&lt;class 'transformers.models.xlm_roberta.tokenization_xlm_roberta_fast.XLMRobertaTokenizerFast'>" />
311
+ <pad_token_id value="1" />
312
+ <processed_post_processor_template value="{&quot;single&quot;: {&quot;ids&quot;: [0, -1, 2], &quot;type_ids&quot;: [0, 0, 0]}, &quot;pair&quot;: {&quot;ids&quot;: [0, -1, 2, 2, -2, 2], &quot;type_ids&quot;: [0, 0, 0, 0, 0, 0]}}" />
313
+ <sentencepiece_version value="0.2.0" />
314
+ <skip_special_tokens value="True" />
315
+ <streaming_detokenizer value="False" />
316
+ <tokenizer_output_type value="i64" />
317
+ <tokenizers_version value="0.21.1" />
318
+ <transformers_version value="4.48.3" />
319
+ <use_max_padding value="False" />
320
+ <use_sentencepiece_backend value="False" />
321
+ <utf8_replace_mode value="replace" />
322
+ <with_detokenizer value="True" />
323
+ </rt_info>
324
+ </net>
openvino_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dc21b1c6835583f59a49ce7b3c15f9fef9ecd49e7d4bee9ce0373a62204efe10
3
+ size 1112181940
openvino_model.xml ADDED
The diff for this file is too large to render. See raw diff
 
openvino_tokenizer.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c32cd4dd9fcb191cb5ecd5b0e3ac9d7d102eb806145e0acd00281395e1b1c142
3
+ size 5319028
openvino_tokenizer.xml ADDED
@@ -0,0 +1,923 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ <?xml version="1.0"?>
2
+ <net name="tokenizer" version="11">
3
+ <layers>
4
+ <layer id="0" name="Parameter_10078" type="Parameter" version="opset1">
5
+ <data shape="?" element_type="string" />
6
+ <output>
7
+ <port id="0" precision="STRING" names="Parameter_10078">
8
+ <dim>-1</dim>
9
+ </port>
10
+ </output>
11
+ </layer>
12
+ <layer id="1" name="Constant_10187" type="Const" version="opset1">
13
+ <data element_type="i32" shape="" offset="0" size="4" />
14
+ <output>
15
+ <port id="0" precision="I32" />
16
+ </output>
17
+ </layer>
18
+ <layer id="2" name="Constant_10188" type="Const" version="opset1">
19
+ <data element_type="i32" shape="" offset="4" size="4" />
20
+ <output>
21
+ <port id="0" precision="I32" />
22
+ </output>
23
+ </layer>
24
+ <layer id="3" name="Constant_10189" type="Const" version="opset1">
25
+ <data element_type="i32" shape="1" offset="0" size="4" />
26
+ <output>
27
+ <port id="0" precision="I32">
28
+ <dim>1</dim>
29
+ </port>
30
+ </output>
31
+ </layer>
32
+ <layer id="4" name="Constant_10084" type="Const" version="opset1">
33
+ <data element_type="i64" shape="" offset="8" size="8" />
34
+ <output>
35
+ <port id="0" precision="I64" />
36
+ </output>
37
+ </layer>
38
+ <layer id="5" name="StringTensorUnpack_10079" type="StringTensorUnpack" version="opset15">
39
+ <input>
40
+ <port id="0" precision="STRING">
41
+ <dim>-1</dim>
42
+ </port>
43
+ </input>
44
+ <output>
45
+ <port id="1" precision="I32">
46
+ <dim>-1</dim>
47
+ </port>
48
+ <port id="2" precision="I32">
49
+ <dim>-1</dim>
50
+ </port>
51
+ <port id="3" precision="U8">
52
+ <dim>-1</dim>
53
+ </port>
54
+ </output>
55
+ </layer>
56
+ <layer id="6" name="ShapeOf_10080" type="ShapeOf" version="opset3">
57
+ <data output_type="i64" />
58
+ <input>
59
+ <port id="0" precision="I32">
60
+ <dim>-1</dim>
61
+ </port>
62
+ </input>
63
+ <output>
64
+ <port id="1" precision="I64">
65
+ <dim>1</dim>
66
+ </port>
67
+ </output>
68
+ </layer>
69
+ <layer id="7" name="Constant_10081" type="Const" version="opset1">
70
+ <data element_type="i64" shape="" offset="8" size="8" />
71
+ <output>
72
+ <port id="0" precision="I64" />
73
+ </output>
74
+ </layer>
75
+ <layer id="8" name="Constant_10082" type="Const" version="opset1">
76
+ <data element_type="i64" shape="" offset="8" size="8" />
77
+ <output>
78
+ <port id="0" precision="I64" />
79
+ </output>
80
+ </layer>
81
+ <layer id="9" name="Gather_10083" type="Gather" version="opset8">
82
+ <data batch_dims="0" />
83
+ <input>
84
+ <port id="0" precision="I64">
85
+ <dim>1</dim>
86
+ </port>
87
+ <port id="1" precision="I64" />
88
+ <port id="2" precision="I64" />
89
+ </input>
90
+ <output>
91
+ <port id="3" precision="I64" />
92
+ </output>
93
+ </layer>
94
+ <layer id="10" name="Constant_10085" type="Const" version="opset1">
95
+ <data element_type="i64" shape="" offset="16" size="8" />
