rossieRuby commited on
Commit
8dc2bb8
·
verified ·
1 Parent(s): 32a7a8d

Add new SentenceTransformer model

Browse files
1_Pooling/config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "word_embedding_dimension": 768,
3
+ "pooling_mode_cls_token": false,
4
+ "pooling_mode_mean_tokens": true,
5
+ "pooling_mode_max_tokens": false,
6
+ "pooling_mode_mean_sqrt_len_tokens": false,
7
+ "pooling_mode_weightedmean_tokens": false,
8
+ "pooling_mode_lasttoken": false,
9
+ "include_prompt": true
10
+ }
README.md ADDED
@@ -0,0 +1,535 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ tags:
3
+ - sentence-transformers
4
+ - sentence-similarity
5
+ - feature-extraction
6
+ - generated_from_trainer
7
+ - dataset_size:10000
8
+ - loss:CosineSimilarityLoss
9
+ - dataset_size:4996
10
+ base_model: rossieRuby/nyayadrishti-bert
11
+ widget:
12
+ - source_sentence: Under what article of the Constitution of India was the petition
13
+ filed?
14
+ sentences:
15
+ - ition No. 13009 of 1983. (Under article 32 of the Constitution of India) Gopal
16
+ Subra aniam (A.C) for the Petitioner Dr. Y.S. Chitale, A.V. Rangam and Mrs. Sarla
17
+ Chandra for the Respondent.
18
+ - The High Court held that (1) the appellant was in possession of the lands before
19
+ the passing of the decree; (2) the suits had not abated and the Board of Revenue
20
+ had no jurisdiction to set aside the proceedings, in the suits ' and (3) the applications
21
+ for restitution were not maintainable. The High Court, however, held that (1)
22
+ appeals against the orders for restitution lay to the revenue court, (2) the civil
23
+ court had no jurisdiction to entertain the appeals and (3) the respondent was
24
+ not estopped from raising the contention. Accordingly on March 26, 1965 the High
25
+ Court allowed the second appeals, set aside the order of the Additional Civil
26
+ Judge and returned the memoranda of appeals for presentation to the proper court.
27
+ - The question whether respondent No. 4 Ram Nath Singh and his son Vijendra Singh
28
+ are guilty of contumacious and wilful disregard of this Court 's order must depend
29
+ on the precise meaning of the words `status quo as in the High Court '. There
30
+ is not much of a controversy as to the scope and effect of the status quo order
31
+ passed by this Court. Shri L.N. Sinha, learned counsel appearing for the appellant
32
+ submitted that the words 'status quo as in the High Court ' mean status quo as
33
+ prevailing between the parties when the matter was pending in the High Court and
34
+ not after the High Court had passed the impugned judgment and disposed of the
35
+ writ petition.
36
+ - source_sentence: Who delivered the order of the Court?
37
+ sentences:
38
+ - 'N: Review Petition (Criminal) Nos. 24 1 242 of 1989. IN Criminal Appeal Nos.
39
+ 544 545 of 1986. Mahabir Singh for the Petitioner. A.N. Mulla, S.B. Upadhyay for
40
+ the Respondents. The Order of the Court was delivered by RAY, J.It is very unfortunate
41
+ that a controversy has arisen following the judgment sought to be reviewed in
42
+ Criminal Appeal Nos. 544 45 of 1986 rendered by this Bench on 31st January 1989
43
+ whereby this Court while confirming the conviction of both the respondents/accused
44
+ reduced the sentence of imprisonment in respect of each of the respond ents from
45
+ 10 years to 5 years by invoking the proviso to Section 376(2) of the Indian Penal
46
+ Code observing "the peculiar facts and 498 circumstances of this case coupled
47
+ with the conduct of the victim girl, in our view, do not call for the minimum
48
+ sen tence as prescribed under Section 376(2). " The State of Haryana has filed
49
+ the above petitions seeking review of the judgment and to "pass such other or
50
+ further order(s) as may be necessary in the circumstances of the case. " At the
51
+ outset, we may examine the scope of review of a judgment in a criminal case already
52
+ pronounced by this Court. Article 137 of the Constitution of India gives the power
53
+ to the Supreme Court to review its judgment but such special power is exercisable
54
+ in accordance with, and subject to, the rules of this Court made under Article
55
+ 145 of the Constitution of India. Order XL, Rule 1 of the Supreme Court Rules
56
+ provides: "The Court may review its judgment or order but no application for review
57
+ will be enter tained in a civil proceeding except on the ground mentioned in Order
58
+ XLVII, Rule 1 of the Code and in a criminal proceeding except on the ground of
59
+ an error on the face of the record." This Court in a series of decisions has examined
60
+ the scope of review in criminal cases after the judgment pro nounced or order
61
+ made. Though we are not citing all those decisions, we may refer to a In the case
62
