Hugging Face's logo

faodl
/
model_cca_multilabel_MiniLM-L12-70prop-data-augmented-v02

Text Classification
setfit
Safetensors
sentence-transformers
bert
generated_from_setfit_trainer
text-embeddings-inference
Model card Files
xet
Community
model_cca_multilabel_MiniLM-L12-70prop-data-augmented-v02 / README.md
lgsilvaesilva's picture
lgsilvaesilva
Push model using huggingface_hub.
d4c685a verified
preview code
|
raw
history blame contribute delete
32 kB
metadata 
tags:
  - setfit
  - sentence-transformers
  - text-classification
  - generated_from_setfit_trainer
widget:
  - text: >-
      Extend social protection to informal workers along agrifood value chains
      through portable cash transfers, micro-insurance products, and inclusive
      retirement and disability benefits.
  - text: >-
      strategy for sustainable agriculture and rural development in the
      2021-2030 period has been approved together with with the list of the 41
      national programmes and action plans. the common goal of the strategy is
      of building a commodity-producing agriculture and developing agriculture
      based on local advantages, in the direction of modernity, with high
      productivity, quality, efficiency, sustainability and competitiveness,
      firmly ensuring national food security, making an important contribution
      to socio-economic stability, preventing and combating natural disasters
      and epidemics, protecting the environment, responding to climate change,
      effectively implementing international commitments on reducing greenhouse
      gas emissions; the strategy also aims to improve income, quality of life,
      role and position of people involved in agricultural production; create
      non-agricultural jobs to develop diversified livelihoods, reduce poverty
      sustainably for rural people, ensure equal development opportunities among
      regions; comprehensive and modern rural development associated with the
      process of urbanization, with synchronous infrastructure and social
      services and close to urban areas; the strategy looks to preserve and
      promote the national cultural identity; building green, clean, beautiful
      countryside, security and order are ensured; to develop agriculture and
      rural economy in association with new rural construction in the direction
      of highly efficient ecological agriculture, modern rural areas and
      civilized farmers. specific objectives are, by 2030, the gdp growth rate
      of agro-forestry-fishery will reach an average of 2.5 - 3 percent per
      year, the productivity growth rate of agricultural, forestry and fishery
      workers will reach an average of 5.5 - 6 percent per year. the growth rate
      of the export value of agro-forestry-fishery products will reach an
      average of 5-6 percent per year. fisheries sector is selected as the
      strategic production sector.
  - text: >-
      the uganda npdp works with the two overall critical variables, the
      population and urbanization projections and the national land use balance
      sheet. it provides a basis for integrating the physical and spatial with
      the economic and social issues of national development planning. its core
      elements are the pattern of human settlements, the land uses and natural
      resources for economic activity and the infrastructure networks which
      connect and service them. the strategic orientations are: strategic
      orientation 1: maximizing national economic growth strategic orientation
      2: favoring social and regional equality. strategic orientation 3:
      maximizing supply of agricultural lands. strategic orientation 4:
      maximizing environmental sustainability strategic orientation 5:
      maximizing urbanization and urbanity strategic orientation 6: maximizing
      national and international connectivity.
  - text: >-
      Mechanization investment will be integrated with soil health and water
      management programs, ensuring appropriate machinery selection to avoid
      soil compaction and water wastage.
  - text: >-
      The policy will publish annual import-dependency indices by commodity and
      translate these insights into prioritized investment in domestic
      production, feed safety, and regional trade integration to reduce
      vulnerability and improve food security.
metrics:
  - accuracy
pipeline_tag: text-classification
library_name: setfit
inference: false
base_model: sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2

SetFit with sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2

This is a SetFit model that can be used for Text Classification. This SetFit model uses sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2 as the Sentence Transformer embedding model. A OneVsRestClassifier 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 Sources

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("faodl/model_cca_multilabel_MiniLM-L12-70prop-data-augmented-v02")
# Run inference
preds = model("Mechanization investment will be integrated with soil health and water management programs, ensuring appropriate machinery selection to avoid soil compaction and water wastage.")