96
+ <output>
97
+ <port id="0" precision="I64" />
98
+ </output>
99
+ </layer>
100
+ <layer id="11" name="Range_10086" type="Range" version="opset4">
101
+ <data output_type="i32" />
102
+ <input>
103
+ <port id="0" precision="I64" />
104
+ <port id="1" precision="I64" />
105
+ <port id="2" precision="I64" />
106
+ </input>
107
+ <output>
108
+ <port id="3" precision="I32">
109
+ <dim>-1</dim>
110
+ </port>
111
+ </output>
112
+ </layer>
113
+ <layer id="12" name="Constant_10087" type="Const" version="opset1">
114
+ <data element_type="i64" shape="" offset="16" size="8" />
115
+ <output>
116
+ <port id="0" precision="I64" />
117
+ </output>
118
+ </layer>
119
+ <layer id="13" name="Constant_10088" type="Const" version="opset1">
120
+ <data element_type="i64" shape="" offset="16" size="8" />
121
+ <output>
122
+ <port id="0" precision="I64" />
123
+ </output>
124
+ </layer>
125
+ <layer id="14" name="Add_10089" type="Add" version="opset1">
126
+ <data auto_broadcast="numpy" />
127
+ <input>
128
+ <port id="0" precision="I64" />
129
+ <port id="1" precision="I64" />
130
+ </input>
131
+ <output>
132
+ <port id="2" precision="I64" />
133
+ </output>
134
+ </layer>
135
+ <layer id="15" name="Constant_10090" type="Const" version="opset1">
136
+ <data element_type="i64" shape="" offset="16" size="8" />
137
+ <output>
138
+ <port id="0" precision="I64" />
139
+ </output>
140
+ </layer>
141
+ <layer id="16" name="Range_10091" type="Range" version="opset4">
142
+ <data output_type="i32" />
143
+ <input>
144
+ <port id="0" precision="I64" />
145
+ <port id="1" precision="I64" />
146
+ <port id="2" precision="I64" />
147
+ </input>
148
+ <output>
149
+ <port id="3" precision="I32">
150
+ <dim>-1</dim>
151
+ </port>
152
+ </output>
153
+ </layer>
154
+ <layer id="17" name="Constant_10153" type="Const" version="opset1">
155
+ <data element_type="u8" shape="55" offset="24" size="55" />
156
+ <output>
157
+ <port id="0" precision="U8">
158
+ <dim>55</dim>
159
+ </port>
160
+ </output>
161
+ </layer>
162
+ <layer id="18" name="SpecialTokensSplit_10154" type="SpecialTokensSplit" version="extension">
163
+ <input>
164
+ <port id="0" precision="I32">
165
+ <dim>-1</dim>
166
+ </port>
167
+ <port id="1" precision="I32">
168
+ <dim>-1</dim>
169
+ </port>
170
+ <port id="2" precision="I32">
171
+ <dim>-1</dim>
172
+ </port>
173
+ <port id="3" precision="I32">
174
+ <dim>-1</dim>
175
+ </port>
176
+ <port id="4" precision="U8">
177
+ <dim>-1</dim>
178
+ </port>
179
+ <port id="5" precision="U8">
180
+ <dim>55</dim>
181
+ </port>
182
+ </input>
183
+ <output>
184
+ <port id="6" precision="I32">
185
+ <dim>-1</dim>
186
+ </port>
187
+ <port id="7" precision="I32">
188
+ <dim>-1</dim>
189
+ </port>
190
+ <port id="8" precision="I32">
191
+ <dim>-1</dim>
192
+ </port>
193
+ <port id="9" precision="I32">
194
+ <dim>-1</dim>
195
+ </port>
196
+ <port id="10" precision="U8">
197
+ <dim>-1</dim>
198
+ </port>
199
+ <port id="11" precision="BOOL">
200
+ <dim>-1</dim>
201
+ </port>
202
+ </output>
203
+ </layer>
204
+ <layer id="19" name="Constant_10155" type="Const" version="opset1">
205
+ <data element_type="u8" shape="237539" offset="79" size="237539" />
206
+ <output>
207
+ <port id="0" precision="U8">
208
+ <dim>237539</dim>
209
+ </port>
210
+ </output>
211
+ </layer>
212
+ <layer id="20" name="CharsMapNormalization_10156" type="CharsMapNormalization" version="extension">
213
+ <data add_dummy_prefix="false" remove_extra_whitespaces="true" escape_whitespaces="false" normalization_form="" case_fold="false" nmt="false" />
214
+ <input>
215
+ <port id="0" precision="I32">
216
+ <dim>-1</dim>
217
+ </port>
218
+ <port id="1" precision="I32">
219
+ <dim>-1</dim>
220
+ </port>
221
+ <port id="2" precision="U8">
222
+ <dim>-1</dim>
223
+ </port>
224
+ <port id="3" precision="BOOL">
225
+ <dim>-1</dim>
226
+ </port>
227
+ <port id="4" precision="U8">
228
+ <dim>237539</dim>
229
+ </port>
230
+ </input>
231
+ <output>
232
+ <port id="5" precision="I32">
233
+ <dim>-1</dim>
234
+ </port>
235
+ <port id="6" precision="I32">
236
+ <dim>-1</dim>
237
+ </port>
238
+ <port id="7" precision="U8">
239
+ <dim>-1</dim>
240
+ </port>
241
+ <port id="8" precision="BOOL">