+ of P.N. Eswara Iyer and Ors vs Registrar, Supreme Court of India, ; the Constitution
63
+ Bench of this Court while considering the rule observed thus: "The rule (Order
64
+ XL, Rule 1), on its face affords a wider set of grounds for review for orders
65
+ in civil proceedings, but limits the ground vis a vis criminal proceedings to
66
+ ''errors apparent on the face of the record. ''. " See also Sow Chandra Kanta
67
+ & Anr. vs Sheik Habib, ; and Sheonandan Paswan vs State of Bihar and Or ders,
68
+ In our considered view, when the present matter is examined in the light of the
69
+ decisions referred to above, we find no error apparent on the face of the record
70
+ necessitat ing review of the judgment and as such these review peti tions are
71
+ liable to be dismissed. 499 We have heard the arguments of the learned senior
72
+ coun sel, Mr. Rajinder Sachar who though initially started his arguments on behalf
73
+ of the People ''s Union for Civil Liber ties ultimately advanced his arguments
74
+ on behalf of the State in these review petitions on the representation made by
75
+ Mr. Mahabir Singh, the learned counsel for the State. Mr. R.K.P. Shankar Dass
76
+ who advanced his arguments on behalf of Mahila Sanyukt Morcha stated that his
77
+ arguments may also be treated as supplemental to the arguments of Mr. Rajinder
78
+ Sachar. Mr. Mulla, the learned senior counsel appeared on behalf of the respondents.
79
+ Although we have found that the Review Petitions are liable to be dismissed on
80
+ the ground that there is no error apparent on the face of the record, we, however,
81
+ in view of the elaborate submissions made by the various learned coun sel appearing
82
+ before us, would like to make the following observations. The facts of the case
83
+ are briefly stated in the Criminal Appeals and, therefore, it is not necessary
84
+ to restate the same. Suffice to say that during the course of the '' hearing on
85
+ the appeals on behalf of the respondents/accused, it has been urged by the learned
86
+ defence counsel that the victim Suman Rani was a woman of questionable character
87
+ and easy virtue with lewd and lascivious behaviour and as such her version is
88
+ not worthy of acceptance. After considerable debate on the merits of the case,
89
+ the argument was confined only with regard to the quantum of sentence. after meticu
90
+ lously examining the entire matter, this Court came to the conclusion that the
91
+ proviso to Section 376(2) I.P.C. could be invoked having regard to the peculiar
92
+ facts and circum stances of the case coupled with the conduct of the victim and
93
+ the mandatory sentence provided under the penal provi sion is not called for.
94
+ At this juncture, we would like to point put that the very confirmation of the
95
+ conviction accepting the sole testimony of the victim Suman Rani rejecting the
96
+ arguments of the defence counsel is itself a clear indication that this Court
97
+ was of the view that the character or reputation of the victim has no bearing
98
+ or relevance either in the matter of adjudging the guilt of the accused or imposing
99
+ punishment under Section 376 I.P.C. We would like to state with all emphasis that
100
+ such factors are wholly alien to the very scope and object of Section 376 and
101
+ can never serve either as mitigating or extenuating circumstances for impos ing
102
+ the sub minimum sentence with the aid of the proviso to Section 376(2) of the
103
+ I.P.C. In fact, we have expressed our 500 views in the judgment itself '' stating
104
+ "No doubt an offence of this nature has to be viewed very seriously and has to
105
+ be dealt with condign punishment. " We have neither characterised the victim,
106
+ Suman Rani as a woman of questionable character and easy virtue nor made any reference
107
+ to her character or reputation in any part of our judgment but used the expression
108
+ "conduct" in the lexi graphical meaning for the limited purpose of showing as
109
+ to how Suman Rani had behaved or conducted herself in not telling any one for
110
+ about 5 days about the sexual assault perpetrated on her till she was examined
111
+ on 28.3.1984 by the Sub Inspector of Police (PW 20) in connection with the complaint
112
+ given by Ram Lal (PW 14) on 22.3.1984 against Ravi Shanker. In this connection,
113
+ we make it further clear that we have not used the word ''conduct '' with reference
114
+ to the character or reputation of the victim Suman Rani. Before parting with this
115
+ matter, we would like to ex press that this Court is second to none in upholding
116
+ the decency and dignity of woman hood and we have not expressed any view in our
117
+ judgment that character, reputation or status of a raped victim is a relevant
118
+ factor for considera tion by the Court while awarding the sentence to a rapist.
