SamagraDataGov commited on
Commit
4c3bdaf
1 Parent(s): da542e7

pytorch_model.bin upload/update

Browse files
1_Pooling/config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "word_embedding_dimension": 384,
3
+ "pooling_mode_cls_token": true,
4
+ "pooling_mode_mean_tokens": false,
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 +1,876 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language: []
3
+ library_name: sentence-transformers
4
+ tags:
5
+ - sentence-transformers
6
+ - sentence-similarity
7
+ - feature-extraction
8
+ - generated_from_trainer
9
+ - dataset_size:900
10
+ - loss:GISTEmbedLoss
11
+ base_model: BAAI/bge-small-en-v1.5
12
+ datasets: []
13
+ metrics:
14
+ - cosine_accuracy@1
15
+ - cosine_accuracy@5
16
+ - cosine_accuracy@10
17
+ - cosine_precision@1
18
+ - cosine_precision@5
19
+ - cosine_precision@10
20
+ - cosine_recall@1
21
+ - cosine_recall@5
22
+ - cosine_recall@10
23
+ - cosine_ndcg@5
24
+ - cosine_ndcg@10
25
+ - cosine_ndcg@100
26
+ - cosine_mrr@5
27
+ - cosine_mrr@10
28
+ - cosine_mrr@100
29
+ - cosine_map@100
30
+ - dot_accuracy@1
31
+ - dot_accuracy@5
32
+ - dot_accuracy@10
33
+ - dot_precision@1
34
+ - dot_precision@5
35
+ - dot_precision@10
36
+ - dot_recall@1
37
+ - dot_recall@5
38
+ - dot_recall@10
39
+ - dot_ndcg@5
40
+ - dot_ndcg@10
41
+ - dot_ndcg@100
42
+ - dot_mrr@5
43
+ - dot_mrr@10
44
+ - dot_mrr@100
45
+ - dot_map@100
46
+ widget:
47
+ - source_sentence: What is the purpose of providing Loss Cost information to DAC &
48
+ FW/Concerned States?
49
+ sentences:
50
+ - '''The premium in respect of affected IUs including subsidy to enable settlement
51
+ of claims arising due to above events in respect of all such beneficiaries
52
+ shall be adjusted from the fund already available with ICs as advance upfront
53
+ subsidy (First Instalment) to facilitate compliance of Section 64 B of Insurance
54
+ Act/Regulation of IRDAI 13.2.5 All admissible claims based on Yield data/Post
55
+ harvest losses will be settled on receipt of second installment of Govt. subsidy
56
+ to be paid on the basis of tentative business statistics generated on the portal
57
+ after 15 days of period specified for auto approval of applications on the Portal.
58
+ The remaining Govt. subsidy, if any will be paid after reconciliation of all
59
+ business statistics for the season on portal. 13.2.6 All empanelled insurance
60
+ companies including private insurance companies shall provide free access to
61
+ the Central/State level agencies including CAG authorized to verify the accounts
62
+ and audit in respect of Crop Insurance. 13.2.7 In case, the State Govt.
63
+ subsidizes full farmers'' share of premium, in that case a token amount of at least
64
+ Re. 1 should compulsorily be charged from the farmer to facilitate electronic
65
+ tracking. 13.2.8 State Govt. has to release the State share of premium subsidy
66
+ within 3 months from premium requisition by concerned Insurance Company failing
67
+ which 1% interest per month shall be levied as penalty on the State govt.'''
68
+ - '''The ELI shall be required to apply to NABARD or NCDC, as the case may be, for Guarantee
69
+ Cover in the specified form only provided at **Annexure-II** for credit proposals
70
+ sanctioned by them during any quarter prior to expiry of the following quarter
71
+ viz., application w.r.t. credit facility sanctioned in April–June Quarter must
72
+ be submitted by the ensuing quarter, i.e. July-September to qualify for consideration
73
+ under the Scheme.  NABARD or NCDC, as the case may be, shalli. Scrutinize
74
+ the proposal before sanctioning the Guarantee Cover to the ELI under the Scheme
75
+ in accordance with its Terms and Conditions of respective Implementing Agencies. ii.
76
+ Insofar as it may be considered necessary, for the purposes of the Scheme, inspect
77
+ or call for copies of the Books of Account and other records (including any Book
78
+ of Instructions or Manual or Circulars covering general instructions regarding
79
+ Conduct of Advances) of the Lending Institution or of the Borrower from the Lending
80
+ Institution. iii. Such Inspection shall be carried out either through the officers
81
+ of NABARD or NCDC, as the case may be, or any other officer of DAC&FW specifically
82
+ appointed for the purpose of Inspection. iv. The Investment and Claims Settlement
83
+ Committee (I&CSC) of NABARD or NCDC, as the case may be, shall sanction the Guarantee
84
+ Cover to the concerned Bank based on the findings of the above. v. The ELI shall
85
+ enter into an Agreement with NABARD or NCDC, as the case may be, at the level
86
+ of the Bank Branch as per the format of Agreement decided by NABARD or NCDC, as
87
+ the case may be.'''
88
+ - '''15.1 National Technical Support Unit (NTSU) or Central Program Management
89
+ Unit (CPMU) will be created at the Central level at the earliest to provide support
90
+ and advise on risk classification & rating, development of new products, methodology
91
+ for loss assessment, legal works, workshop/training & capacity building, alongwith
92
+ use of technology including innovation/ replication, digitisation of administration
93
+ of Schemes through Crop Insurance Portal etc. 15.2 NTSU shall calculate Loss
94
+ Cost (LC) i.e. Claims as percentage (%) of Sum Insured (SI) observed in case of
95
+ notified crop(s) in notified unit area of insurance during the preceding 10 similar
96
+ crop seasons (Kharif/ Rabi) along with approximate actuarial premium rate of the
97
+ crops proposed to be notified for the season. This loss cost/premium rate shall
98
+ be based on the latest available yield data in month of January for Kharif crops
99
+ and **July** for Rabi crops and shall be provided to DAC & FW/ Concerned States on
100
+ request before invitation for premium bidding. This calculation to be done by
101
+ NTSU on behalf of DAC & FW is only for official purposes to have information on
102
+ the approximate cost to the IA for covering the risks so as to evaluate the bids
103
+ in proper perspective. 15.3 National Technical Support Unit (NTSU) may also
104
+ develop a suitable methodology for risk classification/premium rating by using
105
+ historical yield data, weather data, use and level of inputs/irrigation/technology
106
+ in crop cultivation, remote sensing data/information etc. for standardization
107
+ of methodology for risk perception and premium rating for crop insurance in the country. 15.4 In
108
+ addition to National Technical Support Unit (NTSU) / Central Programme Management
109
+ Unit (CPMU) at Central level, all States/UTs implementing Schemes should also
110
+ create a separate TSU/PMU at State HQ level with sufficient technical experts/staff
111
+ to ensure proper implementation of the Scheme.'''
