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:

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: 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' 'Insert a modern icon set for the design' 'Add a mountain landscape background to the page'
1	'Show me some background options for my design' 'I need some icon suggestions for this layout' 'Find me some shape options for this design'
2	'Add a social media handle' 'Insert a promotional message' "Add a heading that says 'Welcome'"
3	'Align the footer elements' 'Align everything to the bottom' 'Center the logo and title'
4	'Make the image slide from left' 'Make the image bounce in' 'Add a fade-in animation to the text'
5	'Make the page rotate in' 'Add a zoom-out effect to the entire page' 'Make everything fade in gradually'
6	'Change the color of the car to red' 'Remove the unwanted text overlay' 'Remove the sunglasses and add regular glasses'
7	'Remove the unnecessary decoration' 'Remove the temporary shape' 'Delete the broken image'
8	'How do I group elements together?' 'What text formatting options are there?' 'What are the available filter effects?'
9	'Place the decorations around the title' 'Distribute the text around the focal point' 'Arrange the icons in a circle around the logo'
10	'Make a copy for editing' 'Create a similar page' 'Make a backup of this page'
11	'Copy the image and make it smaller' 'Duplicate the text and change the color' 'Duplicate the logo and flip it'
12	'Copy the icon to the next slide' 'Duplicate the button to page 5' 'Copy the text to the last page'
13	'Fix the typographic errors' 'Fix the letter spacing' 'Improve the paragraph spacing'
14	'Mirror the logo vertically' 'Mirror the icon horizontally' 'Flip the image right to left'
15	'Make an image of a cozy coffee shop' 'Make an image of a modern office space' 'Make an image of a magical forest'
16	'Create a card for a birthday party' 'Create a dark-themed social media post for tech news' 'Generate an Instagram post for a birthday'
17	'Group the navigation elements' 'Group the header elements' 'Group the decorative elements'
18	'Move the element to the bottom left' 'Move the text to the middle' 'Position the logo in the corner'
19	'Apply a cool blue filter' 'Add a dreamy effect' 'Apply a cinematic filter'
20	'Find me some background images' 'Suggest some visual elements' 'Find me overlay effects'
21	'Redo the filter effect' 'Bring back the deleted element' 'Bring back the deleted shape'
22	'Make the object stand alone' 'Remove the background from the icon' 'Remove the background from the photo'
23	'Remove the distracting element' 'Delete the watermark' 'Remove the old building'
24	'Replace the image with a better one' 'Change the background to a different scene' 'Replace the overlay with a gradient'
25	'Replace the menu items' 'Replace the title' 'Change the button text'
26	'Restore the original brightness' 'Reset the image to its initial state' 'Restore the original colors'
27	'Reduce the shape size' 'Increase the image size' 'Reduce the button size'
28	'Make the page portrait orientation' 'Change to business card size' 'Make the page wider'
29	'Rotate the element 60 degrees' 'Rotate the image clockwise' 'Turn the logo 270 degrees'
30	'Scatter the flowers around the border' 'Distribute the shapes randomly' 'Scatter the stars around the title'
31	'Select the main image' 'Choose the heading text' 'Select the logo element'
32	'Make the background transparent' 'Change to a neutral background' 'Change to a light gray background'
33	'Change the blend mode to exclusion' 'Set the blend mode to soft light' 'Set the blend mode to darken'
34	'Add a depth of field blur' 'Blur the entire image' 'Add a motion blur effect'
35	'Change the border width' 'Add a colored border' 'Add a gradient border'
36	'Brighten the dark areas' 'Make the photo more illuminated' 'Make the image more brilliant'
37	'Put the decoration behind' 'Move the icon to the top' 'Put the shape behind the text'
38	'Make the image more striking' 'Enhance the contrast' 'Increase the visual clarity'
39	'Make the image elliptical' 'Make the image circular' 'Make the image diamond-shaped'
40	'Add a colored drop shadow' 'Create a realistic shadow' 'Create a soft drop shadow'
41	'Fill the shape with orange' 'Change the color to teal' 'Fill the element with brown'
42	'Make the subtitle larger' 'Make the description smaller' 'Make the label smaller'
43	'Make the text italic only' 'Make the text emphasized' 'Make the text bold only'
44	'Use a playful font' 'Use a professional font' 'Change to a modern sans-serif font'
45	'Increase the highlight clarity' 'Increase the highlight intensity' 'Enhance the bright spots'
46	'Make the photo cover the background' 'Set the picture as full background' 'Make the photo cover the entire page'
47	'Spread out the letters in the heading' 'Make the letters closer together in the text' 'Tighten the letter spacing in the caption'
48	'Add more line height' 'Make the text more readable' 'Reduce the line height'
49	'Make the text more transparent' 'Reduce the opacity of the overlay' 'Make the element semi-transparent'
50	'Make the sections closer' 'Make the paragraphs closer together' 'Reduce the paragraph spacing'
51	'Make the colors more vivid' 'Increase the color intensity' 'Increase the color richness'
52	'Darken the shadow regions' 'Darken the shadow tones' 'Enhance the shadow details'
53	'Make the details more sharp' 'Increase the image resolution' 'Increase the detail sharpness'
54	'Center the title text' 'Center the logo text' 'Justify the paragraph text'
55	'Create a text container' 'Create a border around the text' 'Add a gradient background to the text'
56	'Create curved text layout' 'Make the text flow in an arc' 'Make the text follow a path'
57	'Make the text into bullet points' 'Make the text into a list with bullets' 'Create an organized list from the text'
58	'Create a shadow behind the heading' 'Add a dramatic text shadow' 'Add a soft glow effect'
59	'Increase the warm color cast' 'Make the image more yellow' 'Add warm undertones to the photo'
60	'Open the image upload tool' 'Show me the file browser' 'Open the upload window'
61	'Go back to the previous state' 'Undo the text edit' 'Revert the color change'
62	'Ungroup the selected items' 'Break up the grouped objects' 'Ungroup the merged elements'