119
+ With the above observations, we dismiss the Review Peti tions. G.N. Petitions
120
+ dismissed.'
121
+ - N.A. Palkhivala, section N. Andley, and J. B. Dadachanji, for the appellant. K.N.
122
+ Rajagopal Sastri and D. Gupta, for the respondent.
123
+ - After considering the submissions advanced by learned counsels for the parties
124
+ we are constrained to hold that Dewan Bishen Dass predecessor of the appellants
125
+ was a Wasidar and the lands in question were wasidari land leased out to him for
126
+ the purpose of constructing buildings. This lease is governed by Ailan No. 10
127
+ as well as by the Lands Grants Act 1960. We affirm the findings of the High Court
128
+ which held the land as Wasidari land. The land was transferred by Purnesh Chandra
129
+ and others, legal representatives of the original lessee Dewan Bishen Dass, in
130
+ favour of the appellants in contravention of the provisions of section 12(A) of
131
+ the Jammu and Kashmir Land Grants Act, 1960. The impugned notice under section
132
+ 4(1) of the Jammu and Kashmir (Public Premises Eviction of Unauthorised Occupants)
133
+ Act is in accordance with law and as such it is valid. Under the said Act as well
134
+ as the rule the appellants are entitled to get compensation of the buildings and
135
+ structures as well as of the improvements made on the land even though they are
136
+ not entitled to get compensation in respect of value of the land.
137
+ - source_sentence: What does Note (1) of the General Information Table state about
138
+ the cost?
139
+ sentences:
140
+ - Among the very few exceptions to this uninterrupted flow of the court process
141
+ is section 494, Cr. Even here, the Public Prosecutor not any executive authority
142
+ is entrusted by the Code with a limited power to withdraw from a prosecution,
143
+ with the (1) Chandler vs Judicial Council of the Tenth Circuit of the U.S. , 1970.
144
+ 48 court 's consent whereupon the case comes to a close. What the law has ignited,
145
+ the law alone shall extinguish.
146
+ - Serial "Tamas" depicts the Hindu Muslim tension and sikhmuslim tension before
147
+ the partition of India. It further shows how the killings and looting took place
148
+ between these communities before the pre independence at Lahore. "Tamas" is based
149
+ on a book written by Sree Bhisham Sahni. It depicts the period prior to partition
150
+ and how communal violence was generated by fundamentalists and extremists in both
151
+ communities and how innocent persons were duped into serving the ulterior purpose
152
+ of fundamentalists and communities of both sides and how an innocent boy is seduced
153
+ to violence resulting in his harming both communities. It further shows how extremist
154
+ elements in both communities infused tension and hatred for their own ends. That
155
+ is how the two learned Judges of the High Court of Bombay mentioned hereinbefore
156
+ have viewed it. They have also seen that realisation ultimately dawns as to the
157
+ futility of it all and finally how inherent goodness in human mind triumphs and
158
+ both communities learn to live in amity. They saw that the people learnt this
159
+ lesson in a hard way. This is the opinion expressed by two experienced Judges
160
+ of the High Court after viewing the serial.
161
+ - 'General Information Table Note: (1) The cost shown in the column 4 is only estimated
162
+ cost. It will increase or decrease according to the rise or fall in the price
163
+ at the time of completion of the property. Note: (2) The data given in the above
164
+ mentioned table can be amended as felt neces sary. The last paragraph of the letter
165
+ dated 19/20.1.84 (Annexure ''D '') reads thus: "If you want to buy the house on
166
+ the above price/instalment then you must send by 28.1.1984 your written acceptance
167
+ on the annexed proforma to the Registration Section of this office."'
168
+ - source_sentence: Were the donees of the life estate adults or minors at the time
169
+ of the will?
170
+ sentences:
171
+ - The donees of the life estate were minors at the date of the will and there was
172
+ no knowing when they would get married and how many children each would have.