112
+ - source_sentence: What is the purpose of the National Crop Insurance Portal?
113
+ sentences:
114
+ - '''| Table No. |
115
+ Topic |\n|--------------------------------------------------------------------------------------|-----------------------------------------------------------------------------|\n|
116
+ Para | |\n|
117
+ Number | |\n|
118
+ Table 1 |
119
+ Premium rate payable by the farmer |\n|
120
+ Table 2 |
121
+ Seasonality Discipline |\n|
122
+ Table 3 |
123
+ Calculation of crop-wise Sum Insured |\n|
124
+ Table 4 |
125
+ Number of CCEs to be conducted at IU level |\n|
126
+ Table 5 |
127
+ Yield Calculation for multi-picking crop |\n|
128
+ Table 6 | |\n|
129
+ Types of resolution to be used for vegetation index derivation for different | |\n|
130
+ levels of analysis | |\n|
131
+ 19.7 | |\n|
132
+ Table 7 |
133
+ Calculation of Threshold Yield |\n|
134
+ Table 8 | |\n|
135
+ Detailed Procedure for On Account Payment of Claims due to Mid-Season | |\n|
136
+ Adversity | |\n|
137
+ 21.2.7.2 | |\n|
138
+ Table 9 | |\n|
139
+ Detailed Procedure and Timelines for payment of Prevented / Failed Sowing | |\n|
140
+ and Prevented Planting / Germination Claims | |\n|
141
+ 21.3.6.1 | |\n|
142
+ Table 10 | |\n|
143
+ Detailed Procedure and Timelines for payment of | |\n|
144
+ Post Harvest Loss | |\n| | |\n|
145
+ 21.4.8.1 | |\n|
146
+ Table 11 | |\n|
147
+ Detailed Procedure and Timelines for payment of | |\n|
148
+ localized claims | |\n| | |\n|
149
+ 21.5.9.1 | |\n|
150
+ Table 12 |
151
+ Criteria for Performance Evaluation and Ranking of Loss Assessment Agencies |\n|
152
+ Table 13 | |\n|
153
+ Illustration for classification of risks, clustering/clubbing of risks and districts
154
+ | |\n|
155
+ and determination of L1 bidder | |\n|
156
+ Annexure 1 | |\n|
157
+ Table 14 | |\n|
158
+ Penalty and performance Monitoring/Evaluation of Empanelled Insurance | |\n|
159
+ Companies | |\n|
160
+ Annexure 2 | |\n|
161
+ Table 15 |
162
+ Illustrative Evaluation of Efficiency of Nodal Department of State |'''
163
+ - '''i. Soon after executing MoU, the RSA will be required to submit detailed activity-wise
164
+ action plan for the calendar year to NABARD, RO indicating, inter alia the number
165
+ of training/capacity building programmes to be conducted for POPI, FPOs etc. nature
166
+ and duration of training and other support activities required for successful
167
+ implementation of the scheme. ii. The Resource Support Agency will be designing
168
+ and organising capacity building of the Producer Organization Promoting Institutions
169
+ (POPIs), providing necessary training and handholding support to POPIs for undertaking
170
+ the promotion of FPOs in the State. RSA jointly with POPI will arrange for training
171
+ and capacity building of FPO Directors, Lead Farmers and CEO of FPO as per requirement. iii. The
172
+ RSAs will be required to guide/ oversee the overall implementation of the scheme
173
+ and assist POPI / FPOs in value addition, market linkage, accessing storage facilities
174
+ and other support services required by the FPOs for effective value chain development. iv. RSA
175
+ will work under the overall supervision and guidance of NABARD and organize various
176
+ training programmes on a group approach, as per the guidelines. Each training/capacity
177
+ building programme to be organized by RSA shall require prior approval of NABARD
178
+ for extending financial support.'''
179
+ - '''notification or /and on National Crop Insurance Portal multiplied by sown area
180
+ for notified crop. 3.1.3 Special efforts shall be made to ensure maximum
181
+ coverage of SC/ ST/ Women farmers under the Scheme. Further Panchayat Raj Institutions
182
+ (PRIs) may be involved in extension and awareness creation amongst farmers
183
+ and obtaining feed-back of farmers about the implementation of the Scheme 3.1.4 The
184
+ implementing Insurance Company selected as L1 will be responsible for taking necessary measures
185
+ to ensure at least 10% incremental increase in coverage of non-loanee farmers.
186
+ However other empanelled Insurance Companies which have participated in the bidding
187
+ and are keen for enrolment of non loanee farmers in the cluster may also be allowed
188
+ to enrol non-loanee farmers at L1 premium rate. The interested companies have
189
+ to inform their willingness in writing within seven days of finalisation of tender/issuance
190
+ of work order to L1. It will however be the responsibility of all the Insurance
191
+ Companies engaged in this process to ensure that duplicate enrolment does not
192
+ happen in the given cluster/district. Engaging companies other than L1 for enrolling
193
+ non loanee farmers will be taken up on a pilot basis in Districts notified by
194
+ State Govt. They shall enrol non loanee farmers as per conditions laid down
195
+ in Para 17.5. 3.1.5 These Insurance Company will maintain separate data of
196
+ such non loanee farmers covered by them and enter the said data on the portal
197
+ as per seasonality discipline detailed in Para 16.2. They shall be liable for
198
+ payment of claims to such farmers. 3.1.6 The exchange of information, co-witnessing
199
+ of CCEs and sharing of yield data etc for the cluster by Government/NCIP will
200
+ be limited to L1 Company only and it will be binding on other companies to accept
201
+ it. However, the requisition for payment of Government subsidy in respect of non-loanee enrolled
202
+ by them will be submitted directly to the Govt designated agency.'''
