Edit model card

SetFit with BAAI/bge-small-en-v1.5

This is a SetFit model that can be used for Text Classification. This SetFit model uses BAAI/bge-small-en-v1.5 as the Sentence Transformer embedding model. A LogisticRegression instance is used for classification.

The model has been trained using an efficient few-shot learning technique that involves:

  1. Fine-tuning a Sentence Transformer with contrastive learning.
  2. Training a classification head with features from the fine-tuned Sentence Transformer.

Model Details

Model Description

Model Sources

Model Labels

Label Examples
3
  • ' \n\n \n\nHINDALCO INDUSTRIES LTE...\nHIRARUD POWER\n\np lGEn-T2 Te- 31-J\n\n \n \n \n\nUbATE OTIM\n\n \n\n \n\n28-21-2022\n\n \n\nao e301\n\n \n\n \n\n1 ints MG}\n\n \n\n \n \n\n \n\n \n\ntute\n\n \n\n \n\n \n\nrey\n\n \n \n \n \n\n \n\n \n\n4\n\x0c'
  • 'HINDALCO INDUSTRIES LTD.,\n\n \n\nJ HIRAKUD POWER\nASH WEIGH ERIDGE PRINTING DATE : 09-01-2021\nASH FROM :- 01-Nov-20 To - 30-Nov-20 PAGENUMBER: 1\nSLNO TRENG LDATE” PARTY "ODATE OTIME INDATE INTIME TARE GROSS HET” We OLOAD SLD OpERATOR\n\n \n\n \n\nKAY KAY BRI\n. Ipce,\n11-20-724895 ORQIN4428 24-11-2020 IDC(B) —-M-11-2020 10:10 4-11-2020 09:08 10.65 27.91 «17.26 19.00 000 6 N.PANDA\n11-20-724979 « OROON4428 4-11-2020 IDC(B) «4-11-2030 19:34 24-11-2020 17:25 10.65 27.04 16.39 19.00 «8G 64) ON. PANDA\n11-20-725215 ODISF8190 26-11-2020 IDC(B) 26-11-2020.0748 36-11-2020 O1-01 14.12 33.95 19.83 1998 «= 00 6 ITU\naie 11-20-725261 ODISG7595. 26-11-2028 IDC(B) 26-11-2020 12:05 26-11-2020 10-48 11.39 27.38 ««15.99 1760 000 2 N.PANDA\n11-20-725376 OD15G7595 27-11-2020 IDCCB) 27-11-2020. 04:20 27-11-2020 03:22 11.36 © 30.05 «18.69 1760 «1869 § = ~ITU\n11-20-725445 OROINE428 27-11-2020 IDCCB) 77-11-2020 16:39 27-11-2020 1:55 10.58 27.33 16.75 19.00 «8.00 6 N.PANDA\n11-20-725559 — OD15G7595_ 28-11-2020 JDC(B) 28-11-2020 10:08 28-11-2020 08:39 1148 28.88 1742 1760 600 2 N.PANDA\nLi 31-20-725591 ODISFS190 28-11-2020 DCB) 98-11-2020 1:15 28-11-2020 10:21 14.31 36.77) 2.96 19.96 21.96 2 N. PANDA\n11-20-725866 ODISG7595 29-11-2020 JDC(B) 29-11-2020 23:21 29-11-2020 21:09 11.25 77.60 16.35 1760 900 2 N.PANDA\nPARTY TRIPS- 9 77 26641 160.64 167.31 40.65\nAREATRIPS- 9 266.41 160.64 167.31\n\n \n\n \n\n \n\n \n\x0c'
  • ' \n\n" G1-20-679262 ODITF4129 26-01-2000 AK.S 26-01-2000 16:18 6-01-2000 1448 1038 2858 «1918 14.88 1818\n\n. 10-19-665660 ODI7F4129 30-10-2019 AK.S 31-10-2019 00:3\n\nHINDALCO INDUSTRIES LTD.,\nHIRAKUD POWER\n\nPRINTING DATE : 21-03-2020\nFROM :- 03-Jul-19 To - 31-Jan-20 PAGENUMBER . 1\n\n \n\n \n\nLDATE PARTY ODATE OTIME INDATE INTIME TARE\n\n \n\nSHREESAI BRK &EL\nAKS\n\nG1-20-676454 «= GDI7F4229 (05-01-2000) AKS 06-01-2020 00:42\n01-20-676764 «= ODI7F4229 08-01-2020 AK.S 08-01-2020 19:18\nG1-20-677590 ODITF4229 15-01-WAKS 15-01-2020 0511 15-91-\nO1-20-677704 = ODITF4129 15-01-2020 ARS 15-01-2020 23:42 15-01-2020 19:22 10.67 27.99 17.32 14.88) 17.33\nGi -20-67784- ODITF4129 16-01-2020 ARS = 17-01-2020 01:00 16-01-2020 23:19 1065 2882 1517 14.88 15.17\nO1-20-678816 ODITF4i29 24-01-2020 ARS = 4-01-2020 02:46 4-01-2020 01:43 10.23 28.56 15.53 14.88 = 1533\n\n \n\n10.89 77.04 16.15 14.88
0
