metadata
tags:
- setfit
- sentence-transformers
- text-classification
- generated_from_setfit_trainer
widget:
- text: >-
hasAdditionalInformation: 8pg Self Cover Brochure Flat 17" x 8.5"
Accordion Folded 4.25" x 8.5" Ink - 4/4, 4CP + OA Satin AQ Stock - 100#
Gloss Text Finishing - Score, Trim, Accordion Fold, Shrink Wrap in 25's,
hasArtworkDoubleSidedStatus: Double Sided Different, hasCoatingOrSealer:
Aqueous, hasColourDetails: Prints 4/4, 4CP + OA Satin AQ, hasCreatedDate:
2024-04-12, hasCustomerHomeCountry: United States, hasCustomerID: 19754,
hasCustomerName: Kenvue, Inc.(Kenvue - Skillman), hasCutting: Trim to
size, hasElementID: 3229443, hasElementTitle: NTG General Sun Brochure
2024, hasFinishedSizeHeight: 8.5, hasFinishedSizeWidth: 4.25,
hasFlatSizeHeight: 8.5, hasFlatSizeWidth: 17, hasFscPaperBeenSpecified:
No, hasInternalID: b7dd2ed8-0a5f-4e57-8d8a-4aa41afa9d84,
hasMachineFinishing: Yes, hasMachineFinishingDetails: Score, Trim,
Accordion Fold, Shrink Wrap in 25's, hasMaterialCategory: Paper,
hasMaterialDescription: 100# Gloss Text, hasMaterialRecycledPercentage:
0%, hasMaterialThicknessOrWeight: 100, hasMaterialType: Paper and board,
hasMaterialUnitOfMeasure: GSM, hasNumberOfVersions: 1,
hasPackingRequirements: Delivers to Encompass Elements, hasPrice: 45 USD,
hasPrintedSides: Double sided, hasProductCategory: Booklets & Brochures,
hasProofType: PDF digital proof, hasQuantity: 200000,
hasRecycledContentBeenOffered: N/A, hasSendToDetails: PDF PROOF to Rick
Stoy ( [email protected] ), hasSupplierName: HH IC Preflight Fees(HH
IC Preflight Fees), hasTotalColours: 4, hasTotalColoursFace: 4,
hasUnitOfMeasure: Millimetres (mm),
- text: >-
hasArtworkDoubleSidedStatus: Double Sided Different, hasCreatedDate:
2024-01-05, hasCustomerHomeCountry: United States, hasCustomerID: 32045,
hasCustomerName: Rent-A-Center, Inc.(Rent-A-Center), hasCutting: Trim to
size, hasElementID: 3046855, hasElementTitle: Loyalty Postcards - 4
Versions , hasFinishedSizeHeight: 4.5, hasFinishedSizeWidth: 6,
hasFlatSizeHeight: 4.5, hasFlatSizeWidth: 6, hasFscPaperBeenSpecified: No,
hasInternalID: 7bad5a0e-1e7b-4a1a-94e6-97e6ccd30055, hasMaterialCategory:
Paper, hasMaterialDescription: 80# Dull Cover,
hasMaterialRecycledPercentage: 0%, hasMaterialThicknessOrWeight: 80,
hasMaterialType: Paper and board, hasMaterialUnitOfMeasure: Pounds (lbs),
hasNumberOfVersions: 4, hasPrice: 156.3 USD, hasPrintedSides: Double
sided, hasProductCategory: Loose Print, hasProofType: PDF digital proof,
hasQuantity: 48, hasRecycledContentBeenOffered: N/A, hasSupplierName:
American Printing & Mailing(American Printing & Mailing - HHGSP PI),
hasTotalColours: 4, hasUnitOfMeasure: Inches (in),
- text: >-
hasAdditionalInformation: QTY – 250 + Overs Size Flat 2,625" x 2.375"
Folded 2.625" x 1.1875" Prints 1/1 Stock - 80# Gloss Text Finishing –