Evaluation

Metrics

Label	Accuracy
all	0.6240

Uses

Direct Use for Inference

First install the SetFit library:

pip install setfit

Then you can load this model and run inference.

python

from setfit import SetFitModel

# Download from the 🤗 Hub
model = SetFitModel.from_pretrained("setfit_model_id")
# Run inference
preds = model("Position the text at the top")

Training Details

Training Set Metrics

Training set	Min	Median	Max
Word count	3	5.2243	12

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

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: True
use_amp: False
warmup_proportion: 0.1
l2_weight: 0.1
seed: 42
eval_max_steps: -1
load_best_model_at_end: False

Training Results

Epoch	Step	Training Loss	Validation Loss
0.0001	1	0.1774	-
0.0036	50	0.1629	-
0.0073	100	0.1464	-
0.0109	150	0.1147	-
0.0146	200	0.0798	-
0.0182	250	0.0552	-
0.0218	300	0.0391	-
0.0255	350	0.0271	-
0.0291	400	0.0272	-
0.0328	450	0.018	-
0.0364	500	0.015	-
0.0400	550	0.0136	-
0.0437	600	0.012	-
0.0473	650	0.0105	-
0.0510	700	0.0094	-
0.0546	750	0.0087	-
0.0583	800	0.0061	-
0.0619	850	0.0078	-
0.0655	900	0.0068	-
0.0692	950	0.0066	-
0.0728	1000	0.0053	-
0.0765	1050	0.0057	-
0.0801	1100	0.0065	-
0.0837	1150	0.0054	-
0.0874	1200	0.0058	-
0.0910	1250	0.006	-
0.0947	1300	0.0048	-
0.0983	1350	0.0038	-
0.1019	1400	0.0034	-
0.1056	1450	0.0037	-
0.1092	1500	0.006	-
0.1129	1550	0.0047	-
0.1165	1600	0.0042	-
0.1201	1650	0.0038	-
0.1238	1700	0.0036	-
0.1274	1750	0.0049	-
0.1311	1800	0.0019	-
0.1347	1850	0.003	-
0.1384	1900	0.003	-
0.1420	1950	0.0024	-
0.1456	2000	0.0023	-
0.1493	2050	0.002	-
0.1529	2100	0.0033	-
0.1566	2150	0.0032	-
0.1602	2200	0.0048	-
0.1638	2250	0.004	-
0.1675	2300	0.0032	-
0.1711	2350	0.0033	-
0.1748	2400	0.0036	-
0.1784	2450	0.0031	-
0.1820	2500	0.0024	-
0.1857	2550	0.0016	-
0.1893	2600	0.0024	-
0.1930	2650	0.0034	-
0.1966	2700	0.0022	-
0.2002	2750	0.0021	-
0.2039	2800	0.0022	-
0.2075	2850	0.0012	-
0.2112	2900	0.0022	-
0.2148	2950	0.001	-
0.2185	3000	0.0007	-
0.2221	3050	0.0011	-
0.2257	3100	0.0008	-
0.2294	3150	0.0008	-
0.2330	3200	0.0016	-
0.2367	3250	0.0026	-
0.2403	3300	0.0018	-
0.2439	3350	0.0021	-
0.2476	3400	0.001	-
0.2512	3450	0.002	-
0.2549	3500	0.0017	-
0.2585	3550	0.0011	-
0.2621	3600	0.0007	-
0.2658	3650	0.0019	-
0.2694	3700	0.0023	-