  • ', TAX INVOICE\n\n8looin Combustion India Pvt Ltd\nv-44' Near Quality Circie\nA160 C Bhoecurt\n\nPraerr 17 F026\nSITUINOUIN 2 AADCBI ibe les\nate Name Moharashla, Gode 27\n\nCIM US89U-:Mm2007PTC172730\n\nvoniacl 920-60 1242 10, 9850754646\n\nE-Maut OPatekar@bloomeng com\n\nCons.nce (Ship to)\n\nHindatco Industries Limited (Odisha)\nHirakuo Smetter, Hlrakud - 768016, Sambalpur\nOdisha\nGSTIN/UIN\nPANIIT Ne\nStale Name\ntis eo (Lallaoe\nHindalco tndustrtes Limited (Odisha)\n\nHirakud Smelter Hirakud - 768016, Samboilpur\nOdssha\n\n21A4AACH1201R1ZZ7\nAAACH1201R\nOdisha, Code 21\n\nGSTIAN‘UIN 21AAACH1201R1Z2Z\nPAN TI No AAATCH1201R\n\nSte Name Odisha Code .'1\nyee Al sappy Odisha\n\n \n\nDesrnotion of\n‘s Goods and Senices\n\n3h\n\n- Bumer Spares (Mtr)\n\n2On Poe OS Wire igncian Qia-1 Smm2\n2 Surner Spares\n\n47 4762R0005, Gas Cycle Valve\n= Bummer Spares\n\not 122050118 Pressure Connector 1°\n\nPACKING & FORWARDING\n(GST Output @ 18%\n\nTotal\nTerms & Cenditons\n\nAmcunt Chargeable in words}\n\nINR Three Lakh Fifty Thousand Two Hundred Fifty Three\nand Sixty Two passe Only\n\n‘ops, SPAN AADCB1916E\nsti 9\nNe der urs that tenon ice snes he at tua price of ine\n\nJods ueser bed end tna! da" garticu ars are ‘vue ard coi-eci\n\n22\0292V2\n\n(DUPLICATE FOR TRANSPORTER)\n\n'nvoice NO\n22-23/2595/M123\nDeltvery Note\n\nReference No & Date\nABW ATA tor\n\nfh\n\nDated\n\n31-Oct-22\nMode/Terms of Payment\n15 DAYS\n\nOther References\n\nBuyer's Order No Dated\n13602313947 die. 17 082022, 1-Oct-22\nDispatch Doc No : Delvary Notes Date\nDispatched through Destination\nVessal/Flight No Place of receipt by shipper\nCity/Port of Laadmng City/Port of Discharge\nTerms of Delvery\n15 NOV 2022\nHSNY/SAC Quant ty Rate per Amount\n64769000 100.00 MT B3250 MT 83,250.00\n84169000 10NOS 13,96750 NOS 1,39,675.09\n84169000 2NOS 24,040.00 NOS 63,080.00\n2.91 005 00\n938549 2 % 5,820.10\n18 % 53,428 52\n3,50,253.62\nE&OE\n\nConipany s Bank Deta''s\n\nBan Name\n\nAic No\n\nBranch & \FS Code\nfor\n\nThus is a Computer Generateu Invaic2\n\n4010 ICICi Bank CA AIC No 215205003976\n215205003076\nBhosari & ICICO0021\n\n \n\nBloom\n\n \n\x0c'
  • "b'Invoice\n\nNASFUND LTD Invoice date\nP. O Box 5791 2023-07-14\nBoroko\n\nDue date\n\nNational Capital District\n\nPapua New Guinea 2028-07-14\n\n \n\npropertypng....\n\nProperty PNG Limited 01062023 (2023-06-08) (2023-06-13) (2023-06-21)\n\nP.O. Box 1067\nBOROKO, NCD\n\nPapua New Guinea Cee\n\nou\n\n \n\n \n\n \n\n \n\n \n\nInvoice number COGExernpt BY. ey (\nAttention: Mr. Fiona Saascee COI 1-72109 TIN 501080230 Cae :
1
  • "GSTIN: 21AAACH1201R1ZZ\nHINDALCO INDUSTRIES LIMITED ou\n\n \n\n \n\n \n\n \n\n \n\n \n\n \n\n \n\n \n\n \n\n \n\n- HIRAKUD POWER, HIRAKUD-768 016,DIST.:SAMBALPUR (ODISHA) GST Rangeltivision: Sambalpur\n= PHONE: 0663-2481965, FAX: 0663-2481342 GST Commissionerate -Cuttack\n~~ PURCHASE ORDER\n‘Vendor Code: S065 P.O/No: =P/PO/SRV/1819/0759 pee Zeon\nMis SK. MANSUR ALI ee\n‘Order Type: PURCHASE ORDER:\nNEAR GURUDWARA PADA Effective From 01/01/2019 To 31/12/2019\nHIRAKUD Price Basis :\n‘Transportation arrangement :\n‘SAMBALPUR, ORISSA, IN 768016 lee ens oar\nEmail: sk.mansur.ai@73@gmail.com Gamer :\nFax: () Currency 2 INR\nContact: MANSOOR ALi () 9776531785 Hindalco Contact Person : AMARESH MISHRA,\n+ GSTIN:21ALGPASG7SE1ZJ —State:21 - Odisha Email of Contact Person: materialshkdpower@adityabirta.com\n(Order Unit of Value\n‘Quantity Measurement (Rs)\n330.000 MDY 1098170.70\n98835.36\n98835.96\na qaese4t.42\nYramentTem: iSDesGem 8 : GRAND TOTAL: 1295841.42\nBiling Address: Hindalco Industries Ltd, Hirakud Power Hiekus - 76804 coats mercer Rene\n{Hard copy of bills to be forwarded at shipping address]\nShipping address : Hindalco Industries Ltd, Hirakud Power, Hirakud ~ T65016, Sombalpur (ODISHA)\n_ Noles:\n4. THE ABOVE PRICE IS BASED ON ATTACHED LTS CALCULATION.\n2. AS PER LTS, WAGES WILL BE PAID FOR NATIONAL HOLIDAYS & FESTIVAL INCASE A LABOUR IS REQUIRED TO PERFORM\nHIS HOB ON A HOLIDAY MENTIONED ‘AS PER LAW THE CALCULATION FOR THAT DAY WILL BE APPLICABLE OVER AND ABOVE THE WAGES\n\nHE IS BEING PAID (BASIC+ DA X 2+ 1 HRA+ 1 DUST ALLOWANCE + 1 CANTEEN ALLOWANCE). CONTRACTOR PROFIT WILL BE @9.5% & FIX MARGIN.\n3, SHIFT ALLOWANCE WILL BE PAID EXTRA AS PER LTS.\n\n4. PRO-RATA DEDUCTION WILL BE MADE FOR ABSENTISM AND DETERIORATION IN QUALITY OF SERVICE AS DECIDED BY THE AUTHORITY.\n\n5. PAYMENT ON MONTHLY BASIS WILL BE RELEASED WITHIN 15 DAYS OF SUBMISSION OF INVOICE.\n\n6. 10% OF TOTAL BILL VALUE WILL BE RETAINED BY THE COMPANY EVERY MONTH AND WiLL BE RELEASED ANNUALLY ONLY AFTER PAYING\nBONUS & OTHER DUES TO THE LABOURERS ENGAGED BY YOU.\n\n7. IT 1S MANDATORY TO PUT GST NUMBER IN THE INVOICE.\n\nOTHER TERMS :-\n1. JOB SHALL BE EXECUTED AS PER THE SCHEDULE ISSUED BY THE DEPARTMENT ( TO BE COLLECTED IN ADVANCE).\n2. YOU / CONTRACTOR'S SUPERVISOR SHALL MEET THE DEPARTMENTAL SUPERVISOR AT REGULAR INTERVAL DECIDED BY THE DEPARTMENT TO\n\nCONFIRM CHANGE IN SCHEDULE, IF ANY.\n\n3. YOU SHALL SUBMIT THE DAILY REPORT AFTER COMPLETION OF JOS IN THE PRESCRIBED FORMAT AS DIRECTED BY THE DEPARTMENT.\n\n4, NO EXTRA CLAIM TO BE MADE FOR CHANGE IN SCHEDULE FOR SIMILAR TYPE OF JOB.\n\n§. REFUSAL OF ANY JOB SIMILAR TO THE JOB ENLISTED IN THE ORDER OR FAILURE TO EXECUTE THE JOB AS PER SCHEDULE WILL BE LIAGLE FOR\n\nAND SHALL KEEP THE SUPPORTING DOCUMENT WHICH WILL BE VERIFIED BY HR FROM TIME TO TIME.\n\n8. CLEANING OF AREAS WITHOUT CONSUMABLE MATERIAL WILL BE TREATED AS NO CLEANING AND AMOUNT RELATED TO THAT JOB WILL BE\n DEDUCTED FOR THAT PARTICULAR OCCASION.\n\n9. MANPOWER SHOULD BE DEPLOYED AS PER REQUIREMENT OF THE MANAGEMENT.\n\n~ NOTE:\n4.NH & PH WILL BE APPICABLE AS PER THE LTS LAW AND PF,ESIC.BONUS, CONTRACTOR PROFIT (9.5%) WILL ATTRACT.\n\nRegd. Office:\nHINDALCO INDUSTRIES LIMITED Page 1 of 2\nAhura Centre, 1st Floor, ' Wing, Mahskali Caves Road Andhori(East), Mumbai 400033, In India,\nTel No: 91 22 6691 7000, Email: hindalco@adityabiria.com, Website: www.adityabiria.com\n\nIdentity No: L27020MH1958PLC011238\n\n4\ o\n\x0c"
  • ')\n\nHINDALCO INDUSTRIES LIMITED\n\nHIRAKUD POWER, HIRAKUD-768 016,DIST :SAMBALPUR (ODISHA)\n
2
  • "b'\xe2\x80\x9cnasfund &&\n\nMemorandum\nTo: Accounts Payables Office\nDate: 15 February 2024\nFrom: Niasul. K. Lillie\nRe: Binding Clear Cover & Hard Cover for HQ\n\n \n\n \n\n \n\n \n\n \n\nPAYEE Theodist Ltd\n\nPURPOSE Supply Binding Clear Cover & Hard Cover for NSF HQ (NIL STOCK on Clear\nCover)\n\nAMOUNT K1,290.01\n\nMethod of Purchase Order\n\nPayment\n\n \n\n \n\n \n\n \n\n