203
+ - source_sentence: Are credit facilities covered under other schemes eligible for
204
+ Guarantee Cover under the Scheme?
205
+ sentences:
206
+ - ''' (vi) A farmer can be member in more than one FPO with different produce clusters
207
+ but he/she will be eligible only once(for any one FPO that he/she is a member)
208
+ for the matching equity grant up to his/her share. (vii) In the Board of Directors
209
+ (BoD) and Governing Body (GB), as the case may be, there shall be adequate representation
210
+ of women farmer member(s) and there should be minimum one woman member. (viii) It
211
+ has a duly constituted Management Committee responsible for the business of the
212
+ FPO. (ix) It has a business plan and budget for next 18 months that is based
213
+ on a sustainable, revenue model as may be determined by the Implementing Agency.'''
214
+ - '''i. There shall be an Investment and Claims Settlement Committee (I&CSC) in NABARD
215
+ and NCDC. The I&CSC in NABARD will be chaired by its Deputy Managing Director
216
+ (Dy.MD) or as nominated by Chairman, NABARD in case Dy. MD post is vacant ; while
217
+ in NCDC, it will be chaired by its Managing Director (MD). The respective Committee
218
+ may have also representative of States in rotation, Banking Institutions and relevant
219
+ subject matter Experts apart from a representative of DAC&FW. NABARD may also
220
+ co-opt a representative from SFAC. The procedure, frequency of conduct of meeting
221
+ etc. will be decided by respective Organization as may be required.  The I&CSC
222
+ in NABARD and NCDC will apply their due diligence to scrutinize and accord the
223
+ financial sanction to the proposals received from ELIs for Credit Guarantee Cover.
224
+ NABARD and NCDC will submit by tenth day of next month the detailed progress
225
+ of sanctioned credit guarantee cover of the previous month to DAC&FW on prescribed
226
+ pro-forma. The N-PMAFSC shall be final authority in all policy and strategic decision
227
+ making in this regard. ii. The I&CSC shall have discretion to take decisions
228
+ in all matters concerning the Fund and its administration, set standards for risk
229
+ exposure, design the financial structure of the Fund and profit targets, including
230
+ special consideration for FPO with good track record of recovery. iii. The
231
+ I&CSC shall deliberate upon Policy Issues and Strategy for overall development
232
+ of the Fund and make recommendations to the NABARD and NCDC, as the case may be,
233
+ for their approval. The NABARD and NCDC will accord approval and inform N-PMAFSC. iv.'''
234
+ - '''The following credit facilities shall not be eligible for Guarantee Cover under
235
+ the Scheme: i. Any credit facility which has been sanctioned by the ELI against
236
+ collateral security and/ or third party guarantee. ii. Any credit facility in respect
237
+ of which risks are additionally covered under any scheme operated/administered
238
+ by Reserve Bank of India/or by the Government/or by any general insurer or any
239
+ other person or association of persons carrying on the business of insurance,
240
+ guarantee or indemnity. iii. Any credit facility, which does not conform to,
241
+ or is in any way inconsistent with, the provisions of any law, or with any directives
242
+ or instructions issued by the Central Government or the Reserve Bank of India,
243
+ which is, for the time being, in force. iv. Any credit facility granted to any
244
+ borrower, who has himself availed of any other credit facility covered under
245
+ this scheme or under the schemes mentioned in clause (i), (ii) and (iii) above
246
+ at any point in time. v. Any credit facility that is overdue for repayment/ NPA
247
+ taken over by the ELI from any other lender or any other default converted into
248
+ a credit facility. vi. Any credit facility which is overdue for repayment. vii.
249
+ Any credit facility which has been rescheduled or restructured on becoming overdue
250
+ for repayment.'''
251
+ - source_sentence: What is the purpose of the SLCCCI meeting?
252
+ sentences:
253
+ - '''(i) It will regularly monitor and review the progress of FPO development and functioning
254
+ by holding its regular meetings. (ii) It will work out a strategy for all stakeholders
255
+ including the Implementing Agencies (SFAC, NABARD and NCDC), institutions engaged
256
+ in formation and promotion of FPOs and State Government machinery engaged in agricultural
257
+ and rural development to work synergistically to achieve the objective. (iii) It
258
+ will identify the constraints in implementation of scheme and communicate to DAC&FW
259
+ and N-PMAFSC for taking the appropriate policy decision, if so required. (iv) It
260
+ will direct respective State Government departments to help in identification of
261
+ produce clusters for recommendation to N-PMAFSC and assist in mobilization of
262
+ farmers to form the FPOs. (v) It will formulate an effective extension mechanism
263
+ to be undertaken through existing State extension machinery down the line. (vi)
264
+ Most importantly, the committee will coordinate with respective State Government
265
+ departments to facilitate FPOs in getting the license/registration for inputs,
266
+ shops/spaces in the mandis and also availing the assistance for development of
267
+ various infrastructures relating to production and postproduction activities.
268
+ The committee will also ensure that FPOs be associated/involved in all the farmers''
269
+ centric schemes of the Government to the extent feasible. (vii) It will facilitate
270
+ in making available land at appropriate place for development of common facility
271
+ center and also custom hiring center. It will also strategize and prioritize
272
+ for linking of Common Facility Centre (CFC) with e-NAM or with any other e-trading
273
+ platform.'''