242
+ <dim>-1</dim>
243
+ </port>
244
+ </output>
245
+ </layer>
246
+ <layer id="21" name="Constant_10158" type="Const" version="opset1">
247
+ <data element_type="u8" shape="5" offset="237618" size="5" />
248
+ <output>
249
+ <port id="0" precision="U8">
250
+ <dim>5</dim>
251
+ </port>
252
+ </output>
253
+ </layer>
254
+ <layer id="22" name="Constant_10160" type="Const" version="opset1">
255
+ <data element_type="u8" shape="1" offset="237623" size="1" />
256
+ <output>
257
+ <port id="0" precision="U8">
258
+ <dim>1</dim>
259
+ </port>
260
+ </output>
261
+ </layer>
262
+ <layer id="23" name="RegexNormalization_10161" type="RegexNormalization" version="extension">
263
+ <data global_replace="true" />
264
+ <input>
265
+ <port id="0" precision="I32">
266
+ <dim>-1</dim>
267
+ </port>
268
+ <port id="1" precision="I32">
269
+ <dim>-1</dim>
270
+ </port>
271
+ <port id="2" precision="U8">
272
+ <dim>-1</dim>
273
+ </port>
274
+ <port id="3" precision="BOOL">
275
+ <dim>-1</dim>
276
+ </port>
277
+ <port id="4" precision="U8">
278
+ <dim>5</dim>
279
+ </port>
280
+ <port id="5" precision="U8">
281
+ <dim>1</dim>
282
+ </port>
283
+ </input>
284
+ <output>
285
+ <port id="6" precision="I32">
286
+ <dim>-1</dim>
287
+ </port>
288
+ <port id="7" precision="I32">
289
+ <dim>-1</dim>
290
+ </port>
291
+ <port id="8" precision="U8">
292
+ <dim>-1</dim>
293
+ </port>
294
+ <port id="9" precision="BOOL">
295
+ <dim>-1</dim>
296
+ </port>
297
+ </output>
298
+ </layer>
299
+ <layer id="24" name="Constant_10163" type="Const" version="opset1">
300
+ <data element_type="u8" shape="1" offset="237623" size="1" />
301
+ <output>
302
+ <port id="0" precision="U8">
303
+ <dim>1</dim>
304
+ </port>
305
+ </output>
306
+ </layer>
307
+ <layer id="25" name="Constant_10165" type="Const" version="opset1">
308
+ <data element_type="u8" shape="3" offset="237624" size="3" />
309
+ <output>
310
+ <port id="0" precision="U8">
311
+ <dim>3</dim>
312
+ </port>
313
+ </output>
314
+ </layer>
315
+ <layer id="26" name="RegexNormalization_10166" type="RegexNormalization" version="extension">
316
+ <data global_replace="true" />
317
+ <input>
318
+ <port id="0" precision="I32">
319
+ <dim>-1</dim>
320
+ </port>
321
+ <port id="1" precision="I32">
322
+ <dim>-1</dim>
323
+ </port>
324
+ <port id="2" precision="U8">
325
+ <dim>-1</dim>
326
+ </port>
327
+ <port id="3" precision="BOOL">
328
+ <dim>-1</dim>
329
+ </port>
330
+ <port id="4" precision="U8">
331
+ <dim>1</dim>
332
+ </port>
333
+ <port id="5" precision="U8">
334
+ <dim>3</dim>
335
+ </port>
336
+ </input>
337
+ <output>
338
+ <port id="6" precision="I32">
339
+ <dim>-1</dim>
340
+ </port>
341
+ <port id="7" precision="I32">
342
+ <dim>-1</dim>
343
+ </port>
344
+ <port id="8" precision="U8">
345
+ <dim>-1</dim>
346
+ </port>
347
+ <port id="9" precision="BOOL">
348
+ <dim>-1</dim>
349
+ </port>
350
+ </output>
351
+ </layer>
352
+ <layer id="27" name="Constant_10168" type="Const" version="opset1">
353
+ <data element_type="u8" shape="11" offset="237627" size="11" />
354
+ <output>
355
+ <port id="0" precision="U8">
356
+ <dim>11</dim>
357
+ </port>
358
+ </output>
359
+ </layer>
360
+ <layer id="28" name="Constant_10170" type="Const" version="opset1">
361
+ <data element_type="u8" shape="5" offset="237638" size="5" />
362
+ <output>
363
+ <port id="0" precision="U8">
364
+ <dim>5</dim>
365
+ </port>
366
+ </output>
367
+ </layer>
368
+ <layer id="29" name="RegexNormalization_10171" type="RegexNormalization" version="extension">
369
+ <data global_replace="true" />
370
+ <input>
371
+ <port id="0" precision="I32">
372
+ <dim>-1</dim>
373
+ </port>
374
+ <port id="1" precision="I32">
375
+ <dim>-1</dim>
376
+ </port>
377
+ <port id="2" precision="U8">
378
+ <dim>-1</dim>
379
+ </port>
380
+ <port id="3" precision="BOOL">
381
+ <dim>-1</dim>
382
+ </port>
383
+ <port id="4" precision="U8">
384
+ <dim>11</dim>
385
+ </port>
386
+ <port id="5" precision="U8">
387
+ <dim>5</dim>
388
+ </port>
389
+ </input>
390
+ <output>
391
+ <port id="6" precision="I32">
392
+ <dim>-1</dim>
393
+ </port>
394
+ <port id="7" precision="I32">
395
+ <dim>-1</dim>
396
+ </port>
397
+ <port id="8" precision="U8">