Die cut to shape ( round corners ) Apply strip of fugitive glue on Blank
Back Panel, Fold and Seal with dot of Fugitive Glue,
hasArtworkDoubleSidedStatus: Double Sided Different, hasColourDetails:
1/1, Black, hasCreatedDate: 2024-02-08, hasCustomerHomeCountry: United
States, hasCustomerID: 19754, hasCustomerName: Kenvue, Inc.(Kenvue -
Skillman), hasCutting: Trim to size, hasElementID: 3111960,
hasElementTitle: Project Pismo NTG Ultra Sheer Face Liquid Folded Panel
FILLED , hasFinishedSizeHeight: 1.1875, hasFinishedSizeWidth: 2.625,
hasFlatSizeHeight: 2.375, hasFlatSizeWidth: 2.625,
hasFscPaperBeenSpecified: No, hasInternalID:
3948f154-5211-4628-b0cf-ac69cd6597d1, hasMachineFinishing: Yes,
hasMachineFinishingDetails: Die cut folding panel and fold in half. seal
closed with 1 removable low tack glue dot. Apply kleenstick strip with
liner to back panel leaving liner in tact for application,
hasMaterialCategory: Paper, hasMaterialDescription: 80# Gloss Text,
hasMaterialRecycledPercentage: 0%, hasMaterialThicknessOrWeight: 80,
hasMaterialType: Paper and board, hasMaterialUnitOfMeasure: GSM,
hasNumberOfVersions: 1, hasPackingRequirements: Deliver to Skillman,
hasPrice: 1390 USD, hasPrintedSides: Double sided, hasProductCategory:
Booklets & Brochures, hasProofType: PDF digital proof, hasQuantity: 250,
hasQuantityPerVersion: 250, hasRecycledContentBeenOffered: No,
hasSendToDetails: Send PDF PROOF to Shawn Beury ( [email protected] ),
hasSupplierName: Hatteras Press(Hatteras Press -HHGSP - PI),
hasTotalColours: 1, hasUnitOfMeasure: Inches (in),
- text: >-
hasAdditionalInformation: 2,400 packs of 25 of Trifold brochure,
hasColourDetails: 4/4 + Bleed, hasCreatedDate: 2024-09-20,
hasCustomerHomeCountry: United States, hasCustomerID: 26760,
hasCustomerName: Elanco Animal Health(Elanco Animal Health), hasCutting:
Trim to size, hasElementID: 3522746, hasElementTitle: EM-US-20-0069, Dog
Health Brochure Reprint, hasFinishedSizeHeight: 8.5, hasFinishedSizeWidth:
3.67, hasFlatSizeHeight: 8.5, hasFlatSizeWidth: 11,
hasFscPaperBeenSpecified: No, hasInternalID:
bab8cdea-2c32-4de2-b783-3616742e6533, hasMachineFinishing: Yes,
hasMachineFinishingDetails: Trim, Score, Fold & s/w in packs of 25,
hasMaterialCategory: Paper, hasMaterialDescription: Dull Cover,
hasMaterialRecycledPercentage: 0%, hasMaterialThicknessOrWeight: 80,
hasMaterialType: Paper and board, hasMaterialUnitOfMeasure: Pounds (lbs),
hasNumberOfVersions: 1, hasPackingRequirements: Carton pack s/w packs of
25 - Label Cartons EM-US-20-0069, Dog Health Brochure with item barcode
and Qty per carton - & Ship to DFS, hasPrice: 2045 USD, hasPrintedSides:
Double sided, hasProductCategory: Booklets & Brochures, hasProofType: PDF
digital proof, hasQuantity: 5750, hasRecycledContentBeenOffered: N/A,
hasSupplierName: Modern Litho – Kansas City(Modernlitho -James Printing.