Training Details

Training Set Metrics

Training set Min Median Max
Word count 1 55.4334 951

Training Hyperparameters

  • batch_size: (16, 16)
  • num_epochs: (2, 2)
  • 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
  • l2_weight: 0.01
  • seed: 42
  • eval_max_steps: -1
  • load_best_model_at_end: False

Training Results

Epoch Step Training Loss Validation Loss
0.0001 1 0.2114 -
0.0045 50 0.2069 -
0.0091 100 0.2029 -
0.0136 150 0.2025 -
0.0181 200 0.1984 -
0.0226 250 0.1848 -
0.0272 300 0.1784 -
0.0317 350 0.176 -
0.0362 400 0.1743 -
0.0408 450 0.1579 -
0.0453 500 0.149 -
0.0498 550 0.1532 -
0.0543 600 0.1551 -
0.0589 650 0.1483 -
0.0634 700 0.1474 -
0.0679 750 0.1444 -
0.0725 800 0.1363 -
0.0770 850 0.1269 -
0.0815 900 0.1541 -
0.0861 950 0.1256 -
0.0906 1000 0.1457 -
0.0951 1050 0.131 -
0.0996 1100 0.1224 -
0.1042 1150 0.1357 -
0.1087 1200 0.1341 -
0.1132 1250 0.1371 -
0.1178 1300 0.1305 -
0.1223 1350 0.1165 -
0.1268 1400 0.1191 -
0.1313 1450 0.1247 -
0.1359 1500 0.1209 -
0.1404 1550 0.129 -
0.1449 1600 0.1161 -
0.1495 1650 0.1215 -
0.1540 1700 0.1213 -
0.1585 1750 0.1193 -
0.1630 1800 0.1126 -
0.1676 1850 0.1253 -
0.1721 1900 0.1135 -
0.1766 1950 0.1032 -
0.1812 2000 0.0998 -
0.1857 2050 0.116 -
0.1902 2100 0.1088 -
0.1947 2150 0.104 -
0.1993 2200 0.1139 -
0.2038 2250 0.1084 -
0.2083 2300 0.1043 -
0.2129 2350 0.1149 -
0.2174 2400 0.1022 -
0.2219 2450 0.1106 -
0.2264 2500 0.1028 -
0.2310 2550 0.0986 -
0.2355 2600 0.0965 -
0.2400 2650 0.1047 -
0.2446 2700 0.1007 -
0.2491 2750 0.0979 -
0.2536 2800 0.0967 -
0.2582 2850 0.0999 -
0.2627 2900 0.1025 -
0.2672 2950 0.0938 -
0.2717 3000 0.0923 -
0.2763 3050 0.0885 -
0.2808 3100 0.0953 -
0.2853 3150 0.0931 -
0.2899 3200 0.095 -
0.2944 3250 0.0945 -
0.2989 3300 0.0919 -
0.3034 3350 0.0975 -
0.3080 3400 0.0906 -
0.3125 3450 0.0977 -
0.3170 3500 0.0952 -
0.3216 3550 0.0851 -
0.3261 3600 0.0883 -
0.3306 3650 0.0852 -
0.3351 3700 0.082 -
0.3397 3750 0.0901 -
0.3442 3800 0.0778 -
0.3487 3850 0.0819 -
0.3533 3900 0.0804 -
0.3578 3950 0.083 -
0.3623 4000 0.0855 -
0.3668 4050 0.0828 -
0.3714 4100 0.0899 -
0.3759 4150 0.0875 -
0.3804 4200 0.0816 -
0.3850 4250 0.09 -
0.3895 4300 0.0782 -
0.3940 4350 0.0831 -
0.3986 4400 0.0795 -
0.4031 4450 0.0807 -
0.4076 4500 0.0809 -
0.4121 4550 0.0763 -
0.4167 4600 0.08 -
0.4212 4650 0.0731 -
0.4257 4700 0.0759 -
0.4303 4750 0.0758 -
0.4348 4800 0.0791 -
0.4393 4850 0.0731 -
0.4438 4900 0.0774 -
0.4484 4950 0.0781 -
0.4529 5000 0.0783 -
0.4574 5050 0.0852 -
0.4620 5100 0.0771 -
0.4665 5150 0.0813 -
