tmp-org/kaufland_v1

SetFit with Alibaba-NLP/gte-multilingual-base

This is a SetFit model trained on the tmp-org/kaufland-0-0 dataset that can be used for Text Classification. This SetFit model uses Alibaba-NLP/gte-multilingual-base 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 Type: SetFit
• Sentence Transformer body: Alibaba-NLP/gte-multilingual-base
• Classification head: a LogisticRegression instance
• Maximum Sequence Length: 8192 tokens
• Number of Classes: 20 classes
• Training Dataset: tmp-org/kaufland-0-0

Model Sources

• Repository: SetFit on GitHub
• Paper: Efficient Few-Shot Learning Without Prompts
• Blogpost: SetFit: Efficient Few-Shot Learning Without Prompts

Model Labels

Label	Examples
Home_Home	'Hallo Kristin (Hallo Kristin) [TextView
Coupons_Filiale	'Akti\u200bvieren () [TextView
Angebote_Kaufland Card Angebote	'KNÜLLER () [TextView
Angebote_Angebote	'Aktuelle Angebote (Aktuelle Angebote) [TextView
Other_Prospekt	' (Seite 2 von 64) [View
Other_Digitale Kassenbons	' (Zurück) [View
Other_Kaufland Card	' (Close button) [View
Online-Marktplatz_Online-Marktplatz	'Profil (Profil) [TextView
Coupons_Partner	'Alle (18) () [TextView
Coupons_Marktplatz	'Coupons (Coupons) [TextView
Profil_Profil	'Profil (Profil) [TextView
Other_Coupon details	' (Zurück) [View
Other_Other	" (Close button) [View
Other_Einkaufsliste	'Hallo Kristin (Hallo Kristin) [TextView
Other_Rezepte	' (Zurück) [ImageButton
Other_Angebot details	' (Zurück) [View
Other_Loading	'\ue916 () [TextView
Other_Treuepunkte	'Sammle Treuepunkte mit jedem Filialeinkauf und jeder Bestellung im Online-Marktplatz. Außerdem kannst durch die Teilnahme am „Glücksrad“ oder an unserem „Geschenk des Tages“ Treuepunkte erhalten. \n\nLös deine verfügbaren Treuepunkte ein, um Coupons zu aktivieren und spar bei deinem nächsten Einkauf online und in deiner Filiale. () [TextView
Online-Marktplatz_Loading	'Online-Marktplatz (Online-Marktplatz) [TextView
Angebote_Loading	'Aktuelle Angebote (Aktuelle Angebote) [TextView

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("tmp-org/kaufland_v1")
# Run inference
preds = model("Zahnzusatzversicherung () [WebView|WebView]")

Training Details

Training Set Metrics

Training set	Min	Median	Max
Word count	3	134.6818	3133

Label	Training Sample Count
Angebote_Angebote	16
Angebote_Kaufland Card Angebote	16
Angebote_Loading	2
Coupons_Filiale	16
Coupons_Marktplatz	16
Coupons_Partner	16
Home_Home	16
Online-Marktplatz_Loading	3
Online-Marktplatz_Online-Marktplatz	15
Other_Angebot details	11
Other_Coupon details	5
Other_Digitale Kassenbons	12
Other_Einkaufsliste	3
Other_Kaufland Card	11
Other_Loading	4
Other_Other	13
Other_Prospekt	16
Other_Rezepte	12
Other_Treuepunkte	4
Profil_Profil	13

Training Hyperparameters

• batch_size: (4, 4)
• num_epochs: (1, 1)
• max_steps: -1
• sampling_strategy: undersampling
• 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
• l2_weight: 0.01
• seed: 4242
• eval_max_steps: -1
• load_best_model_at_end: False

Training Results

Epoch	Step	Training Loss	Validation Loss
0.0013	1	0.2528	-
0.0628	50	0.1965	-
0.1256	100	0.143	-
0.1884	150	0.1238	-
0.2513	200	0.1006	-
0.3141	250	0.0849	-
0.3769	300	0.1106	-
0.4397	350	0.0844	-
0.5025	400	0.0738	-
0.5653	450	0.0642	-
0.6281	500	0.0654	-
0.6910	550	0.0566	-
0.7538	600	0.0702	-
0.8166	650	0.0524	-
0.8794	700	0.0557	-
0.9422	750	0.0519	-

Framework Versions

• Python: 3.12.6
• SetFit: 1.1.2
• Sentence Transformers: 5.2.2
• Transformers: 4.57.1
• PyTorch: 2.10.0+cu128
• Datasets: 3.6.0
• Tokenizers: 0.22.2

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}
}