0.2731	3750	0.0022	-
0.2767	3800	0.0015	-
0.2803	3850	0.0016	-
0.2840	3900	0.0017	-
0.2876	3950	0.0041	-
0.2913	4000	0.0028	-
0.2949	4050	0.0032	-
0.2986	4100	0.004	-
0.3022	4150	0.0025	-
0.3058	4200	0.0026	-
0.3095	4250	0.0024	-
0.3131	4300	0.0015	-
0.3168	4350	0.0013	-
0.3204	4400	0.0026	-
0.3240	4450	0.0017	-
0.3277	4500	0.0015	-
0.3313	4550	0.0013	-
0.3350	4600	0.0012	-
0.3386	4650	0.0009	-
0.3422	4700	0.0008	-
0.3459	4750	0.0009	-
0.3495	4800	0.0005	-
0.3532	4850	0.0005	-
0.3568	4900	0.001	-
0.3604	4950	0.0005	-
0.3641	5000	0.0003	-
0.3677	5050	0.0011	-
0.3714	5100	0.0006	-
0.3750	5150	0.0007	-
0.3786	5200	0.0006	-
0.3823	5250	0.0007	-
0.3859	5300	0.0005	-
0.3896	5350	0.001	-
0.3932	5400	0.0006	-
0.3969	5450	0.0011	-
0.4005	5500	0.0009	-
0.4041	5550	0.001	-
0.4078	5600	0.001	-
0.4114	5650	0.0011	-
0.4151	5700	0.0007	-
0.4187	5750	0.0008	-
0.4223	5800	0.0009	-
0.4260	5850	0.0004	-
0.4296	5900	0.0007	-
0.4333	5950	0.0005	-
0.4369	6000	0.0011	-
0.4405	6050	0.0007	-
0.4442	6100	0.0007	-
0.4478	6150	0.0003	-
0.4515	6200	0.0004	-
0.4551	6250	0.0006	-
0.4587	6300	0.0003	-
0.4624	6350	0.001	-
0.4660	6400	0.0006	-
0.4697	6450	0.0009	-
0.4733	6500	0.0008	-
0.4770	6550	0.0009	-
0.4806	6600	0.0005	-
0.4842	6650	0.0009	-
0.4879	6700	0.0009	-
0.4915	6750	0.0012	-
0.4952	6800	0.0004	-
0.4988	6850	0.0005	-
0.5024	6900	0.0009	-
0.5061	6950	0.0014	-
0.5097	7000	0.0005	-
0.5134	7050	0.0007	-
0.5170	7100	0.0009	-
0.5206	7150	0.0011	-
0.5243	7200	0.001	-
0.5279	7250	0.0021	-
0.5316	7300	0.0015	-
0.5352	7350	0.001	-
0.5388	7400	0.001	-
0.5425	7450	0.0018	-
0.5461	7500	0.0009	-
0.5498	7550	0.0008	-
0.5534	7600	0.0004	-
0.5571	7650	0.0007	-
0.5607	7700	0.0009	-
0.5643	7750	0.0011	-
0.5680	7800	0.0006	-
0.5716	7850	0.0016	-
0.5753	7900	0.0016	-
0.5789	7950	0.0009	-
0.5825	8000	0.0017	-
0.5862	8050	0.0017	-
0.5898	8100	0.001	-
0.5935	8150	0.001	-
0.5971	8200	0.0005	-
0.6007	8250	0.0008	-
0.6044	8300	0.0003	-
0.6080	8350	0.0005	-
0.6117	8400	0.0006	-
0.6153	8450	0.0006	-
0.6189	8500	0.0007	-
0.6226	8550	0.0006	-
0.6262	8600	0.0007	-
0.6299	8650	0.0008	-
0.6335	8700	0.0005	-
0.6372	8750	0.001	-
0.6408	8800	0.0011	-
0.6444	8850	0.0008	-
0.6481	8900	0.0008	-
0.6517	8950	0.0007	-
0.6554	9000	0.0006	-
0.6590	9050	0.0008	-
0.6626	9100	0.0004	-
0.6663	9150	0.0007	-
