README.md

metadata

tags:
  - setfit
  - sentence-transformers
  - text-classification
  - generated_from_setfit_trainer
widget:
  - text: I need some icon suggestions for this layout
  - text: Tighten the letter spacing
  - text: Group the logo and title together
  - text: Create a photo of a mountain landscape
  - text: Mirror the logo vertically
metrics:
  - accuracy
pipeline_tag: text-classification
library_name: setfit
inference: true
base_model: nomic-ai/nomic-embed-text-v1.5
model-index:
  - name: SetFit with nomic-ai/nomic-embed-text-v1.5
    results:
      - task:
          type: text-classification
          name: Text Classification
        dataset:
          name: Unknown
          type: unknown
          split: test
        metrics:
          - type: accuracy
            value: 0.29854096520763185
            name: Accuracy

SetFit with nomic-ai/nomic-embed-text-v1.5

This is a SetFit model that can be used for Text Classification. This SetFit model uses nomic-ai/nomic-embed-text-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 Type: SetFit
• Sentence Transformer body: nomic-ai/nomic-embed-text-v1.5
• Classification head: a LogisticRegression instance
• Maximum Sequence Length: 8192 tokens
• Number of Classes: 63 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
0	'Add a corporate presentation background'
1	'I need some icon suggestions for this layout'
2	"Add a heading that says 'Welcome'"
3	'Distribute the shapes across the page'
4	'Add a zoom animation to the logo'
5	'Make everything fade in gradually'
6	'Remove the unwanted text overlay'
7	'Remove the background shape'
8	'What shape options do you have?'
9	'Distribute the buttons around the center image'
10	'Duplicate the page structure'
11	'Duplicate the icon and move it'
12	'Copy the content to the final page'
13	'Fix the letter spacing'
14	'Mirror the logo vertically'
15	'Create a photo of a mountain landscape'
16	'Create a card for a birthday party'
17	'Group the logo and title together'
18	'Move the image to the center'
19	'Apply a vintage filter to the photo'
20	'Find me texture patterns'
21	'Restore the previous color'
22	'Remove the background from the illustration'
23	'Delete the old sign'
24	'Replace the old logo with a new one'
25	'Replace the tagline'
26	'Reset the image adjustments'
27	'Scale the text up'
28	'Resize to LinkedIn post dimensions'
29	'Turn the logo 270 degrees'
30	'Scatter the particles around the text'
31	'Select the footer content'
32	'Change to a light gray background'
33	'Change the blend mode to color burn'
34	'Blur the logo slightly'
35	'Add a double border to the image'
36	'Make the photo more illuminated'
37	'Send the logo to the back'
38	'Increase the image contrast'
39	'Crop to a polygon'
40	'Add a gradient shadow'
41	'Change the background color to green'
42	'Increase the menu font size'
43	'Make the text bold and strikethrough'
44	'Use a decorative font for the logo'
45	'Brighten the light areas'
46	'Set the picture as full background'
47	'Tighten the letter spacing'
48	'Add more line height'
49	'Reduce the opacity of the overlay'
50	'Reduce the paragraph spacing'
51	'Make the colors more vibrant'
52	'Make the shadows deeper'
53	'Increase the image sharpness'
54	'Left align the subtitle'
55	'Create a text container'
56	'Create text in a spiral pattern'
57	'Convert to bullet point format'
58	'Add a colored shadow to the text'
59	'Increase the warm color cast'
60	'I want to upload a new image'
61	'Revert the opacity change'
62	'Break apart the grouped elements'

Evaluation

Metrics

Label	Accuracy
all	0.2985

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("Tighten the letter spacing")

Training Details

Training Set Metrics

Training set	Min	Median	Max
Word count	3	5.2857	8

Label	Training Sample Count
0	1
1	1
2	1
3	1
4	1
5	1
6	1
7	1
8	1
9	1
10	1
11	1
12	1
13	1
14	1
15	1
16	1
17	1
18	1
19	1
20	1
21	1
22	1
23	1
24	1
25	1
26	1
27	1
28	1
29	1
30	1
31	1
32	1
33	1
34	1
35	1
36	1
37	1
38	1
39	1
40	1
41	1
42	1
43	1
44	1
45	1
46	1
47	1
48	1
49	1
50	1
51	1
52	1
53	1
54	1
55	1
56	1
57	1
58	1
59	1
60	1
61	1
62	1

Training Hyperparameters

• batch_size: (64, 64)
• num_epochs: (1, 1)
• 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
• l2_weight: 0.01
• seed: 42
• eval_max_steps: -1
• load_best_model_at_end: False

Training Results

Epoch	Step	Training Loss	Validation Loss
0.0161	1	0.1282	-
0.8065	50	0.0118	-

Framework Versions

• Python: 3.12.11
• SetFit: 1.1.3
• Sentence Transformers: 5.1.0
• Transformers: 4.54.1
• PyTorch: 2.7.1
• Datasets: 4.0.0
• Tokenizers: 0.21.4

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}
}