398
+ <dim>-1</dim>
399
+ </port>
400
+ <port id="9" precision="BOOL">
401
+ <dim>-1</dim>
402
+ </port>
403
+ </output>
404
+ </layer>
405
+ <layer id="30" name="Constant_10173" type="Const" version="opset1">
406
+ <data element_type="u8" shape="3" offset="237624" size="3" />
407
+ <output>
408
+ <port id="0" precision="U8">
409
+ <dim>3</dim>
410
+ </port>
411
+ </output>
412
+ </layer>
413
+ <layer id="31" name="RegexSplit_10174" type="RegexSplit" version="extension">
414
+ <data behaviour="mergedwithnext" invert="false" max_splits="-1" />
415
+ <input>
416
+ <port id="0" precision="I32">
417
+ <dim>-1</dim>
418
+ </port>
419
+ <port id="1" precision="I32">
420
+ <dim>-1</dim>
421
+ </port>
422
+ <port id="2" precision="I32">
423
+ <dim>-1</dim>
424
+ </port>
425
+ <port id="3" precision="I32">
426
+ <dim>-1</dim>
427
+ </port>
428
+ <port id="4" precision="U8">
429
+ <dim>-1</dim>
430
+ </port>
431
+ <port id="5" precision="BOOL">
432
+ <dim>-1</dim>
433
+ </port>
434
+ <port id="6" precision="U8">
435
+ <dim>3</dim>
436
+ </port>
437
+ </input>
438
+ <output>
439
+ <port id="7" precision="I32">
440
+ <dim>-1</dim>
441
+ </port>
442
+ <port id="8" precision="I32">
443
+ <dim>-1</dim>
444
+ </port>
445
+ <port id="9" precision="I32">
446
+ <dim>-1</dim>
447
+ </port>
448
+ <port id="10" precision="I32">
449
+ <dim>-1</dim>
450
+ </port>
451
+ <port id="11" precision="U8">
452
+ <dim>-1</dim>
453
+ </port>
454
+ <port id="12" precision="BOOL">
455
+ <dim>-1</dim>
456
+ </port>
457
+ </output>
458
+ </layer>
459
+ <layer id="32" name="Constant_10176" type="Const" version="opset1">
460
+ <data element_type="i32" shape="250002" offset="237643" size="1000008" />
461
+ <output>
462
+ <port id="0" precision="I32">
463
+ <dim>250002</dim>
464
+ </port>
465
+ </output>
466
+ </layer>
467
+ <layer id="33" name="Constant_10178" type="Const" version="opset1">
468
+ <data element_type="i32" shape="250002" offset="1237651" size="1000008" />
469
+ <output>
470
+ <port id="0" precision="I32">
471
+ <dim>250002</dim>
472
+ </port>
473
+ </output>
474
+ </layer>
475
+ <layer id="34" name="Constant_10180" type="Const" version="opset1">
476
+ <data element_type="u8" shape="2581345" offset="2237659" size="2581345" />
477
+ <output>
478
+ <port id="0" precision="U8">
479
+ <dim>2581345</dim>
480
+ </port>
481
+ </output>
482
+ </layer>
483
+ <layer id="35" name="Constant_10181_compressed" type="Const" version="opset1">
484
+ <data element_type="f16" shape="250002" offset="4819004" size="500004" />
485
+ <output>
486
+ <port id="0" precision="FP16">
487
+ <dim>250002</dim>
488
+ </port>
489
+ </output>
490
+ </layer>
491
+ <layer id="36" name="Constant_10181" type="Convert" version="opset1">
492
+ <data destination_type="f32" />
493
+ <rt_info>
494
+ <attribute name="decompression" version="0" />
495
+ </rt_info>
496
+ <input>
497
+ <port id="0" precision="FP16">
498
+ <dim>250002</dim>
499
+ </port>
500
+ </input>
501
+ <output>
502
+ <port id="1" precision="FP32">
503
+ <dim>250002</dim>
504
+ </port>
505
+ </output>
506
+ </layer>
507
+ <layer id="37" name="UnigramTokenizer_10182" type="UnigramTokenizer" version="extension">
508
+ <data byte_fallback="false" unk_token_id="3" fuse_unk="true" min_score="inf" />
509
+ <input>
510
+ <port id="0" precision="I32">
511
+ <dim>-1</dim>
512
+ </port>
513
+ <port id="1" precision="I32">
514
+ <dim>-1</dim>
515
+ </port>
516
+ <port id="2" precision="I32">
517
+ <dim>-1</dim>
518
+ </port>
519
+ <port id="3" precision="I32">
520
+ <dim>-1</dim>
521
+ </port>
522
+ <port id="4" precision="U8">
523
+ <dim>-1</dim>
524
+ </port>
525
+ <port id="5" precision="I32">
526
+ <dim>250002</dim>
527
+ </port>
528
+ <port id="6" precision="I32">
529
+ <dim>250002</dim>
530
+ </port>
531
+ <port id="7" precision="U8">
532
+ <dim>2581345</dim>
533
+ </port>
534
+ <port id="8" precision="FP32">
535
+ <dim>250002</dim>
536
+ </port>
537
+ </input>
538
+ <output>
539
+ <port id="9" precision="I32">
540
+ <dim>-1</dim>
541
+ </port>
542
+ <port id="10" precision="I32">
543
+ <dim>-1</dim>
544
+ </port>
545
+ <port id="11" precision="I32">
546
+ <dim>-1</dim>
547
+ </port>
548
+ </output>
549
+ </layer>
550