173
+ It would therefore be reason able to expect that the testatrix would so arrange
174
+ her affairs that each of the foster children should get half of the income of
175
+ the property for life and that their children should succeed to the respective
176
+ interest of their parents. It is hardly likely that the testatrix would know the
177
+ difference between joint tenants and tenants in common and she would naturally
178
+ be eager to treat the foster children as her own children so that the heirs of
179
+ the foster children would take share and share alike the properties be in divided
180
+ per stirpes among them.
181
+ - 'The learned Solicitor General who appeared for the Commissioner of Income tax,
182
+ West Bengal, combated the contentions raised by ''Mr. Chatterjee on a two fold
183
+ ground: (1) In the first instance, without questioning the jurisdiction of this
184
+ Court to grant special leave against an order of an Income tax Tribunal, he argued
185
+ that such leave should not be granted when remedies provided by the Income tax
186
+ Act itself were available for correcting errors of the Tribunal, and had been
187
+ taken but without success. It was said that the power conferred on this Court
188
+ by article 136 of the Constitution being an extraordinary power, its exercise
189
+ should be limited to cases of patent and glaring errors of procedure, or where
190
+ there has been a failure of justice because of the violation of the rules of natural
191
+ justice or like causes but that this discretionary power should not be exercised
192
+ for the purpose of reviewing findings of fact when the law dealing with the subject
193
+ has declared those findings as final and conclusive.'
194
+ - According ly, the Appellate Tribunal did not specifically examine the alternative
195
+ position whether the process of calendering of the type and kind adopted by the
196
+ appellant really shared the common element or characteristic possessed by the
197
+ other processes specifically enumerated. Therefore, if it is to be held that the
198
+ expression "any other process" in Sec. 2(f)(v) must be understood and construed
199
+ ejus dem generis, then the question whether the "process" of calendering employed
200
+ in the present case belongs to the same genus as the processes envisaged in the
201
+ preceding expressions in the section would have to be examined afresh.
202
+ - source_sentence: What was the basis of the loan advances to the assessee?
203
+ sentences:
204
+ - 'It was, however, stated by the Tribunal that taking into account the correspondence
205
+ and the documents referred to earlier it was satisfied with the assessee ''s case
206
+ that the transfer of shares to London at issue price or at par was throughout
207
+ the basis of the advances of loans to the assessee. It is necessary to reproduce
208
+ paragraph 31 of the order of the Tribunal : "In October 1953, there was no mention
209
+ of any capital gains tax being revived. At that time the asses 144 see could not
210
+ have had any idea of avoiding or reducing any liability to capital gains tax.
211
+ The learned counsel for the department laid some emphasis on the fact that there
212
+ was no enforceable arrangement. The question as to whether there was an enforceable
213
+ arrangement or not is not really material. What we have to find out is whether
214
+ the object in putting through these transactions of taking over the shares at
215
+ par or at issue price was one of avoidance or reduction of liability to capital
216
+ gains tax. That object does not get established by the mere absence of an enforceable
217
+ arrangement. Having regard to the assessee being the subsidiary of I.C.I., there
218
+ is nothing surprising about the arrangement not being so formal or not being put
219
+ through after complying with all the necessary legal formalities. The absence
220
+ of formal agreement is thus understandable in this context and cannot by itself
221
+ suggest anything in favour of the department. Businessmen are not always motivated
222
+ by legalistic considerations. Even taking that the arrangement was only binding
223
+ morally and not legally, still so long as the assessee wanted to fulfil a moral
224
+ obligation and had not the capital gains tax in mind, it cannot be said that the
225
+ transaction was entered into with the object of avoidance or reduction of liability
226
+ to capital gains tax".'
227
+ - A. V. Viswanatha Sastri, R. K. Garg, M. K. Ramamurthi, D. P. Singh and section
228
+ C. Agarwala, for the appellants. M. C. Setalvad, Attorney General for India. B.
229
+ P. Rajgarhia and K. K. Sinha, for the respondents.
230
+ - The requirement of obtaining opinion E; of the Advisory Board is an additional
231
+ safeguard over and above the safeguard afforded to the, detenu of Making a representation
232
+ against the order of detention. The opinion of the Advisory Board even if given
233
+ after consideration of the representation is no substitute for the consideration
234
+ of the representation by the detaining authority. This Court pointed out in Khairul
235
+ Haque vs The State of West Bengal(1).