274
+ - '''| S.No | Activity | Kharif |\n|-------------------------------------------------|------------------|------------------|\n|
275
+ 1 | 15 | |\n|
276
+ th | | |\n|
277
+ November | 1 | |\n|
278
+ st | | |\n|
279
+ June | | |\n|
280
+ Nodal Department of | | |\n|
281
+ States/UTs | | |\n|
282
+ Conduct of SLCCCI meeting to take decision | | |\n|
283
+ for notification of Crops and areas, , adoption | | |\n|
284
+ of Level of Indemnity and to inform crop wise | | |\n|
285
+ Scale of Finance etc for drafting of Tender | | |\n|
286
+ documents | | |\n|
287
+ 2 | 30 | |\n|
288
+ th | | |\n|
289
+ November | 15 | |\n|
290
+ th | | |\n|
291
+ June | | |\n|
292
+ Nodal Department of | | |\n|
293
+ States/UTs | | |\n|
294
+ Uploading of requisite information/data on | | |\n|
295
+ crop insurance Portal and Issuing of tender | | |\n|
296
+ documents | | |\n|
297
+ 3 | | |\n|
298
+ Finalization of Tender and award of work by | | |\n|
299
+ States/UTs | | |\n|
300
+ 31 | | |\n|
301
+ st | | |\n|
302
+ December | 15 | |\n|
303
+ th | | |\n|
304
+ July | States/UT | |\n|
305
+ 4 | 31 | |\n|
306
+ st | | |\n|
307
+ January | 31 | |\n|
308
+ st | | |\n|
309
+ July | | |\n|
310
+ Digitization of notification and downloading/ | | |\n|
311
+ issuance of notification from crop insurance | | |\n|
312
+ Portal for circulation amongst stakeholders | | |\n|
313
+ By State/UTs and | | |\n|
314
+ concerned Insurance | | |\n|
315
+ Companies | | |\n|
316
+ 5 | | |\n|
317
+ Awareness/ sensitization/training | | |\n|
318
+ programmes by State Govt. and ICs/IAs | | |\n|
319
+ From 15 | | |\n|
320
+ th | | |\n|
321
+ March | 15 | |\n|
322
+ th | | |\n|
323
+ September | | |\n|
324
+ By State/UTs and | | |\n|
325
+ concerned Insurance | | |\n|
326
+ Companies | | |\n|
327
+ 6 | | |\n|
328
+ Start of enrolment of farmers for the | | |\n|
329
+ season(as per crop calendar) | | |\n|
330
+ From 1st April | From 1st October | All Stakeholders
331
+ |\n| Farmers/Bank | 7 | |\n|
332
+ Cut-off date for intimation of change of | | |\n|
333
+ insured crop by the loanee farmer | | |\n|
334
+ 2 working days prior | | |\n|
335
+ to cut-off date for | | |\n|
336
+ debit/collection of | | |\n|
337
+ premium from | | |\n|
338
+ farmers | | |\n|
339
+ 2 working days prior | | |\n|
340
+ to cut-off date for | | |\n|
341
+ debit/collection of | | |\n|
342
+ premium from | | |\n|
343
+ farmers | | |'''
344
+ - '''14.2.1 Considering the significance and strategic roles of the State Government
345
+ and its machinery down the line in synergizing the efforts in mobilizing the
346
+ farmers, offering various services relating to production and post-production
347
+ and also closely and periodically review the developmental and functioning including constraints
348
+ faced by FPOs, a State Level Consultative Committee called SLCC is constituted.
349
+ The Composition of the Consultative Committee is as under- i. Addl. Chief
350
+ Secy/Secy. I/c Agriculture/ Agriculture Marketing----Chairperson ii. Secretary
351
+ of line Ministries/Departments ------------------------------Members (Horticulture,
352
+ Animal Husbandry& Fisheries, Co-operation, Marketing, Rural Development, Panchayati
353
+ Raj) (In State/UT where there is no post of ACS/PS; and Secretary, I/c Agriculture
354
+ is Chairperson, members may be Directors of line Departments) iii. Representative
355
+ of SFAC ---------------------------------------------------Member iv. Representative
356
+ of NCDC -------------------------------------------------Member v. Representative
357
+ of NABARD ----------------------------------Member Secretary* vi. Convener,
358
+ SLBC -----------------------------------------------------------Member vii. Two
359
+ Experts from Agriculture Universities/Institutions ----------------Members Note:
360
+ (i) Chairman may co-opt additional Member(s) as per requirement and In-charge
361
+ of DMI of respective State/UT may be invited in the meeting as special invitee.
362
+ Two prominent farmer members or representatives of FPOs in the State/UT be nominated
363
+ as member of the Committee by Government of India. *(ii)Representative of NABARD
364
+ will be Member Secretary to convene and coordinate the meetings of SLCC, however,
365
+ in State (s) where there is presence of NCDC and State level SFAC and they are
366
+ playing lead and significant roles in formation and promotion of FPOs, their
367
+ representatives may be Member Secretary in place of NABARD). In case of any
368
+ dispute, decision of N-PMAFSC in this regard shall be final. (iii)State
369
+ Governments/ UT Administration will issue order for formation of State Level
370
+ Consultative Committee.'''
371
+ - source_sentence: What is the requirement for banks regarding Aadhaar for crop loans
372
+ and Kisan Credit Card (KCC) under the Interest Subvention Scheme?
373
+ sentences:
374
+ - '''notification or /and on National Crop Insurance Portal multiplied by sown area
375
+ for notified crop. 3.1.3 Special efforts shall be made to ensure maximum
376
+ coverage of SC/ ST/ Women farmers under the Scheme. Further Panchayat Raj Institutions
377
+ (PRIs) may be involved in extension and awareness creation amongst farmers
378
+ and obtaining feed-back of farmers about the implementation of the Scheme 3.1.4 The
379
+ implementing Insurance Company selected as L1 will be responsible for taking necessary measures
380
+ to ensure at least 10% incremental increase in coverage of non-loanee farmers.
381
+ However other empanelled Insurance Companies which have participated in the bidding
382
+ and are keen for enrolment of non loanee farmers in the cluster may also be allowed
383
+ to enrol non-loanee farmers at L1 premium rate. The interested companies have
384
+ to inform their willingness in writing within seven days of finalisation of tender/issuance
385
+ of work order to L1. It will however be the responsibility of all the Insurance
386
+ Companies engaged in this process to ensure that duplicate enrolment does not
387
+ happen in the given cluster/district. Engaging companies other than L1 for enrolling
388
+ non loanee farmers will be taken up on a pilot basis in Districts notified by
389
+ State Govt. They shall enrol non loanee farmers as per conditions laid down
390
+ in Para 17.5. 3.1.5 These Insurance Company will maintain separate data of
391
+ such non loanee farmers covered by them and enter the said data on the portal
392
+ as per seasonality discipline detailed in Para 16.2. They shall be liable for
393
+ payment of claims to such farmers. 3.1.6 The exchange of information, co-witnessing
394
+ of CCEs and sharing of yield data etc for the cluster by Government/NCIP will
395
+ be limited to L1 Company only and it will be binding on other companies to accept
396
+ it. However, the requisition for payment of Government subsidy in respect of non-loanee enrolled
397
+ by them will be submitted directly to the Govt designated agency.'''