+ <layer id="38" name="Subtract_10183" type="Subtract" version="opset1">
551
+ <data auto_broadcast="numpy" />
552
+ <input>
553
+ <port id="0" precision="I32">
554
+ <dim>-1</dim>
555
+ </port>
556
+ <port id="1" precision="I32">
557
+ <dim>-1</dim>
558
+ </port>
559
+ </input>
560
+ <output>
561
+ <port id="2" precision="I32">
562
+ <dim>-1</dim>
563
+ </port>
564
+ </output>
565
+ </layer>
566
+ <layer id="39" name="Constant_10184" type="Const" version="opset1">
567
+ <data element_type="i32" shape="" offset="5319008" size="4" />
568
+ <output>
569
+ <port id="0" precision="I32" />
570
+ </output>
571
+ </layer>
572
+ <layer id="40" name="Minimum_10185" type="Minimum" version="opset1">
573
+ <data auto_broadcast="numpy" />
574
+ <input>
575
+ <port id="0" precision="I32">
576
+ <dim>-1</dim>
577
+ </port>
578
+ <port id="1" precision="I32" />
579
+ </input>
580
+ <output>
581
+ <port id="2" precision="I32">
582
+ <dim>-1</dim>
583
+ </port>
584
+ </output>
585
+ </layer>
586
+ <layer id="41" name="Add_10186" type="Add" version="opset1">
587
+ <data auto_broadcast="numpy" />
588
+ <input>
589
+ <port id="0" precision="I32">
590
+ <dim>-1</dim>
591
+ </port>
592
+ <port id="1" precision="I32">
593
+ <dim>-1</dim>
594
+ </port>
595
+ </input>
596
+ <output>
597
+ <port id="2" precision="I32">
598
+ <dim>-1</dim>
599
+ </port>
600
+ </output>
601
+ </layer>
602
+ <layer id="42" name="Constant_10190" type="Const" version="opset1">
603
+ <data element_type="i32" shape="" offset="0" size="4" />
604
+ <output>
605
+ <port id="0" precision="I32" />
606
+ </output>
607
+ </layer>
608
+ <layer id="43" name="Constant_10191" type="Const" version="opset1">
609
+ <data element_type="i32" shape="" offset="4" size="4" />
610
+ <output>
611
+ <port id="0" precision="I32" />
612
+ </output>
613
+ </layer>
614
+ <layer id="44" name="Constant_10192" type="Const" version="opset1">
615
+ <data element_type="i32" shape="1" offset="5319012" size="4" />
616
+ <output>
617
+ <port id="0" precision="I32">
618
+ <dim>1</dim>
619
+ </port>
620
+ </output>
621
+ </layer>
622
+ <layer id="45" name="Constant_10193" type="Const" version="opset1">
623
+ <data element_type="i32" shape="3" offset="5319016" size="12" />
624
+ <output>
625
+ <port id="0" precision="I32">
626
+ <dim>3</dim>
627
+ </port>
628
+ </output>
629
+ </layer>
630
+ <layer id="46" name="CombineSegments_10194" type="CombineSegments" version="extension">
631
+ <input>
632
+ <port id="0" precision="I32" />
633
+ <port id="1" precision="I32" />
634
+ <port id="2" precision="I32">
635
+ <dim>1</dim>
636
+ </port>
637
+ <port id="3" precision="I32">
638
+ <dim>-1</dim>
639
+ </port>
640
+ <port id="4" precision="I32">
641
+ <dim>-1</dim>
642
+ </port>
643
+ <port id="5" precision="I32">
644
+ <dim>-1</dim>
645
+ </port>
646
+ <port id="6" precision="I32" />
647
+ <port id="7" precision="I32" />
648
+ <port id="8" precision="I32">
649
+ <dim>1</dim>
650
+ </port>
651
+ <port id="9" precision="I32">
652
+ <dim>3</dim>
653
+ </port>
654
+ </input>
655
+ <output>
656
+ <port id="10" precision="I32">
657
+ <dim>-1</dim>
658
+ </port>
659
+ <port id="11" precision="I32">
660
+ <dim>-1</dim>
661
+ </port>
662
+ <port id="12" precision="I32">
663
+ <dim>-1</dim>
664
+ </port>
665
+ <port id="13" precision="I32">
666
+ <dim>-1</dim>
667
+ </port>
668
+ <port id="14" precision="I32">
669
+ <dim>-1</dim>
670
+ </port>
671
+ <port id="15" precision="I32">
672
+ <dim>-1</dim>
673
+ </port>
674
+ </output>
675
+ </layer>
676
+ <layer id="47" name="Subtract_10195" type="Subtract" version="opset1">
677
+ <data auto_broadcast="numpy" />
678
+ <input>
679
+ <port id="0" precision="I32">
680
+ <dim>-1</dim>
681
+ </port>
682
+ <port id="1" precision="I32">
683
+ <dim>-1</dim>
684
+ </port>
685
+ </input>
686
+ <output>
687
+ <port id="2" precision="I32">
688
+ <dim>-1</dim>
689
+ </port>
690
+ </output>
691
+ </layer>
692
+ <layer id="48" name="Constant_10196" type="Const" version="opset1">
693
+ <data element_type="i32" shape="" offset="0" size="4" />
694
+ <output>
695
+ <port id="0" precision="I32" />
696
+ </output>
697
+ </layer>
698
+ <layer id="49" name="ReduceMax_10197" type="ReduceMax" version="opset1">