236
+ pipeline_tag: sentence-similarity
237
+ library_name: sentence-transformers
238
+ ---
239
+
240
+ # SentenceTransformer based on rossieRuby/nyayadrishti-bert
241
+
242
+ This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [rossieRuby/nyayadrishti-bert](https://huggingface.co/rossieRuby/nyayadrishti-bert). It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
243
+
244
+ ## Model Details
245
+
246
+ ### Model Description
247
+ - **Model Type:** Sentence Transformer
248
+ - **Base model:** [rossieRuby/nyayadrishti-bert](https://huggingface.co/rossieRuby/nyayadrishti-bert) <!-- at revision 1fb57b531f242c78a01a6abd89b545f078f5dffa -->
249
+ - **Maximum Sequence Length:** 512 tokens
250
+ - **Output Dimensionality:** 768 dimensions
251
+ - **Similarity Function:** Cosine Similarity
252
+ <!-- - **Training Dataset:** Unknown -->
253
+ <!-- - **Language:** Unknown -->
254
+ <!-- - **License:** Unknown -->
255
+
256
+ ### Model Sources
257
+
258
+ - **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
259
+ - **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
260
+ - **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
261
+
262
+ ### Full Model Architecture
263
+
264
+ ```
265
+ SentenceTransformer(
266
+ (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel
267
+ (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
268
+ )
269
+ ```
270
+
271
+ ## Usage
272
+
273
+ ### Direct Usage (Sentence Transformers)
274
+
275
+ First install the Sentence Transformers library:
276
+
277
+ ```bash
278
+ pip install -U sentence-transformers
279
+ ```
280
+
281
+ Then you can load this model and run inference.
282
+ ```python
283
+ from sentence_transformers import SentenceTransformer
284
+
285
+ # Download from the 🤗 Hub
286
+ model = SentenceTransformer("rossieRuby/nyayadrishti-bert-v2")
287
+ # Run inference
288
+ sentences = [
289
+ 'What was the basis of the loan advances to the assessee?',
290
+ 'It was, however, stated by the Tribunal that taking into account the correspondence and the documents referred to earlier it was satisfied with the assessee \'s case that the transfer of shares to London at issue price or at par was throughout the basis of the advances of loans to the assessee. It is necessary to reproduce paragraph 31 of the order of the Tribunal : "In October 1953, there was no mention of any capital gains tax being revived. At that time the asses 144 see could not have had any idea of avoiding or reducing any liability to capital gains tax. The learned counsel for the department laid some emphasis on the fact that there was no enforceable arrangement. The question as to whether there was an enforceable arrangement or not is not really material. What we have to find out is whether the object in putting through these transactions of taking over the shares at par or at issue price was one of avoidance or reduction of liability to capital gains tax. That object does not get established by the mere absence of an enforceable arrangement. Having regard to the assessee being the subsidiary of I.C.I., there is nothing surprising about the arrangement not being so formal or not being put through after complying with all the necessary legal formalities. The absence of formal agreement is thus understandable in this context and cannot by itself suggest anything in favour of the department. Businessmen are not always motivated by legalistic considerations. Even taking that the arrangement was only binding morally and not legally, still so long as the assessee wanted to fulfil a moral obligation and had not the capital gains tax in mind, it cannot be said that the transaction was entered into with the object of avoidance or reduction of liability to capital gains tax".',
291
+ 'The requirement of obtaining opinion E; of the Advisory Board is an additional safeguard over and above the safeguard afforded to the, detenu of Making a representation against the order of detention. The opinion of the Advisory Board even if given after consideration of the representation is no substitute for the consideration of the representation by the detaining authority. This Court pointed out in Khairul Haque vs The State of West Bengal(1).',
292
+ ]
293
+ embeddings = model.encode(sentences)
294
+ print(embeddings.shape)
295
+ # [3, 768]
296
+
297
+ # Get the similarity scores for the embeddings
298
+ similarities = model.similarity(embeddings, embeddings)
299
+ print(similarities.shape)
300
+ # [3, 3]
301
+ ```
302
+
303
+ <!--
304
+ ### Direct Usage (Transformers)
305
+
306
+ <details><summary>Click to see the direct usage in Transformers</summary>
307
+
308
+ </details>
309
+ -->
310
+
311
+ <!--
312
+ ### Downstream Usage (Sentence Transformers)
313
+
314
+ You can finetune this model on your own dataset.