398
+ - '''The protocol of timelines defined above shall also be applicable to the actual
399
+ Yield data being provided by the concerned State department in batches, i.e. Crop-wise-District-wise
400
+ lots of data and timelines for each batch of data shall be counted separately/batch
401
+ wise. Unnecessary delay in finalization of yield data will attract penal interest
402
+ on due claim amount payable to the farmers as decided by the SLCC in the matter. 19.3 The
403
+ existing Technical Advisory Committee (TAC) comprising of representatives from
404
+ Directorate of Economics and Statistics, DAC&FW, MNCFC, NSSO, IASRI, will be
405
+ further augmented and constituted under the chairmanship of **Additional Secretary,
406
+ DAC&FW** and additional members depending on case to case basis will be drawn
407
+ from various organizations namely, relevant institution under Indian Council of
408
+ Agriculture Research (ICAR), Indian Agriculture Research Institute (IARI), National
409
+ Remote Sensing Centre (NRSC), Space Applications Centre (SAC), Central Statistical
410
+ Organization(CSO), Insurance Regulatory and Development Authority of India (IRDAI),
411
+ Reserve Bank of India (RBI), National Bank for Agriculture & Rural Development
412
+ (NABARD), India Meteorological Department (IMD) or any other such organisation. 19.4 The
413
+ role of TAC shall be to review the technical matters related to execution of the
414
+ Scheme. In case, the matter requires further focused deliberations, the TAC will
415
+ refer the issue to Technical Agency (as defined in the SOP for yield dispute).
416
+ Technical Agency (TA) will compulsorily take input/reports/relevant information
417
+ from concerned State Department and Insurance Company and may also opt for representatives
418
+ of other organizations/experts (if required) with permission of chairman of TAC
419
+ and follow the procedure as defined in the SOP. 19.5 State shall also constitute
420
+ the State Level Technical Advisory Committee (STAC) on similar pattern to resolve
421
+ disputes.'''
422
+ - '''6.3.1 Aadhaar has been made mandatory for availing Crop insurance from Kharif
423
+ 2017 season onwards. Therefore, all banks are advised to mandatorily obtain
424
+ Aadhaar number of their farmers and the same applies for non-loanee farmers enrolled through banks/Insurance companies/insurance intermediaries. 6.3.2 Farmers
425
+ not having Aadhaar ID may also enrol under PMFBY subject to their enrolment for Aadhaar
426
+ and submission of proof of such enrolment as per notification No. 334.dated 8th
427
+ February, 2017 issued by GOI under Section 7 of Aadhaar Act 2016(Targeted Delivery
428
+ of Financial and other Subsidies, Benefits and Services). Copy of the notification
429
+ may be perused on www.pmfby.gov.in. This may be subject to further directions
430
+ issued by Govt. from time to time. 6.3.3 All banks have to compulsorily take
431
+ Aadhaar/Aadhaar enrolment number as per notification under Aadhaar Act before
432
+ sanction of crop loan/KCC under Interest Subvention Scheme. Hence the coverage of
433
+ loanee farmers without Aadhaar does not arise and such accounts need to be reviewed
434
+ by the concerned bank branch regularly.'''
435
+ pipeline_tag: sentence-similarity
436
+ model-index:
437
+ - name: SentenceTransformer based on BAAI/bge-small-en-v1.5
438
+ results:
439
+ - task:
440
+ type: information-retrieval
441
+ name: Information Retrieval
442
+ dataset:
443
+ name: val evaluator
444
+ type: val_evaluator
445
+ metrics:
446
+ - type: cosine_accuracy@1
447
+ value: 0.53
448
+ name: Cosine Accuracy@1
449
+ - type: cosine_accuracy@5
450
+ value: 0.88
451
+ name: Cosine Accuracy@5
452
+ - type: cosine_accuracy@10
453
+ value: 0.94
454
+ name: Cosine Accuracy@10
455
+ - type: cosine_precision@1
456
+ value: 0.53
457
+ name: Cosine Precision@1
458
+ - type: cosine_precision@5
459
+ value: 0.176
460
+ name: Cosine Precision@5
461
+ - type: cosine_precision@10
462
+ value: 0.09399999999999997
463
+ name: Cosine Precision@10
464
+ - type: cosine_recall@1
465
+ value: 0.53
466
+ name: Cosine Recall@1
467
+ - type: cosine_recall@5
468
+ value: 0.88
469
+ name: Cosine Recall@5
470
+ - type: cosine_recall@10
471
+ value: 0.94
472
+ name: Cosine Recall@10
473
+ - type: cosine_ndcg@5
474
+ value: 0.7258950722710406
475
+ name: Cosine Ndcg@5
476
+ - type: cosine_ndcg@10
477
+ value: 0.7457311797109036
478
+ name: Cosine Ndcg@10
479
+ - type: cosine_ndcg@100
480
+ value: 0.7599721862628874
481
+ name: Cosine Ndcg@100
482
+ - type: cosine_mrr@5
483
+ value: 0.6736666666666666
484
+ name: Cosine Mrr@5
485
+ - type: cosine_mrr@10
486
+ value: 0.6821071428571428
487
+ name: Cosine Mrr@10
488
+ - type: cosine_mrr@100
489
+ value: 0.6857238570180357
490
+ name: Cosine Mrr@100
491
+ - type: cosine_map@100
492
+ value: 0.685723857018036
493
+ name: Cosine Map@100
494
+ - type: dot_accuracy@1
495
+ value: 0.53
496
+ name: Dot Accuracy@1
497
+ - type: dot_accuracy@5
498
+ value: 0.88
499
+ name: Dot Accuracy@5
500
+ - type: dot_accuracy@10
501
+ value: 0.94
502
+ name: Dot Accuracy@10
503
+ - type: dot_precision@1
504
+ value: 0.53
505
+ name: Dot Precision@1
506
+ - type: dot_precision@5
507
+ value: 0.176
508
+ name: Dot Precision@5
509
+ - type: dot_precision@10
510
+ value: 0.09399999999999997
511
+ name: Dot Precision@10
512
+ - type: dot_recall@1
513
+ value: 0.53
514
+ name: Dot Recall@1
515
+ - type: dot_recall@5
516
+ value: 0.88
517
+ name: Dot Recall@5
518
+ - type: dot_recall@10
519
+ value: 0.94
520
+ name: Dot Recall@10
521
+ - type: dot_ndcg@5
522
+ value: 0.7258950722710406
523
+ name: Dot Ndcg@5
524
+ - type: dot_ndcg@10
525
+ value: 0.7457311797109036
526
+ name: Dot Ndcg@10
527
+ - type: dot_ndcg@100
528
+ value: 0.7599721862628874
529
+ name: Dot Ndcg@100
530
+ - type: dot_mrr@5
531
+ value: 0.6736666666666666
532
+ name: Dot Mrr@5