699
+ <data keep_dims="false" />
700
+ <input>
701
+ <port id="0" precision="I32">
702
+ <dim>-1</dim>
703
+ </port>
704
+ <port id="1" precision="I32" />
705
+ </input>
706
+ <output>
707
+ <port id="2" precision="I32" />
708
+ </output>
709
+ </layer>
710
+ <layer id="50" name="Constant_10198" type="Const" version="opset1">
711
+ <data element_type="i32" shape="" offset="4" size="4" />
712
+ <output>
713
+ <port id="0" precision="I32" />
714
+ </output>
715
+ </layer>
716
+ <layer id="51" name="RaggedToDense_10199" type="RaggedToDense" version="extension">
717
+ <data pad_right="true" m_pad_max_length="false" />
718
+ <input>
719
+ <port id="0" precision="I32">
720
+ <dim>-1</dim>
721
+ </port>
722
+ <port id="1" precision="I32">
723
+ <dim>-1</dim>
724
+ </port>
725
+ <port id="2" precision="I32">
726
+ <dim>-1</dim>
727
+ </port>
728
+ <port id="3" precision="I32" />
729
+ <port id="4" precision="I32" />
730
+ </input>
731
+ <output>
732
+ <port id="5" precision="I32">
733
+ <dim>-1</dim>
734
+ <dim>-1</dim>
735
+ </port>
736
+ <port id="6" precision="BOOL">
737
+ <dim>-1</dim>
738
+ <dim>-1</dim>
739
+ </port>
740
+ </output>
741
+ </layer>
742
+ <layer id="52" name="Convert_10200" type="Convert" version="opset1">
743
+ <data destination_type="i32" />
744
+ <input>
745
+ <port id="0" precision="BOOL">
746
+ <dim>-1</dim>
747
+ <dim>-1</dim>
748
+ </port>
749
+ </input>
750
+ <output>
751
+ <port id="1" precision="I32">
752
+ <dim>-1</dim>
753
+ <dim>-1</dim>
754
+ </port>
755
+ </output>
756
+ </layer>
757
+ <layer id="53" name="Convert_10200.0" type="Convert" version="opset1">
758
+ <data destination_type="i64" />
759
+ <input>
760
+ <port id="0" precision="I32">
761
+ <dim>-1</dim>
762
+ <dim>-1</dim>
763
+ </port>
764
+ </input>
765
+ <output>
766
+ <port id="1" precision="I64" names="attention_mask">
767
+ <dim>-1</dim>
768
+ <dim>-1</dim>
769
+ </port>
770
+ </output>
771
+ </layer>
772
+ <layer id="55" name="RaggedToDense_10199.0" type="Convert" version="opset1">
773
+ <data destination_type="i64" />
774
+ <input>
775
+ <port id="0" precision="I32">
776
+ <dim>-1</dim>
777
+ <dim>-1</dim>
778
+ </port>
779
+ </input>
780
+ <output>
781
+ <port id="1" precision="I64" names="input_ids">
782
+ <dim>-1</dim>
783
+ <dim>-1</dim>
784
+ </port>
785
+ </output>
786
+ </layer>
787
+ <layer id="56" name="Result_10203" type="Result" version="opset1" output_names="input_ids">
788
+ <input>
789
+ <port id="0" precision="I64">
790
+ <dim>-1</dim>
791
+ <dim>-1</dim>
792
+ </port>
793
+ </input>
794
+ </layer>
795
+ <layer id="54" name="Result_10205" type="Result" version="opset1" output_names="attention_mask">
796
+ <input>
797
+ <port id="0" precision="I64">
798
+ <dim>-1</dim>
799
+ <dim>-1</dim>
800
+ </port>
801
+ </input>
802
+ </layer>
803
+ </layers>
804
+ <edges>
805
+ <edge from-layer="0" from-port="0" to-layer="5" to-port="0" />
806
+ <edge from-layer="1" from-port="0" to-layer="46" to-port="0" />
807
+ <edge from-layer="2" from-port="0" to-layer="46" to-port="1" />
808
+ <edge from-layer="3" from-port="0" to-layer="46" to-port="2" />
809
+ <edge from-layer="4" from-port="0" to-layer="11" to-port="0" />
810
+ <edge from-layer="5" from-port="1" to-layer="6" to-port="0" />
811
+ <edge from-layer="5" from-port="3" to-layer="18" to-port="4" />
812
+ <edge from-layer="5" from-port="2" to-layer="18" to-port="3" />
813
+ <edge from-layer="5" from-port="1" to-layer="18" to-port="2" />
814
+ <edge from-layer="6" from-port="1" to-layer="9" to-port="0" />
815
+ <edge from-layer="7" from-port="0" to-layer="9" to-port="1" />
816
+ <edge from-layer="8" from-port="0" to-layer="9" to-port="2" />
817
+ <edge from-layer="9" from-port="3" to-layer="14" to-port="0" />
818
+ <edge from-layer="9" from-port="3" to-layer="11" to-port="1" />
819
+ <edge from-layer="10" from-port="0" to-layer="11" to-port="2" />
820
+ <edge from-layer="11" from-port="3" to-layer="18" to-port="0" />
821
+ <edge from-layer="12" from-port="0" to-layer="16" to-port="0" />
822
+ <edge from-layer="13" from-port="0" to-layer="14" to-port="1" />
823
+ <edge from-layer="14" from-port="2" to-layer="16" to-port="1" />
824
+ <edge from-layer="15" from-port="0" to-layer="16" to-port="2" />