0.4710 5200 0.0795 -
0.4755 5250 0.0725 -
0.4801 5300 0.0751 -
0.4846 5350 0.0756 -
0.4891 5400 0.0715 -
0.4937 5450 0.0643 -
0.4982 5500 0.0675 -
0.5027 5550 0.0769 -
0.5072 5600 0.0761 -
0.5118 5650 0.0739 -
0.5163 5700 0.0716 -
0.5208 5750 0.0706 -
0.5254 5800 0.0719 -
0.5299 5850 0.0721 -
0.5344 5900 0.068 -
0.5389 5950 0.0626 -
0.5435 6000 0.0679 -
0.5480 6050 0.0713 -
0.5525 6100 0.0692 -
0.5571 6150 0.0728 -
0.5616 6200 0.0622 -
0.5661 6250 0.0686 -
0.5707 6300 0.073 -
0.5752 6350 0.0563 -
0.5797 6400 0.0621 -
0.5842 6450 0.0699 -
0.5888 6500 0.0691 -
0.5933 6550 0.0676 -
0.5978 6600 0.0621 -
0.6024 6650 0.0693 -
0.6069 6700 0.0708 -
0.6114 6750 0.0672 -
0.6159 6800 0.0728 -
0.6205 6850 0.0629 -
0.625 6900 0.0694 -
0.6295 6950 0.063 -
0.6341 7000 0.0591 -
0.6386 7050 0.0663 -
0.6431 7100 0.0722 -
0.6476 7150 0.0576 -
0.6522 7200 0.0604 -
0.6567 7250 0.0632 -
0.6612 7300 0.0709 -
0.6658 7350 0.0649 -
0.6703 7400 0.0611 -
0.6748 7450 0.0597 -
0.6793 7500 0.0712 -
0.6839 7550 0.0668 -
0.6884 7600 0.0664 -
0.6929 7650 0.0664 -
0.6975 7700 0.0622 -
0.7020 7750 0.0601 -
0.7065 7800 0.0582 -
0.7111 7850 0.0622 -
0.7156 7900 0.0648 -
0.7201 7950 0.064 -
0.7246 8000 0.0624 -
0.7292 8050 0.0622 -
0.7337 8100 0.0596 -
0.7382 8150 0.0633 -
0.7428 8200 0.0532 -
0.7473 8250 0.0565 -
0.7518 8300 0.0724 -
0.7563 8350 0.0559 -
0.7609 8400 0.064 -
0.7654 8450 0.0603 -
0.7699 8500 0.059 -
0.7745 8550 0.0543 -
0.7790 8600 0.0568 -
0.7835 8650 0.0638 -
0.7880 8700 0.0578 -
0.7926 8750 0.0692 -
0.7971 8800 0.0608 -
0.8016 8850 0.0652 -
0.8062 8900 0.061 -
0.8107 8950 0.0581 -
0.8152 9000 0.0627 -
0.8197 9050 0.0656 -
0.8243 9100 0.0579 -
0.8288 9150 0.0626 -
0.8333 9200 0.0587 -
0.8379 9250 0.0625 -
0.8424 9300 0.051 -
0.8469 9350 0.0553 -
0.8514 9400 0.0507 -
0.8560 9450 0.0521 -
0.8605 9500 0.0548 -
0.8650 9550 0.0536 -
0.8696 9600 0.0517 -
0.8741 9650 0.0569 -
0.8786 9700 0.0572 -
0.8832 9750 0.0553 -
0.8877 9800 0.0567 -
0.8922 9850 0.0594 -
0.8967 9900 0.0598 -
0.9013 9950 0.0548 -
0.9058 10000 0.0563 -
0.9103 10050 0.0466 -
0.9149 10100 0.0561 -
0.9194 10150 0.0533 -
0.9239 10200 0.0569 -
0.9284 10250 0.0484 -
0.9330 10300 0.0563 -
0.9375 10350 0.0597 -
0.9420 10400 0.0556 -
0.9466 10450 0.0542 -
0.9511 10500 0.0528 -
0.9556 10550 0.0652 -
0.9601 10600 0.0541 -
0.9647 10650 0.0581 -
0.9692 10700 0.0529 -
0.9737 10750 0.0497 -
0.9783 10800 0.0591 -
0.9828 10850 0.055 -
0.9873 10900 0.0464 -
0.9918 10950 0.0529 -
0.9964 11000 0.0562 -
1.0009 11050 0.0508 -
1.0054 11100 0.0447 -