315
+
316
+ <details><summary>Click to expand</summary>
317
+
318
+ </details>
319
+ -->
320
+
321
+ <!--
322
+ ### Out-of-Scope Use
323
+
324
+ *List how the model may foreseeably be misused and address what users ought not to do with the model.*
325
+ -->
326
+
327
+ <!--
328
+ ## Bias, Risks and Limitations
329
+
330
+ *What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
331
+ -->
332
+
333
+ <!--
334
+ ### Recommendations
335
+
336
+ *What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
337
+ -->
338
+
339
+ ## Training Details
340
+
341
+ ### Training Dataset
342
+
343
+ #### Unnamed Dataset
344
+
345
+ * Size: 4,996 training samples
346
+ * Columns: <code>sentence_0</code>, <code>sentence_1</code>, and <code>label</code>
347
+ * Approximate statistics based on the first 1000 samples:
348
+ | | sentence_0 | sentence_1 | label |
349
+ |:--------|:----------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------|:--------------------------------------------------------------|
350
+ | type | string | string | float |
351
+ | details | <ul><li>min: 6 tokens</li><li>mean: 15.18 tokens</li><li>max: 38 tokens</li></ul> | <ul><li>min: 10 tokens</li><li>mean: 149.23 tokens</li><li>max: 512 tokens</li></ul> | <ul><li>min: 1.0</li><li>mean: 1.0</li><li>max: 1.0</li></ul> |
352
+ * Samples:
353
+ | sentence_0 | sentence_1 | label |
354
+ |:--------------------------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:-----------------|
355
+ | <code>Who died first among the testators?</code> | <code>of the three testators, Theyi Amma died first the exact date of her death does not appear and is not very material and Kunhan Kaimal died thereafter sometime in 1930. It is the case of Kesavan Kaimal that in the events which had happened, he had become entitled by survivorship to all the properties disposed of by the will, including those of Kunhan Kaimal, and on this footing he conveyed on October 14,1938, seven items of properties, of which three belonged to Kunhan Kaimal, to one Sankarankutti Kaimal and on October 16, 1944, another three items of properties which belonged to Kunhan Kaimal, to Kalyani and Vijayan.</code> | <code>1.0</code> |
356
+ | <code>Why was the annual rental value of Rs. 7200 not considered a proper method of computation?</code> | <code>213 the claimants ' The annual rental value of the land acquired, namely, Rs. 7 200/ will also not furnish a proper method of computation because that was a rent fixed in 1944 when that land was not of such great value as it had acquired at the time when sec. 4 notification was issued. A perusal of the correspondence between the owners of the land and the Deputy Commissioner of Ranchi would show that the land owners had given it at consessional rate to ' the Military authorities having regard to the purpose for which it was being put to use.</code> | <code>1.0</code> |
357
+ | <code>What is the date of the Income tax Appellate Tribunal's judgment and order?</code> | <code>Appeals Nos. 776 and 777 of 1957. Appeals by special leave from the judgment and order dated September 25, 1956, of the Bombay High Court in Income tax Application No. 48 of 1956; and from the judgment and order dated March 17,1954, of the Income tax Appellate Tribunal, Bombay, in E.P.T.A. Nos. 757, 903 and 944 of 1948 49, respectively.</code> | <code>1.0</code> |
358
+ * Loss: [<code>CosineSimilarityLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cosinesimilarityloss) with these parameters:
359
+ ```json
360
+ {
361
+ "loss_fct": "torch.nn.modules.loss.MSELoss"
362
+ }
363
+ ```
364
+
365
+ ### Training Hyperparameters
366
+ #### Non-Default Hyperparameters
367
+
368
+ - `per_device_train_batch_size`: 32
369
+ - `per_device_eval_batch_size`: 32
370
+ - `num_train_epochs`: 1
371
+ - `multi_dataset_batch_sampler`: round_robin
372
+
373
+ #### All Hyperparameters
374
+ <details><summary>Click to expand</summary>
375
+
376
+ - `overwrite_output_dir`: False
377
+ - `do_predict`: False
378
+ - `eval_strategy`: no
379
+ - `prediction_loss_only`: True