825
+ <edge from-layer="16" from-port="3" to-layer="18" to-port="1" />
826
+ <edge from-layer="17" from-port="0" to-layer="18" to-port="5" />
827
+ <edge from-layer="18" from-port="6" to-layer="31" to-port="0" />
828
+ <edge from-layer="18" from-port="7" to-layer="31" to-port="1" />
829
+ <edge from-layer="18" from-port="11" to-layer="20" to-port="3" />
830
+ <edge from-layer="18" from-port="10" to-layer="20" to-port="2" />
831
+ <edge from-layer="18" from-port="9" to-layer="20" to-port="1" />
832
+ <edge from-layer="18" from-port="8" to-layer="20" to-port="0" />
833
+ <edge from-layer="19" from-port="0" to-layer="20" to-port="4" />
834
+ <edge from-layer="20" from-port="5" to-layer="23" to-port="0" />
835
+ <edge from-layer="20" from-port="6" to-layer="23" to-port="1" />
836
+ <edge from-layer="20" from-port="7" to-layer="23" to-port="2" />
837
+ <edge from-layer="20" from-port="8" to-layer="23" to-port="3" />
838
+ <edge from-layer="21" from-port="0" to-layer="23" to-port="4" />
839
+ <edge from-layer="22" from-port="0" to-layer="23" to-port="5" />
840
+ <edge from-layer="23" from-port="9" to-layer="26" to-port="3" />
841
+ <edge from-layer="23" from-port="8" to-layer="26" to-port="2" />
842
+ <edge from-layer="23" from-port="7" to-layer="26" to-port="1" />
843
+ <edge from-layer="23" from-port="6" to-layer="26" to-port="0" />
844
+ <edge from-layer="24" from-port="0" to-layer="26" to-port="4" />
845
+ <edge from-layer="25" from-port="0" to-layer="26" to-port="5" />
846
+ <edge from-layer="26" from-port="6" to-layer="29" to-port="0" />
847
+ <edge from-layer="26" from-port="7" to-layer="29" to-port="1" />
848
+ <edge from-layer="26" from-port="8" to-layer="29" to-port="2" />
849
+ <edge from-layer="26" from-port="9" to-layer="29" to-port="3" />
850
+ <edge from-layer="27" from-port="0" to-layer="29" to-port="4" />
851
+ <edge from-layer="28" from-port="0" to-layer="29" to-port="5" />
852
+ <edge from-layer="29" from-port="6" to-layer="31" to-port="2" />
853
+ <edge from-layer="29" from-port="7" to-layer="31" to-port="3" />
854
+ <edge from-layer="29" from-port="8" to-layer="31" to-port="4" />
855
+ <edge from-layer="29" from-port="9" to-layer="31" to-port="5" />
856
+ <edge from-layer="30" from-port="0" to-layer="31" to-port="6" />
857
+ <edge from-layer="31" from-port="7" to-layer="37" to-port="0" />
858
+ <edge from-layer="31" from-port="8" to-layer="37" to-port="1" />
859
+ <edge from-layer="31" from-port="9" to-layer="37" to-port="2" />
860
+ <edge from-layer="31" from-port="10" to-layer="37" to-port="3" />
861
+ <edge from-layer="31" from-port="11" to-layer="37" to-port="4" />
862
+ <edge from-layer="32" from-port="0" to-layer="37" to-port="5" />
863
+ <edge from-layer="33" from-port="0" to-layer="37" to-port="6" />
864
+ <edge from-layer="34" from-port="0" to-layer="37" to-port="7" />
865
+ <edge from-layer="35" from-port="0" to-layer="36" to-port="0" />
866
+ <edge from-layer="36" from-port="1" to-layer="37" to-port="8" />
867
+ <edge from-layer="37" from-port="9" to-layer="46" to-port="3" />
868
+ <edge from-layer="37" from-port="11" to-layer="46" to-port="5" />
869
+ <edge from-layer="37" from-port="9" to-layer="41" to-port="0" />
870
+ <edge from-layer="37" from-port="9" to-layer="38" to-port="1" />
871
+ <edge from-layer="37" from-port="10" to-layer="38" to-port="0" />
872
+ <edge from-layer="38" from-port="2" to-layer="40" to-port="0" />
873
+ <edge from-layer="39" from-port="0" to-layer="40" to-port="1" />
874
+ <edge from-layer="40" from-port="2" to-layer="41" to-port="1" />
875
+ <edge from-layer="41" from-port="2" to-layer="46" to-port="4" />
876
+ <edge from-layer="42" from-port="0" to-layer="46" to-port="6" />
877
+ <edge from-layer="43" from-port="0" to-layer="46" to-port="7" />
878
+ <edge from-layer="44" from-port="0" to-layer="46" to-port="8" />
879
+ <edge from-layer="45" from-port="0" to-layer="46" to-port="9" />
880
+ <edge from-layer="46" from-port="10" to-layer="47" to-port="1" />
881
+ <edge from-layer="46" from-port="12" to-layer="51" to-port="2" />
882
+ <edge from-layer="46" from-port="11" to-layer="51" to-port="1" />
883
+ <edge from-layer="46" from-port="10" to-layer="51" to-port="0" />