533
+ - type: dot_mrr@10
534
+ value: 0.6821071428571428
535
+ name: Dot Mrr@10
536
+ - type: dot_mrr@100
537
+ value: 0.6857238570180357
538
+ name: Dot Mrr@100
539
+ - type: dot_map@100
540
+ value: 0.685723857018036
541
+ name: Dot Map@100
542
+ ---
543
+
544
+ # SentenceTransformer based on BAAI/bge-small-en-v1.5
545
+
546
+ This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [BAAI/bge-small-en-v1.5](https://huggingface.co/BAAI/bge-small-en-v1.5). It maps sentences & paragraphs to a 384-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
547
+
548
+ ## Model Details
549
+
550
+ ### Model Description
551
+ - **Model Type:** Sentence Transformer
552
+ - **Base model:** [BAAI/bge-small-en-v1.5](https://huggingface.co/BAAI/bge-small-en-v1.5) <!-- at revision 5c38ec7c405ec4b44b94cc5a9bb96e735b38267a -->
553
+ - **Maximum Sequence Length:** 512 tokens
554
+ - **Output Dimensionality:** 384 tokens
555
+ - **Similarity Function:** Cosine Similarity
556
+ <!-- - **Training Dataset:** Unknown -->
557
+ <!-- - **Language:** Unknown -->
558
+ <!-- - **License:** Unknown -->
559
+
560
+ ### Model Sources
561
+
562
+ - **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
563
+ - **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
564
+ - **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
565
+
566
+ ### Full Model Architecture
567
+
568
+ ```
569
+ SentenceTransformer(
570
+ (0): Transformer({'max_seq_length': 512, 'do_lower_case': True}) with Transformer model: BertModel
571
+ (1): Pooling({'word_embedding_dimension': 384, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
572
+ (2): Normalize()
573
+ )
574
+ ```
575
+
576
+ ## Usage
577
+
578
+ ### Direct Usage (Sentence Transformers)
579
+
580
+ First install the Sentence Transformers library:
581
+
582
+ ```bash
583
+ pip install -U sentence-transformers
584
+ ```
585
+
586
+ Then you can load this model and run inference.
587
+ ```python
588
+ from sentence_transformers import SentenceTransformer
589
+
590
+ # Download from the 🤗 Hub
591
+ model = SentenceTransformer("SamagraDataGov/embedding_finetuned")
592
+ # Run inference
593
+ sentences = [
594
+ 'What is the requirement for banks regarding Aadhaar for crop loans and Kisan Credit Card (KCC) under the Interest Subvention Scheme?',
595
+ "'6.3.1 Aadhaar has been made mandatory for availing Crop insurance from Kharif 2017 season onwards. Therefore, all banks are advised to mandatorily obtain Aadhaar number of their farmers and the same applies for non-loanee farmers enrolled through banks/Insurance companies/insurance intermediaries. 6.3.2 Farmers not having Aadhaar ID may also enrol under PMFBY subject to their enrolment for Aadhaar and submission of proof of such enrolment as per notification No. 334.dated 8th February, 2017 issued by GOI under Section 7 of Aadhaar Act 2016(Targeted Delivery of Financial and other Subsidies, Benefits and Services). Copy of the notification may be perused on www.pmfby.gov.in. This may be subject to further directions issued by Govt. from time to time. 6.3.3 All banks have to compulsorily take Aadhaar/Aadhaar enrolment number as per notification under Aadhaar Act before sanction of crop loan/KCC under Interest Subvention Scheme. Hence the coverage of loanee farmers without Aadhaar does not arise and such accounts need to be reviewed by the concerned bank branch regularly.'",
596
+ "'The protocol of timelines defined above shall also be applicable to the actual Yield data being provided by the concerned State department in batches, i.e. Crop-wise-District-wise lots of data and timelines for each batch of data shall be counted separately/batch wise. Unnecessary delay in finalization of yield data will attract penal interest on due claim amount payable to the farmers as decided by the SLCC in the matter. 19.3 The existing Technical Advisory Committee (TAC) comprising of representatives from Directorate of Economics and Statistics, DAC&FW, MNCFC, NSSO, IASRI, will be further augmented and constituted under the chairmanship of **Additional Secretary, DAC&FW** and additional members depending on case to case basis will be drawn from various organizations namely, relevant institution under Indian Council of Agriculture Research (ICAR), Indian Agriculture Research Institute (IARI), National Remote Sensing Centre (NRSC), Space Applications Centre (SAC), Central Statistical Organization(CSO), Insurance Regulatory and Development Authority of India (IRDAI), Reserve Bank of India (RBI), National Bank for Agriculture & Rural Development (NABARD), India Meteorological Department (IMD) or any other such organisation. 19.4 The role of TAC shall be to review the technical matters related to execution of the Scheme. In case, the matter requires further focused deliberations, the TAC will refer the issue to Technical Agency (as defined in the SOP for yield dispute). Technical Agency (TA) will compulsorily take input/reports/relevant information from concerned State Department and Insurance Company and may also opt for representatives of other organizations/experts (if required) with permission of chairman of TAC and follow the procedure as defined in the SOP. 19.5 State shall also constitute the State Level Technical Advisory Committee (STAC) on similar pattern to resolve disputes.'",
597
+ ]
598
+ embeddings = model.encode(sentences)
599
+ print(embeddings.shape)
600
+ # [3, 384]
601
+
602
+ # Get the similarity scores for the embeddings
603
+ similarities = model.similarity(embeddings, embeddings)
604
+ print(similarities.shape)
605
+ # [3, 3]
606
+ ```
607
+
608
+ <!--
609
+ ### Direct Usage (Transformers)
610
+
611
+ <details><summary>Click to see the direct usage in Transformers</summary>
612
+
613
+ </details>
614
+ -->
615
+
616
+ <!--
617
+ ### Downstream Usage (Sentence Transformers)
618
+
619
+ You can finetune this model on your own dataset.