Evaluation

Metrics

Label Accuracy
all 1.0

Uses

Direct Use for Inference

First install the SetFit library:

pip install setfit

Then you can load this model and run inference.

from setfit import SetFitModel

# Download from the 🤗 Hub
model = SetFitModel.from_pretrained("Gopal2002/NASFUND_MODEL_V1")
# Run inference
preds = model("| Sth. Mansur i ves cere

09861521380
At- Hirakud, Dist- Sambalpur, Odisha-768016
{ORIGINAL FOR RECEIPIENT)

TAX INVOICE
ISSUED UNDER RULE 46 OF CGST / OGST RULES, 2017) _

wvorceno: SX 0-219 /19

 
  
  
 

  
 

DETAILS OF RECEIPIENTS (BILLED TO)
HINDALCO INDUSTRIES LIMITED
HIRAKUD POWER
AT/PO- HIRAKUD,
DIST- SAMBALPUR-768016, ODISHA

 
  
  
   
 

  
    
   

INVOICEDATE: | ¥~ G'

PO NO. : ejeo| seupres] oF € J

  
    
   

  
 

GSTIN : 21AAACH1201R17Z
PAN No. : AAACH1201R
DETAILS OF CONSIGNEE (SHIPPED TO)
HINDALCO INDUSTRIES LIMITED

   
    
  
 

      
 

PODATE: 9Q~-'-'%

 
   

DATE OF ISSUE :

      

PERIOD OF SERVICE: 14-$°\"% fo 16-64-19

 
 
 

  
 

af uh Dea VC cy )* OR gre

b)L Om egy cd.