380
+ - `per_device_train_batch_size`: 32
381
+ - `per_device_eval_batch_size`: 32
382
+ - `per_gpu_train_batch_size`: None
383
+ - `per_gpu_eval_batch_size`: None
384
+ - `gradient_accumulation_steps`: 1
385
+ - `eval_accumulation_steps`: None
386
+ - `torch_empty_cache_steps`: None
387
+ - `learning_rate`: 5e-05
388
+ - `weight_decay`: 0.0
389
+ - `adam_beta1`: 0.9
390
+ - `adam_beta2`: 0.999
391
+ - `adam_epsilon`: 1e-08
392
+ - `max_grad_norm`: 1
393
+ - `num_train_epochs`: 1
394
+ - `max_steps`: -1
395
+ - `lr_scheduler_type`: linear
396
+ - `lr_scheduler_kwargs`: {}
397
+ - `warmup_ratio`: 0.0
398
+ - `warmup_steps`: 0
399
+ - `log_level`: passive
400
+ - `log_level_replica`: warning
401
+ - `log_on_each_node`: True
402
+ - `logging_nan_inf_filter`: True
403
+ - `save_safetensors`: True
404
+ - `save_on_each_node`: False
405
+ - `save_only_model`: False
406
+ - `restore_callback_states_from_checkpoint`: False
407
+ - `no_cuda`: False
408
+ - `use_cpu`: False
409
+ - `use_mps_device`: False
410
+ - `seed`: 42
411
+ - `data_seed`: None
412
+ - `jit_mode_eval`: False
413
+ - `use_ipex`: False
414
+ - `bf16`: False
415
+ - `fp16`: False
416
+ - `fp16_opt_level`: O1
417
+ - `half_precision_backend`: auto
418
+ - `bf16_full_eval`: False
419
+ - `fp16_full_eval`: False
420
+ - `tf32`: None
421
+ - `local_rank`: 0
422
+ - `ddp_backend`: None
423
+ - `tpu_num_cores`: None
424
+ - `tpu_metrics_debug`: False
425
+ - `debug`: []
426
+ - `dataloader_drop_last`: False
427
+ - `dataloader_num_workers`: 0
428
+ - `dataloader_prefetch_factor`: None
429
+ - `past_index`: -1
430
+ - `disable_tqdm`: False
431
+ - `remove_unused_columns`: True
432
+ - `label_names`: None
433
+ - `load_best_model_at_end`: False
434
+ - `ignore_data_skip`: False
435
+ - `fsdp`: []
436
+ - `fsdp_min_num_params`: 0
437
+ - `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
438
+ - `tp_size`: 0
439
+ - `fsdp_transformer_layer_cls_to_wrap`: None
440
+ - `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
441
+ - `deepspeed`: None
442
+ - `label_smoothing_factor`: 0.0
443
+ - `optim`: adamw_torch
444
+ - `optim_args`: None
445
+ - `adafactor`: False
446
+ - `group_by_length`: False
447
+ - `length_column_name`: length
448
+ - `ddp_find_unused_parameters`: None
449
+ - `ddp_bucket_cap_mb`: None
450
+ - `ddp_broadcast_buffers`: False
451
+ - `dataloader_pin_memory`: True
452
+ - `dataloader_persistent_workers`: False
453
+ - `skip_memory_metrics`: True
454
+ - `use_legacy_prediction_loop`: False
455
+ - `push_to_hub`: False
456
+ - `resume_from_checkpoint`: None
457
+ - `hub_model_id`: None
458
+ - `hub_strategy`: every_save
459
+ - `hub_private_repo`: None
460
+ - `hub_always_push`: False
461
+ - `gradient_checkpointing`: False
462
+ - `gradient_checkpointing_kwargs`: None
463
+ - `include_inputs_for_metrics`: False
464
+ - `include_for_metrics`: []
465
+ - `eval_do_concat_batches`: True
466
+ - `fp16_backend`: auto
467
+ - `push_to_hub_model_id`: None
468
+ - `push_to_hub_organization`: None
469
+ - `mp_parameters`:
470
+ - `auto_find_batch_size`: False
471
+ - `full_determinism`: False
472
+ - `torchdynamo`: None
473
+ - `ray_scope`: last
474
+ - `ddp_timeout`: 1800
475
+ - `torch_compile`: False
476
+ - `torch_compile_backend`: None
477
+ - `torch_compile_mode`: None
478
+ - `include_tokens_per_second`: False
479
+ - `include_num_input_tokens_seen`: False
480
+ - `neftune_noise_alpha`: None
481
+ - `optim_target_modules`: None
482
+ - `batch_eval_metrics`: False
483
+ - `eval_on_start`: False
484
+ - `use_liger_kernel`: False
485
+ - `eval_use_gather_object`: False
486
+ - `average_tokens_across_devices`: False
487
+ - `prompts`: None
488
+ - `batch_sampler`: batch_sampler
489
+ - `multi_dataset_batch_sampler`: round_robin
490
+
491
+ </details>
492
+
493
+ ### Framework Versions
494
+ - Python: 3.11.12
495
+ - Sentence Transformers: 4.1.0
496
+ - Transformers: 4.51.3
497
+ - PyTorch: 2.6.0+cu124
498
+ - Accelerate: 1.6.0
499
+ - Datasets: 2.14.4
500
+ - Tokenizers: 0.21.1
501
+
502
+ ## Citation
503
+
504