620
+
621
+ <details><summary>Click to expand</summary>
622
+
623
+ </details>
624
+ -->
625
+
626
+ <!--
627
+ ### Out-of-Scope Use
628
+
629
+ *List how the model may foreseeably be misused and address what users ought not to do with the model.*
630
+ -->
631
+
632
+ ## Evaluation
633
+
634
+ ### Metrics
635
+
636
+ #### Information Retrieval
637
+ * Dataset: `val_evaluator`
638
+ * Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)
639
+
640
+ | Metric | Value |
641
+ |:--------------------|:-----------|
642
+ | cosine_accuracy@1 | 0.53 |
643
+ | cosine_accuracy@5 | 0.88 |
644
+ | cosine_accuracy@10 | 0.94 |
645
+ | cosine_precision@1 | 0.53 |
646
+ | cosine_precision@5 | 0.176 |
647
+ | cosine_precision@10 | 0.094 |
648
+ | cosine_recall@1 | 0.53 |
649
+ | cosine_recall@5 | 0.88 |
650
+ | cosine_recall@10 | 0.94 |
651
+ | cosine_ndcg@5 | 0.7259 |
652
+ | cosine_ndcg@10 | 0.7457 |
653
+ | cosine_ndcg@100 | 0.76 |
654
+ | cosine_mrr@5 | 0.6737 |
655
+ | cosine_mrr@10 | 0.6821 |
656
+ | cosine_mrr@100 | 0.6857 |
657
+ | **cosine_map@100** | **0.6857** |
658
+ | dot_accuracy@1 | 0.53 |
659
+ | dot_accuracy@5 | 0.88 |
660
+ | dot_accuracy@10 | 0.94 |
661
+ | dot_precision@1 | 0.53 |
662
+ | dot_precision@5 | 0.176 |
663
+ | dot_precision@10 | 0.094 |
664
+ | dot_recall@1 | 0.53 |
665
+ | dot_recall@5 | 0.88 |
666
+ | dot_recall@10 | 0.94 |
667
+ | dot_ndcg@5 | 0.7259 |
668
+ | dot_ndcg@10 | 0.7457 |
669
+ | dot_ndcg@100 | 0.76 |
670
+ | dot_mrr@5 | 0.6737 |
671
+ | dot_mrr@10 | 0.6821 |
672
+ | dot_mrr@100 | 0.6857 |
673
+ | dot_map@100 | 0.6857 |
674
+
675
+ <!--
676
+ ## Bias, Risks and Limitations
677
+
678
+ *What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
679
+ -->
680
+
681
+ <!--
682
+ ### Recommendations
683
+
684
+ *What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
685
+ -->
686
+
687
+ ## Training Details
688
+
689
+ ### Training Hyperparameters
690
+ #### Non-Default Hyperparameters
691
+
692
+ - `eval_strategy`: steps
693
+ - `gradient_accumulation_steps`: 4
694
+ - `learning_rate`: 1e-05
695
+ - `weight_decay`: 0.01
696
+ - `num_train_epochs`: 1.0
697
+ - `warmup_ratio`: 0.1
698
+ - `load_best_model_at_end`: True
699
+
700
+ #### All Hyperparameters
701
+ <details><summary>Click to expand</summary>
702
+
703
+ - `overwrite_output_dir`: False
704
+ - `do_predict`: False
705
+ - `eval_strategy`: steps
706
+ - `prediction_loss_only`: True
707
+ - `per_device_train_batch_size`: 8
708
+ - `per_device_eval_batch_size`: 8
709
+ - `per_gpu_train_batch_size`: None
710
+ - `per_gpu_eval_batch_size`: None
711
+ - `gradient_accumulation_steps`: 4
712
+ - `eval_accumulation_steps`: None
713
+ - `learning_rate`: 1e-05
714
+ - `weight_decay`: 0.01
715
+ - `adam_beta1`: 0.9
716
+ - `adam_beta2`: 0.999
717
+ - `adam_epsilon`: 1e-08
718
+ - `max_grad_norm`: 1.0
719
+ - `num_train_epochs`: 1.0
720
+ - `max_steps`: -1
721
+ - `lr_scheduler_type`: linear
722
+ - `lr_scheduler_kwargs`: {}
723
+ - `warmup_ratio`: 0.1
724
+ - `warmup_steps`: 0
725
+ - `log_level`: passive
726
+ - `log_level_replica`: warning
727
+ - `log_on_each_node`: True
728
+ - `logging_nan_inf_filter`: True
729
+ - `save_safetensors`: True
730
+ - `save_on_each_node`: False
731
+ - `save_only_model`: False
732
+ - `restore_callback_states_from_checkpoint`: False
733
+ - `no_cuda`: False
734
+ - `use_cpu`: False
735
+ - `use_mps_device`: False
736
+ - `seed`: 42
737
+ - `data_seed`: None
738
+ - `jit_mode_eval`: False
739
+ - `use_ipex`: False
740
+ - `bf16`: False
741
+ - `fp16`: False
742
+ - `fp16_opt_level`: O1
743
+ - `half_precision_backend`: auto
744
+ - `bf16_full_eval`: False
745
+ - `fp16_full_eval`: False
746
+ - `tf32`: None
747
+ - `local_rank`: 0
748
+ - `ddp_backend`: None
749
+ - `tpu_num_cores`: None
750
+ - `tpu_metrics_debug`: False
751
+ - `debug`: []
752
+ - `dataloader_drop_last`: False
753
+ - `dataloader_num_workers`: 0
754
+ - `dataloader_prefetch_factor`: None
755
+ - `past_index`: -1
756
+ - `disable_tqdm`: False
757
+ - `remove_unused_columns`: True
758
+ - `label_names`: None
759
+ - `load_best_model_at_end`: True
760
+ - `ignore_data_skip`: False
761
+ - `fsdp`: []
762
+ - `fsdp_min_num_params`: 0
763
+ - `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
764
+ - `fsdp_transformer_layer_cls_to_wrap`: None
765
+ - `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
766
+ - `deepspeed`: None
767
+ - `label_smoothing_factor`: 0.0
768
+ - `optim`: adamw_torch
769
+ - `optim_args`: None
770
+ - `adafactor`: False
771
+ - `group_by_length`: False
772
+ - `length_column_name`: length
773
+ - `ddp_find_unused_parameters`: None
774
+ - `ddp_bucket_cap_mb`: None
775
+ - `ddp_broadcast_buffers`: False
776
+ - `dataloader_pin_memory`: True
777
+ - `dataloader_persistent_workers`: False
778
+ - `skip_memory_metrics`: True
779
+ - `use_legacy_prediction_loop`: False
780
+ - `push_to_hub`: False
781
+ - `resume_from_checkpoint`: None
