SetFit with intfloat/multilingual-e5-large

This is a SetFit model that can be used for Text Classification. This SetFit model uses intfloat/multilingual-e5-large 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:

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
Sentence Transformer body: intfloat/multilingual-e5-large
Classification head: a LogisticRegression instance
Maximum Sequence Length: 512 tokens
Number of Classes: 12 classes

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
6	'Are there any major whitespace opportunity in terms of Categories x Pack Segments in Cuernavaca?' 'In Colas MS which packsegment is not dominated by KOF in TT HM Orizaba 2022? At what price point we can launch an offering' 'I want to launch a new pack type in csd for kof. Tell me what'
2	"Do any seasonal patterns exist in Jumex's share change in Orizaba?" 'What is the Market share for Resto in colas MS at each size groups in TT HM Orizaba in 2022' 'Which categories have seen the some of the highest Share losses for KOF in Cuernavaca in FY22-21?'
0	'Which packs have driven the shares for the competition in Colas in FY 21-22?' 'Apart from Jugos + NÃ©ctares, Which are the top contributing categoriesXconsumo to the share loss for Jumex in Orizaba in 2021?' 'which pack segment is contributing most to share change for Resto in Orizaba NCBs in 2022'
10	'Which pack segment shows opportunities to drive my market share in NCBS Colas SS?' 'What are my priority pack segments to gain share in NCB Colas SS?' 'What are my priority pack segments to gain share in AGUA Colas SS?'
5	'Where should I play in terms\xa0of flavor in Sabores SS?' 'I want to launch flavored water in onion flavor for kof.' 'What areas should I focus on to grow my market presence?'
7	'Is Fanta a premium brand? How premium are its offerings as compared to other brands in Sabores?' "Is there potential for PPL correction in the packaging and pricing strategy of Tropicana's fruit juice offerings within the Juice category?" 'Is there an opportunity to premiumize any offerings for coca-cola?'
9	'Which industries to prioritize to gain share in AGUA in Cuernavaca?' 'What measures can be taken to maximize headroom in the AGUA market?' 'How much headroom do I have in CSDS'
11	'How can I gain share in NCBS?' 'How should KOF gain share in Colas MS in Cuernavaca? ' 'How can I gain share in CSD Colas MS in Cuernavaca'
8	'Category wise market share' 'What is the ND, WD of KOF in colas' 'Tell me the top 10 SKUs in colas'
3	'What is the difference in offerings for KOF vs the key competitors in xx price bracket within CSD Colas in TT HM?' 'How should KOF gain share in <10 price bracket for NCB in TT HM' 'Which price points to play in?'
1	'what factors contributed to share change for agua?' 'Why is Resto losing share in Cuernavaca Colas SS RET Original?' 'What are the main factors contributing to the share gain of Jumex in Still Drinks MS in Orizaba for FY 2022?'
4	'How has the csd industry evolved in the last two years?' 'Tell me the categories to focus on, for driving growth in future' 'What is the change in industry mix for coca-cola in TT HM Orizaba in 2021 to 2022'

Evaluation

Metrics

Label	Accuracy
all	0.25

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("vgarg/fw_identification_model_e5_large_v5_14_02_24")
# Run inference
preds = model("Why is KOF losing share in Cuernavaca Colas MS RET Original?")

Training Details

Training Set Metrics

Training set	Min	Median	Max
Word count	4	13.5351	28

Label	Training Sample Count
0	10
1	10
2	10
3	8
4	10
5	10
6	10
7	10
8	10
9	10
10	10
11	6

Training Hyperparameters

batch_size: (16, 16)
num_epochs: (3, 3)
max_steps: -1
sampling_strategy: oversampling
num_iterations: 20
body_learning_rate: (2e-05, 2e-05)
head_learning_rate: 2e-05
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
0.0035	1	0.3481	-
0.1754	50	0.1442	-
0.3509	100	0.091	-
0.5263	150	0.0089	-
0.7018	200	0.0038	-
0.8772	250	0.0018	-
1.0526	300	0.001	-
1.2281	350	0.0012	-
1.4035	400	0.0007	-
1.5789	450	0.0007	-
1.7544	500	0.0004	-
1.9298	550	0.0005	-
2.1053	600	0.0006	-
2.2807	650	0.0005	-
2.4561	700	0.0006	-
2.6316	750	0.0004	-
2.8070	800	0.0004	-
2.9825	850	0.0004	-

Framework Versions

Python: 3.10.12
SetFit: 1.0.3
Sentence Transformers: 2.3.1
Transformers: 4.35.2
PyTorch: 2.1.0+cu121
Datasets: 2.17.0
Tokenizers: 0.15.1

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}
}

vgarg
/

fw_identification_model_e5_large_v5_14_02_24