1.0100 11150 0.0493 -
1.0145 11200 0.0526 -
1.0190 11250 0.0476 -
1.0236 11300 0.0469 -
1.0281 11350 0.0465 -
1.0326 11400 0.0484 -
1.0371 11450 0.0492 -
1.0417 11500 0.0518 -
1.0462 11550 0.0554 -
1.0507 11600 0.0533 -
1.0553 11650 0.0558 -
1.0598 11700 0.0517 -
1.0643 11750 0.0536 -
1.0688 11800 0.051 -
1.0734 11850 0.0512 -
1.0779 11900 0.0531 -
1.0824 11950 0.0496 -
1.0870 12000 0.0471 -
1.0915 12050 0.0492 -
1.0960 12100 0.0444 -
1.1005 12150 0.0441 -
1.1051 12200 0.053 -
1.1096 12250 0.048 -
1.1141 12300 0.044 -
1.1187 12350 0.0482 -
1.1232 12400 0.0486 -
1.1277 12450 0.0475 -
1.1322 12500 0.0484 -
1.1368 12550 0.0512 -
1.1413 12600 0.0534 -
1.1458 12650 0.0532 -
1.1504 12700 0.0422 -
1.1549 12750 0.0487 -
1.1594 12800 0.0489 -
1.1639 12850 0.0558 -
1.1685 12900 0.0506 -
1.1730 12950 0.0497 -
1.1775 13000 0.0593 -
1.1821 13050 0.0476 -
1.1866 13100 0.0435 -
1.1911 13150 0.0553 -
1.1957 13200 0.0527 -
1.2002 13250 0.0535 -
1.2047 13300 0.0469 -
1.2092 13350 0.054 -
1.2138 13400 0.0476 -
1.2183 13450 0.0507 -
1.2228 13500 0.0497 -
1.2274 13550 0.0519 -
1.2319 13600 0.0444 -
1.2364 13650 0.0516 -
1.2409 13700 0.0494 -
1.2455 13750 0.0425 -
1.25 13800 0.0487 -
1.2545 13850 0.0509 -
1.2591 13900 0.0523 -
1.2636 13950 0.0446 -
1.2681 14000 0.0478 -
1.2726 14050 0.0439 -
1.2772 14100 0.0518 -
1.2817 14150 0.0505 -
1.2862 14200 0.0497 -
1.2908 14250 0.0554 -
1.2953 14300 0.0482 -
1.2998 14350 0.0427 -
1.3043 14400 0.05 -
1.3089 14450 0.0477 -
1.3134 14500 0.0463 -
1.3179 14550 0.0448 -
1.3225 14600 0.045 -
1.3270 14650 0.0437 -
1.3315 14700 0.0467 -
1.3361 14750 0.0435 -
1.3406 14800 0.043 -
1.3451 14850 0.0486 -
1.3496 14900 0.049 -
1.3542 14950 0.0471 -
1.3587 15000 0.0455 -
1.3632 15050 0.0428 -
1.3678 15100 0.0425 -
1.3723 15150 0.0465 -
1.3768 15200 0.0452 -
1.3813 15250 0.0487 -
1.3859 15300 0.045 -
1.3904 15350 0.0431 -
1.3949 15400 0.0512 -
1.3995 15450 0.0411 -
1.4040 15500 0.0495 -
1.4085 15550 0.0406 -
1.4130 15600 0.0445 -
1.4176 15650 0.0496 -
1.4221 15700 0.0497 -
1.4266 15750 0.0466 -
1.4312 15800 0.0479 -
1.4357 15850 0.0448 -
1.4402 15900 0.0453 -
1.4447 15950 0.0486 -
1.4493 16000 0.0519 -
1.4538 16050 0.06 -
1.4583 16100 0.045 -
1.4629 16150 0.0421 -
1.4674 16200 0.0501 -
1.4719 16250 0.0525 -
1.4764 16300 0.0444 -
1.4810 16350 0.0422 -
1.4855 16400 0.0428 -
1.4900 16450 0.0398 -
1.4946 16500 0.0412 -
1.4991 16550 0.0482 -
1.5036 16600 0.046 -
1.5082 16650 0.0502 -
1.5127 16700 0.048 -
1.5172 16750 0.0447 -
1.5217 16800 0.0419 -
1.5263 16850 0.0429 -