0.6699	9200	0.0007	-
0.6736	9250	0.0002	-
0.6772	9300	0.0007	-
0.6808	9350	0.0007	-
0.6845	9400	0.0006	-
0.6881	9450	0.0007	-
0.6918	9500	0.0009	-
0.6954	9550	0.0007	-
0.6990	9600	0.0006	-
0.7027	9650	0.0009	-
0.7063	9700	0.0005	-
0.7100	9750	0.0006	-
0.7136	9800	0.001	-
0.7173	9850	0.0004	-
0.7209	9900	0.0006	-
0.7245	9950	0.0006	-
0.7282	10000	0.0003	-
0.7318	10050	0.0009	-
0.7355	10100	0.0006	-
0.7391	10150	0.0011	-
0.7427	10200	0.0008	-
0.7464	10250	0.0006	-
0.7500	10300	0.0008	-
0.7537	10350	0.0006	-
0.7573	10400	0.0005	-
0.7609	10450	0.0008	-
0.7646	10500	0.0007	-
0.7682	10550	0.0005	-
0.7719	10600	0.0007	-
0.7755	10650	0.0011	-
0.7791	10700	0.0011	-
0.7828	10750	0.0008	-
0.7864	10800	0.0003	-
0.7901	10850	0.0006	-
0.7937	10900	0.0009	-
0.7973	10950	0.0006	-
0.8010	11000	0.0008	-
0.8046	11050	0.0005	-
0.8083	11100	0.0014	-
0.8119	11150	0.0007	-
0.8156	11200	0.0004	-
0.8192	11250	0.001	-
0.8228	11300	0.0005	-
0.8265	11350	0.0003	-
0.8301	11400	0.0005	-
0.8338	11450	0.0003	-
0.8374	11500	0.0004	-
0.8410	11550	0.0006	-
0.8447	11600	0.0006	-
0.8483	11650	0.0006	-
0.8520	11700	0.0005	-
0.8556	11750	0.0008	-
0.8592	11800	0.0009	-
0.8629	11850	0.0007	-
0.8665	11900	0.0012	-
0.8702	11950	0.0003	-
0.8738	12000	0.0006	-
0.8774	12050	0.0007	-
0.8811	12100	0.0008	-
0.8847	12150	0.0003	-
0.8884	12200	0.0006	-
0.8920	12250	0.0006	-
0.8957	12300	0.0004	-
0.8993	12350	0.0005	-
0.9029	12400	0.0005	-
0.9066	12450	0.0006	-
0.9102	12500	0.0004	-
0.9139	12550	0.0005	-
0.9175	12600	0.0003	-
0.9211	12650	0.0004	-
0.9248	12700	0.0005	-
0.9284	12750	0.0006	-
0.9321	12800	0.0004	-
0.9357	12850	0.0005	-
0.9393	12900	0.0005	-
0.9430	12950	0.0011	-
0.9466	13000	0.0004	-
0.9503	13050	0.0007	-
0.9539	13100	0.0005	-
0.9575	13150	0.0006	-
0.9612	13200	0.0005	-
0.9648	13250	0.0007	-
0.9685	13300	0.0007	-
0.9721	13350	0.0004	-
0.9758	13400	0.0005	-
0.9794	13450	0.0005	-
0.9830	13500	0.0004	-
0.9867	13550	0.0006	-
0.9903	13600	0.0004	-
0.9940	13650	0.0007	-
0.9976	13700	0.0007	-

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

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}
}

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

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

weizhou03/nomic-embed-text-v1.5

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