Inc - HHGSP), hasTotalColours: 4, hasTotalColoursFace: 4,
hasUnitOfMeasure: Inches (in),
- text: >-
hasAdditionalInformation: 299mm x 200mm sheet size. Kiss cut magnets -
Printed 1 colour onto 1mm white magnetic sheets, CAD and kiss cut - see
cutter guide. Layout - 13 deep and 20 wide @ 23mm x 10mm,
hasColourDetails: Digitally printed 1 colour black to face of 1mm white
magnetic sheet, hasCreatedDate: 2024-05-17, hasCustomerHomeCountry: United
Kingdom, hasCustomerID: 37488, hasCustomerName: Imperial Tobacco Ltd
UK(Imperial Tobacco Ltd UK), hasCutting: Cut to shape, hasElementID:
3293613, hasElementTitle: HHG0036F Generic Magnetic Price List Numerals,
hasFinishedSizeHeight: 299, hasFinishedSizeWidth: 200,
hasFscPaperBeenSpecified: No, hasInternalID:
99124112-8ee2-44d4-9170-701acadadf62, hasMachineFinishing: Yes,
hasMachineFinishingDetails: Cad and kiss cut, hasMaterialCategory:
Plastic, hasMaterialDescription: 1mm white Isotropic ferrite powder with
PVC binder, hasMaterialRecycledPercentage: 5%,
hasMaterialThicknessOrWeight: 1, hasMaterialType: PVC,
hasMaterialUnitOfMeasure: Millimetres (mm), hasNumberOfVersions: 1,
hasPackingRequirements: Collate into sets with back boards and SKU magnets
IF required, otherwise box in 50s, hasPressWetProofQuantity: 1, hasPrice:
2100 GBP, hasPrintedSides: Single sided, hasProductCategory:
Indoor/Outdoor Signage, hasProofType: Press / wet proof, hasQuantity: 350,
hasRecycledContentBeenOffered: No, hasSendToDetails: FAO Claire Jackway
Imperial Brands Bristol., hasSupplierName: RDD Europe Ltd(RDD Europe Ltd),
hasTotalColours: 1, hasUnitOfMeasure: Millimetres (mm),
metrics:
- f1_micro
- f1_macro
- f1_weighted
- precision
- accuracy
- recall
pipeline_tag: text-classification
library_name: setfit
inference: false
model-index:
- name: SetFit
results:
- task:
type: text-classification
name: Text Classification
dataset:
name: Northell/ros-classifiers-materials-flat
type: unknown
split: test
metrics:
- type: f1_micro
value: 0
name: F1_Micro
- type: f1_macro
value: 0
name: F1_Macro
- type: f1_weighted
value: 0
name: F1_Weighted
- type: precision
value: .nan
name: Precision
- type: accuracy
value: 0.9772727489471436
name: Accuracy
- type: recall
value: 0
name: Recall
SetFit
This is a SetFit model that can be used for Text Classification. A OneVsRestClassifier instance is used for classification.
The model has been trained using an efficient few-shot learning technique that involves:
- Fine-tuning a Sentence Transformer with contrastive learning.
- Training a classification head with features from the fine-tuned Sentence Transformer.
Model Details
Model Description
- Model Type: SetFit
- Classification head: a OneVsRestClassifier instance
- Maximum Sequence Length: 512 tokens
- Number of Classes: 44 classes
Model Sources
- Repository: SetFit on GitHub
- Paper: Efficient Few-Shot Learning Without Prompts
- Blogpost: SetFit: Efficient Few-Shot Learning Without Prompts
Evaluation
Metrics
| Label | F1_Micro | F1_Macro | F1_Weighted | Precision | Accuracy | Recall |
|---|---|---|---|---|---|---|
| all | 0.0 | 0.0 | 0.0 | nan | 0.9773 | 0.0 |
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("setfit_model_id")
# Run inference
preds = model("hasArtworkDoubleSidedStatus: Double Sided Different, hasCreatedDate: 2024-01-05, hasCustomerHomeCountry: United States, hasCustomerID: 32045, hasCustomerName: Rent-A-Center, Inc.(Rent-A-Center), hasCutting: Trim to size, hasElementID: 3046855, hasElementTitle: Loyalty Postcards - 4 Versions , hasFinishedSizeHeight: 4.5, hasFinishedSizeWidth: 6, hasFlatSizeHeight: 4.5, hasFlatSizeWidth: 6, hasFscPaperBeenSpecified: No, hasInternalID: 7bad5a0e-1e7b-4a1a-94e6-97e6ccd30055, hasMaterialCategory: Paper, hasMaterialDescription: 80# Dull Cover, hasMaterialRecycledPercentage: 0%, hasMaterialThicknessOrWeight: 80, hasMaterialType: Paper and board, hasMaterialUnitOfMeasure: Pounds (lbs), hasNumberOfVersions: 4, hasPrice: 156.3 USD, hasPrintedSides: Double sided, hasProductCategory: Loose Print, hasProofType: PDF digital proof, hasQuantity: 48, hasRecycledContentBeenOffered: N/A, hasSupplierName: American Printing & Mailing(American Printing & Mailing - HHGSP PI), hasTotalColours: 4, hasUnitOfMeasure: Inches (in), ")
Training Details
Training Set Metrics
| Training set | Min | Median | Max |
|---|---|---|---|
| Word count | 77 | 113.9875 | 202 |
Framework Versions
- Python: 3.10.16
- SetFit: 1.1.1
- Sentence Transformers: 3.4.1
- Transformers: 4.49.0
- PyTorch: 2.6.0+cu124
- Datasets: 3.2.0
- Tokenizers: 0.21.0
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}
}