+ ### BibTeX
505
+
506
+ #### Sentence Transformers
507
+ ```bibtex
508
+ @inproceedings{reimers-2019-sentence-bert,
509
+ title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
510
+ author = "Reimers, Nils and Gurevych, Iryna",
511
+ booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
512
+ month = "11",
513
+ year = "2019",
514
+ publisher = "Association for Computational Linguistics",
515
+ url = "https://arxiv.org/abs/1908.10084",
516
+ }
517
+ ```
518
+
519
+ <!--
520
+ ## Glossary
521
+
522
+ *Clearly define terms in order to be accessible across audiences.*
523
+ -->
524
+
525
+ <!--
526
+ ## Model Card Authors
527
+
528
+ *Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
529
+ -->
530
+
531
+ <!--
532
+ ## Model Card Contact
533
+
534
+ *Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
535
+ -->
config.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "BertModel"
4
+ ],
5
+ "attention_probs_dropout_prob": 0.1,
6
+ "bos_token_id": 0,
7
+ "classifier_dropout": null,
8
+ "eos_token_ids": 0,
9
+ "hidden_act": "gelu",
10
+ "hidden_dropout_prob": 0.1,
11
+ "hidden_size": 768,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 3072,
14
+ "layer_norm_eps": 1e-12,
15
+ "max_position_embeddings": 512,
16
+ "model_type": "bert",
17
+ "num_attention_heads": 12,
18
+ "num_hidden_layers": 12,
19
+ "output_past": true,
20
+ "pad_token_id": 0,
21
+ "position_embedding_type": "absolute",
22
+ "torch_dtype": "float32",
23
+ "transformers_version": "4.51.3",
24
+ "type_vocab_size": 2,
25
+ "use_cache": true,
26
+ "vocab_size": 30522
27
+ }
config_sentence_transformers.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "__version__": {
3
+ "sentence_transformers": "4.1.0",
4
+ "transformers": "4.51.3",
5
+ "pytorch": "2.6.0+cu124"
6
+ },
7
+ "prompts": {},
8
+ "default_prompt_name": null,
9
+ "similarity_fn_name": "cosine"
10
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e83ad1f112c1357aade400596f2a8f86a7d5ddc6078e33d03d93c96838313242
3
+ size 437951328
modules.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "idx": 0,
4
+ "name": "0",
5
+ "path": "",
6
+ "type": "sentence_transformers.models.Transformer"
7
+ },
8
+ {
9
+ "idx": 1,
10
+ "name": "1",
11
+ "path": "1_Pooling",
12
+ "type": "sentence_transformers.models.Pooling"
13
+ }
14
+ ]
sentence_bert_config.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "max_seq_length": 512,
3
+ "do_lower_case": false
4
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": {
3
+ "content": "[CLS]",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "mask_token": {
10
+ "content": "[MASK]",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "[PAD]",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "sep_token": {
24
+ "content": "[SEP]",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "unk_token": {
31
+ "content": "[UNK]",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ }
37
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,65 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "100": {
12
+ "content": "[UNK]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "101": {
20
+ "content": "[CLS]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "102": {
28
+ "content": "[SEP]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "103": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "clean_up_tokenization_spaces": true,
45
+ "cls_token": "[CLS]",
46
+ "do_basic_tokenize": true,
47
+ "do_lower_case": true,
48
+ "extra_special_tokens": {},
49
+ "mask_token": "[MASK]",
50
+ "max_length": 512,
51
+ "model_max_length": 512,
52
+ "never_split": null,
53
+ "pad_to_multiple_of": null,
54
+ "pad_token": "[PAD]",
55
+ "pad_token_type_id": 0,
56
+ "padding_side": "right",
57
+ "sep_token": "[SEP]",
58
+ "stride": 0,
59
+ "strip_accents": null,
60
+ "tokenize_chinese_chars": true,
61
+ "tokenizer_class": "BertTokenizer",
62
+ "truncation_side": "right",
63
+ "truncation_strategy": "longest_first",
64
+ "unk_token": "[UNK]"
65
+ }
vocab.txt ADDED
The diff for this file is too large to render. See raw diff