782
+ - `hub_model_id`: None
783
+ - `hub_strategy`: every_save
784
+ - `hub_private_repo`: False
785
+ - `hub_always_push`: False
786
+ - `gradient_checkpointing`: False
787
+ - `gradient_checkpointing_kwargs`: None
788
+ - `include_inputs_for_metrics`: False
789
+ - `eval_do_concat_batches`: True
790
+ - `fp16_backend`: auto
791
+ - `push_to_hub_model_id`: None
792
+ - `push_to_hub_organization`: None
793
+ - `mp_parameters`:
794
+ - `auto_find_batch_size`: False
795
+ - `full_determinism`: False
796
+ - `torchdynamo`: None
797
+ - `ray_scope`: last
798
+ - `ddp_timeout`: 1800
799
+ - `torch_compile`: False
800
+ - `torch_compile_backend`: None
801
+ - `torch_compile_mode`: None
802
+ - `dispatch_batches`: None
803
+ - `split_batches`: None
804
+ - `include_tokens_per_second`: False
805
+ - `include_num_input_tokens_seen`: False
806
+ - `neftune_noise_alpha`: None
807
+ - `optim_target_modules`: None
808
+ - `batch_eval_metrics`: False
809
+ - `batch_sampler`: batch_sampler
810
+ - `multi_dataset_batch_sampler`: proportional
811
+
812
+ </details>
813
+
814
+ ### Training Logs
815
+ | Epoch | Step | Training Loss | loss | val_evaluator_cosine_map@100 |
816
+ |:---------:|:------:|:-------------:|:----------:|:----------------------------:|
817
+ | **0.531** | **15** | **0.5565** | **0.1262** | **0.6857** |
818
+ | 0.9912 | 28 | - | 0.1262 | 0.6857 |
819
+
820
+ * The bold row denotes the saved checkpoint.
821
+
822
+ ### Framework Versions
823
+ - Python: 3.10.14
824
+ - Sentence Transformers: 3.0.1
825
+ - Transformers: 4.41.1
826
+ - PyTorch: 2.3.0+cu121
827
+ - Accelerate: 0.27.2
828
+ - Datasets: 2.19.1
829
+ - Tokenizers: 0.19.1
830
+
831
+ ## Citation
832
+
833
+ ### BibTeX
834
+
835
+ #### Sentence Transformers
836
+ ```bibtex
837
+ @inproceedings{reimers-2019-sentence-bert,
838
+ title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
839
+ author = "Reimers, Nils and Gurevych, Iryna",
840
+ booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
841
+ month = "11",
842
+ year = "2019",
843
+ publisher = "Association for Computational Linguistics",
844
+ url = "https://arxiv.org/abs/1908.10084",
845
+ }
846
+ ```
847
+
848
+ #### GISTEmbedLoss
849
+ ```bibtex
850
+ @misc{solatorio2024gistembed,
851
+ title={GISTEmbed: Guided In-sample Selection of Training Negatives for Text Embedding Fine-tuning},
852
+ author={Aivin V. Solatorio},
853
+ year={2024},
854
+ eprint={2402.16829},
855
+ archivePrefix={arXiv},
856
+ primaryClass={cs.LG}
857
+ }
858
+ ```
859
+
860
+ <!--
861
+ ## Glossary
862
+
863
+ *Clearly define terms in order to be accessible across audiences.*
864
+ -->
865
+
866
+ <!--
867
+ ## Model Card Authors
868
+
869
+ *Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
870
+ -->
871
+
872
+ <!--
873
+ ## Model Card Contact
874
+
875
+ *Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
876
+ -->
config.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "BAAI/bge-small-en-v1.5",
3
+ "architectures": [
4
+ "BertModel"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "classifier_dropout": null,
8
+ "hidden_act": "gelu",
9
+ "hidden_dropout_prob": 0.1,
10
+ "hidden_size": 384,
11
+ "id2label": {
12
+ "0": "LABEL_0"
13
+ },
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 1536,
16
+ "label2id": {
17
+ "LABEL_0": 0
18
+ },
19
+ "layer_norm_eps": 1e-12,
20
+ "max_position_embeddings": 512,
21
+ "model_type": "bert",
22
+ "num_attention_heads": 12,
23
+ "num_hidden_layers": 12,
24
+ "pad_token_id": 0,
25
+ "position_embedding_type": "absolute",
26
+ "torch_dtype": "float32",
27
+ "transformers_version": "4.41.1",
28
+ "type_vocab_size": 2,
29
+ "use_cache": true,
30
+ "vocab_size": 30522
31
+ }
config_sentence_transformers.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "__version__": {
3
+ "sentence_transformers": "3.0.1",
4
+ "transformers": "4.41.1",
5
+ "pytorch": "2.3.0+cu121"
6
+ },
7
+ "prompts": {},
8
+ "default_prompt_name": null,
9
+ "similarity_fn_name": null
10
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d449ea16287f0cc901a0533835435abf64a34a3dad28a43d75e11ab3900a4e58
3
+ size 133462128
modules.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ {
15
+ "idx": 2,
16
+ "name": "2",
17
+ "path": "2_Normalize",
18
+ "type": "sentence_transformers.models.Normalize"
19
+ }
20
+ ]
sentence_bert_config.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "max_seq_length": 512,
3
+ "do_lower_case": true
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 +1,57 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "mask_token": "[MASK]",
49
+ "model_max_length": 512,
50
+ "never_split": null,
51
+ "pad_token": "[PAD]",
52
+ "sep_token": "[SEP]",
53
+ "strip_accents": null,
54
+ "tokenize_chinese_chars": true,
55
+ "tokenizer_class": "BertTokenizer",
56
+ "unk_token": "[UNK]"
57
+ }
vocab.txt ADDED
The diff for this file is too large to render. See raw diff