linkyeu/intent-routing-modernbert-hpo

SetFit with sentence-transformers/all-mpnet-base-v2

This is a SetFit model that can be used for Text Classification. This SetFit model uses sentence-transformers/all-mpnet-base-v2 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: sentence-transformers/all-mpnet-base-v2
• Classification head: a LogisticRegression instance
• Maximum Sequence Length: 384 tokens
• Number of Classes: 6 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
1	'what can your do with your mcp tool db2 schema' 'is this normal?\nGC(47) Pause Young 234M->89M 12.5ms' 'what units is file_size in? what format is mimetype in'
0	'use mongo and postgres mcps to find test data' 'browser screenshot the error page' 'use the linear mcp to find my high priority issues'
2	'java\npublic void process() {\n if (user != null) {\n // stuff\n }\n}\n\nclean this up' "Duplicate class: 'ChunkCacheMetricsTest'" "python\ndef process(data):\n result = []\n for item in data:\n if item['active']:\n result.append(item['value'])\n return result\n\nuse list comprehension"
3	'review security scan then patch critical issues' 'analyze failing tests, group them, fix each group' '\n47 tests failing\n\ncategorize and fix each category'
5	'model architecture not supported, plan alternative approach' 'architecture diagram for wml' 'before we move on let me ask - can we develop a set of documents that break the implementation down into individual tasks - the first one being a simple core set of function that could deliver a a single starting point. and then add layers of function in subsequent tasks to methodically build out th'
4	'yes the linear mcp is finally connected' 'Usually we use a custom linter config here, not the standard one' 'ok it worked after i restarted the postgres mcp server'

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/best_model")
# Run inference
preds = model("lets fix the issues")

Training Details

Training Set Metrics

Training set	Min	Median	Max
Word count	1	12.1876	125

Label	Training Sample Count
0	43
1	80
2	92
3	56
4	64
5	86

Training Hyperparameters

• batch_size: (8, 8)
• num_epochs: (2, 2)
• max_steps: -1
• sampling_strategy: oversampling
• num_iterations: 10
• body_learning_rate: (1e-05, 1e-05)
• head_learning_rate: 0.001
• loss: CosineSimilarityLoss
• distance_metric: cosine_distance
• margin: 0.25
• end_to_end: False
• use_amp: False
• warmup_proportion: 0.05
• l2_weight: 0.01
• seed: 42
• eval_max_steps: -1
• load_best_model_at_end: True

Training Results

Epoch	Step	Training Loss	Validation Loss
0.0009	1	0.5659	-
0.0199	21	-	0.3639
0.0399	42	-	0.3283
0.0475	50	0.3801	-
0.0598	63	-	0.2766
0.0798	84	-	0.2404
0.0950	100	0.2667	-
0.0997	105	-	0.2395
0.1197	126	-	0.2435
0.1396	147	-	0.2417
0.1425	150	0.2294	-
0.1595	168	-	0.2307
0.1795	189	-	0.2256
0.1899	200	0.2286	-
0.1994	210	-	0.2173
0.2194	231	-	0.2139
0.2374	250	0.1829	-
0.2393	252	-	0.2084
0.2593	273	-	0.2046
0.2792	294	-	0.2073
0.2849	300	0.1566	-
0.2991	315	-	0.2012
0.3191	336	-	0.1938
0.3324	350	0.1502	-
0.3390	357	-	0.1934
0.3590	378	-	0.1972
0.3789	399	-	0.2028
0.3799	400	0.1041	-
0.3989	420	-	0.2001
0.4188	441	-	0.1981
0.4274	450	0.1118	-
0.4387	462	-	0.1863
0.4587	483	-	0.1767
0.4748	500	0.0876	-
0.4786	504	-	0.1746
0.4986	525	-	0.1724
0.5185	546	-	0.1731
0.5223	550	0.0575	-
0.5385	567	-	0.1706
0.5584	588	-	0.1743
0.5698	600	0.0524	-
0.5783	609	-	0.1683
0.5983	630	-	0.1635
0.6173	650	0.0401	-
0.6182	651	-	0.1691
0.6382	672	-	0.1659
0.6581	693	-	0.1625
0.6648	700	0.0214	-
0.6781	714	-	0.1649
0.6980	735	-	0.1577
0.7123	750	0.0164	-
0.7179	756	-	0.1595
0.7379	777	-	0.1589
0.7578	798	-	0.1517
0.7597	800	0.0178	-
0.7778	819	-	0.1563
0.7977	840	-	0.1547
0.8072	850	0.0128	-
0.8177	861	-	0.1531
0.8376	882	-	0.1528
0.8547	900	0.0073	-
0.8575	903	-	0.1524

Framework Versions

• Python: 3.12.12
• SetFit: 1.1.3
• Sentence Transformers: 3.4.1
• Transformers: 4.51.3
• PyTorch: 2.5.1+cu124
• Datasets: 4.8.4
• 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}
}