1.5308 16900 0.0533 -
1.5353 16950 0.0482 -
1.5399 17000 0.0519 -
1.5444 17050 0.0503 -
1.5489 17100 0.0432 -
1.5534 17150 0.0388 -
1.5580 17200 0.0537 -
1.5625 17250 0.0477 -
1.5670 17300 0.0444 -
1.5716 17350 0.0407 -
1.5761 17400 0.0463 -
1.5806 17450 0.0417 -
1.5851 17500 0.0403 -
1.5897 17550 0.0481 -
1.5942 17600 0.0485 -
1.5987 17650 0.0462 -
1.6033 17700 0.0383 -
1.6078 17750 0.0429 -
1.6123 17800 0.0413 -
1.6168 17850 0.0421 -
1.6214 17900 0.0409 -
1.6259 17950 0.0436 -
1.6304 18000 0.0468 -
1.6350 18050 0.0446 -
1.6395 18100 0.0389 -
1.6440 18150 0.0443 -
1.6486 18200 0.0489 -
1.6531 18250 0.0489 -
1.6576 18300 0.0498 -
1.6621 18350 0.044 -
1.6667 18400 0.0392 -
1.6712 18450 0.0441 -
1.6757 18500 0.0429 -
1.6803 18550 0.0369 -
1.6848 18600 0.0409 -
1.6893 18650 0.0496 -
1.6938 18700 0.052 -
1.6984 18750 0.0377 -
1.7029 18800 0.0403 -
1.7074 18850 0.0473 -
1.7120 18900 0.0474 -
1.7165 18950 0.0447 -
1.7210 19000 0.0498 -
1.7255 19050 0.0427 -
1.7301 19100 0.0454 -
1.7346 19150 0.0478 -
1.7391 19200 0.0438 -
1.7437 19250 0.0378 -
1.7482 19300 0.0442 -
1.7527 19350 0.0453 -
1.7572 19400 0.0407 -
1.7618 19450 0.0487 -
1.7663 19500 0.0408 -
1.7708 19550 0.0444 -
1.7754 19600 0.0426 -
1.7799 19650 0.044 -
1.7844 19700 0.04 -
1.7889 19750 0.0484 -
1.7935 19800 0.0478 -
1.7980 19850 0.0495 -
1.8025 19900 0.037 -
1.8071 19950 0.0472 -
1.8116 20000 0.0411 -
1.8161 20050 0.0368 -
1.8207 20100 0.042 -
1.8252 20150 0.0359 -
1.8297 20200 0.0452 -
1.8342 20250 0.0504 -
1.8388 20300 0.0513 -
1.8433 20350 0.048 -
1.8478 20400 0.0497 -
1.8524 20450 0.0408 -
1.8569 20500 0.0448 -
1.8614 20550 0.0411 -
1.8659 20600 0.0401 -
1.8705 20650 0.0445 -
1.875 20700 0.0468 -
1.8795 20750 0.0488 -
1.8841 20800 0.0401 -
1.8886 20850 0.0408 -
1.8931 20900 0.0384 -
1.8976 20950 0.0496 -
1.9022 21000 0.0436 -
1.9067 21050 0.0434 -
1.9112 21100 0.041 -
1.9158 21150 0.0452 -
1.9203 21200 0.0454 -
1.9248 21250 0.0447 -
1.9293 21300 0.0476 -
1.9339 21350 0.0399 -
1.9384 21400 0.0397 -
1.9429 21450 0.0399 -
1.9475 21500 0.0462 -
1.9520 21550 0.0452 -
1.9565 21600 0.046 -
1.9611 21650 0.0368 -
1.9656 21700 0.043 -
1.9701 21750 0.0413 -
1.9746 21800 0.0459 -
1.9792 21850 0.0442 -
1.9837 21900 0.0381 -
1.9882 21950 0.0411 -
1.9928 22000 0.0519 -
1.9973 22050 0.0445 -

Framework Versions

  • Python: 3.12.12
  • SetFit: 1.1.3
  • Sentence Transformers: 5.1.2
  • Transformers: 4.57.1
  • PyTorch: 2.8.0+cu126
  • Datasets: 4.0.0
  • Tokenizers: 0.22.1

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}
}