 
 
 

    

SG0004
Se eS

Sea wee. of a Ck.fwd pote. ce. ovev
| x

aged of Rok. Ae gers € eAv cdl

 
 

  

 

   
   

Total Invoice value (In Figures)
Total Invoice value (in words) —

  
  

 

       
 
 

“T wralisa $2.93 §acKs a f : “4 of Fy as nS a @& @er\. | wai

 

CERTIFIED THAT THE PARTICULARS GIVEN ABOVE ARE TRUE AND CORRECT AND THE AMOUNT INDICATED REPRESENTS THE PRICE ACTUALLY
CHARGED BY US AND IS NO FLOW OS ADDITIAL CONDITION DIRECTLY OR INDIRECTLY FROM THE BUYER.

 

 
 
   
 

Our GSTIN : 21ALGPA5675EIZI

 

Our PAN: ALGPAS675E
")

Training Details

Training Set Metrics

Training set Min Median Max
Word count 2 274.2711 699
Label Training Sample Count
0 53
1 53
2 5
3 55

Training Hyperparameters

  • batch_size: (32, 32)
  • num_epochs: (2, 2)
  • max_steps: -1
  • sampling_strategy: oversampling
  • body_learning_rate: (2e-05, 1e-05)
  • head_learning_rate: 0.01
  • loss: CosineSimilarityLoss
  • distance_metric: cosine_distance
  • margin: 0.25
  • end_to_end: False
  • use_amp: False
  • warmup_proportion: 0.1
  • seed: 42
  • eval_max_steps: -1
  • load_best_model_at_end: False

Training Results

Epoch Step Training Loss Validation Loss
0.0017 1 0.2889 -
0.0846 50 0.2092 -
0.1692 100 0.0499 -
0.2538 150 0.0143 -
0.3384 200 0.0113 -
0.4230 250 0.0023 -
0.5076 300 0.0027 -
0.5922 350 0.0025 -
0.6768 400 0.0012 -
0.7614 450 0.0012 -
0.8460 500 0.001 -
0.9306 550 0.0014 -
1.0152 600 0.0011 -
1.0998 650 0.0013 -
1.1844 700 0.0012 -
1.2690 750 0.0008 -
1.3536 800 0.0008 -
1.4382 850 0.0008 -
1.5228 900 0.0008 -
1.6074 950 0.0008 -
1.6920 1000 0.0007 -
1.7766 1050 0.0007 -
1.8613 1100 0.0008 -
1.9459 1150 0.0007 -

Framework Versions

  • Python: 3.10.12
  • SetFit: 1.0.3
  • Sentence Transformers: 2.7.0
  • Transformers: 4.40.1
  • PyTorch: 2.2.1+cu121
  • Datasets: 2.19.0
  • Tokenizers: 0.19.1

Citation

BibTeX

@article{https://doi.org/10.48550/arxiv.2209.11055,
    doi = {10.48550/ARXIV.2209.11055},
    url = {https://arxiv.org/abs/2209.11055},
    author = {Tunstall, Lewis and Reimers, Nils and Jo, Unso Eun Seo and Bates, Luke and Korat, Daniel and Wasserblat, Moshe and Pereg, Oren},
    keywords = {Computation and Language (cs.CL), FOS: Computer and information sciences, FOS: Computer and information sciences},
    title = {Efficient Few-Shot Learning Without Prompts},
    publisher = {arXiv},
    year = {2022},
    copyright = {Creative Commons Attribution 4.0 International}
}
Downloads last month
3
Safetensors
Model size
33.4M params
Tensor type
F32
·

Finetuned from

Evaluation results