884
+ <edge from-layer="46" from-port="11" to-layer="47" to-port="0" />
885
+ <edge from-layer="47" from-port="2" to-layer="49" to-port="0" />
886
+ <edge from-layer="48" from-port="0" to-layer="49" to-port="1" />
887
+ <edge from-layer="49" from-port="2" to-layer="51" to-port="3" />
888
+ <edge from-layer="50" from-port="0" to-layer="51" to-port="4" />
889
+ <edge from-layer="51" from-port="6" to-layer="52" to-port="0" />
890
+ <edge from-layer="51" from-port="5" to-layer="55" to-port="0" />
891
+ <edge from-layer="52" from-port="1" to-layer="53" to-port="0" />
892
+ <edge from-layer="53" from-port="1" to-layer="54" to-port="0" />
893
+ <edge from-layer="55" from-port="1" to-layer="56" to-port="0" />
894
+ </edges>
895
+ <rt_info>
896
+ <add_attention_mask value="True" />
897
+ <add_prefix_space />
898
+ <add_special_tokens value="True" />
899
+ <bos_token_id value="0" />
900
+ <clean_up_tokenization_spaces />
901
+ <detokenizer_input_type value="i64" />
902
+ <eos_token_id value="2" />
903
+ <handle_special_tokens_with_re />
904
+ <max_length />
905
+ <number_of_inputs value="1" />
906
+ <openvino_tokenizers_version value="2025.1.0.0-523-710ddf14de8" />
907
+ <openvino_version value="2025.1.0-18503-6fec06580ab-releases/2025/1" />
908
+ <original_post_processor_template value="{&quot;type&quot;: &quot;TemplateProcessing&quot;, &quot;single&quot;: [{&quot;SpecialToken&quot;: {&quot;id&quot;: &quot;&lt;s>&quot;, &quot;type_id&quot;: 0}}, {&quot;Sequence&quot;: {&quot;id&quot;: &quot;A&quot;, &quot;type_id&quot;: 0}}, {&quot;SpecialToken&quot;: {&quot;id&quot;: &quot;&lt;/s>&quot;, &quot;type_id&quot;: 0}}], &quot;pair&quot;: [{&quot;SpecialToken&quot;: {&quot;id&quot;: &quot;&lt;s>&quot;, &quot;type_id&quot;: 0}}, {&quot;Sequence&quot;: {&quot;id&quot;: &quot;A&quot;, &quot;type_id&quot;: 0}}, {&quot;SpecialToken&quot;: {&quot;id&quot;: &quot;&lt;/s>&quot;, &quot;type_id&quot;: 0}}, {&quot;SpecialToken&quot;: {&quot;id&quot;: &quot;&lt;/s>&quot;, &quot;type_id&quot;: 0}}, {&quot;Sequence&quot;: {&quot;id&quot;: &quot;B&quot;, &quot;type_id&quot;: 0}}, {&quot;SpecialToken&quot;: {&quot;id&quot;: &quot;&lt;/s>&quot;, &quot;type_id&quot;: 0}}], &quot;special_tokens&quot;: {&quot;&lt;/s>&quot;: {&quot;id&quot;: &quot;&lt;/s>&quot;, &quot;ids&quot;: [2], &quot;tokens&quot;: [&quot;&lt;/s>&quot;]}, &quot;&lt;s>&quot;: {&quot;id&quot;: &quot;&lt;s>&quot;, &quot;ids&quot;: [0], &quot;tokens&quot;: [&quot;&lt;s>&quot;]}}}" />
909
+ <original_tokenizer_class value="&lt;class 'transformers.models.xlm_roberta.tokenization_xlm_roberta_fast.XLMRobertaTokenizerFast'>" />
910
+ <pad_token_id value="1" />
911
+ <processed_post_processor_template value="{&quot;single&quot;: {&quot;ids&quot;: [0, -1, 2], &quot;type_ids&quot;: [0, 0, 0]}, &quot;pair&quot;: {&quot;ids&quot;: [0, -1, 2, 2, -2, 2], &quot;type_ids&quot;: [0, 0, 0, 0, 0, 0]}}" />
912
+ <sentencepiece_version value="0.2.0" />
913
+ <skip_special_tokens value="True" />
914
+ <streaming_detokenizer value="False" />
915
+ <tokenizer_output_type value="i64" />
916
+ <tokenizers_version value="0.21.1" />
917
+ <transformers_version value="4.48.3" />
918
+ <use_max_padding value="False" />
919
+ <use_sentencepiece_backend value="False" />
920
+ <utf8_replace_mode value="replace" />
921
+ <with_detokenizer value="True" />
922
+ </rt_info>
923
+ </net>
sentencepiece.bpe.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
3
+ size 5069051
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": true,
26
+ "normalized": true,
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
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:14917dd757b81bc44d4af6b028367351702656670c1954e055dabdfcf21593cf
3
+ size 17082798
tokenizer_config.json ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "250001": {
36
+ "content": "<mask>",
37
+ "lstrip": true,
38
+ "normalized": true,
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
+ "extra_special_tokens": {},
49
+ "mask_token": "<mask>",
50
+ "model_max_length": 512,
51
+ "pad_token": "<pad>",
52
+ "sep_token": "</s>",
53
+ "sp_model_kwargs": {},
54
+ "tokenizer_class": "XLMRobertaTokenizer",
55
+ "unk_token": "<unk>"
56
+ }