Hugging Face's logo

mann2107
/
BCMPIIRAB_MiniLM_ALLNewV2

Text Classification
setfit
Safetensors
sentence-transformers
bert
generated_from_setfit_trainer
Model card Files Community
Edit model card

SetFit with sentence-transformers/all-MiniLM-L6-v2

This is a SetFit model that can be used for Text Classification. This SetFit model uses sentence-transformers/all-MiniLM-L6-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 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("mann2107/BCMPIIRAB_MiniLM_ALLNewV2")
# Run inference
preds = model("Thank you for your email. Please go ahead and issue. Please invoice in KES")

Training Details

Training Set Metrics

Training set Min Median Max
Word count 1 25.6577 136
Label Training Sample Count
0 24
1 24
2 24
3 24
4 24
5 24
6 24
7 24
8 24
9 24
10 24
11 24
12 24
13 24

Training Hyperparameters

  • batch_size: (8, 8)
  • num_epochs: (5, 5)
  • max_steps: -1
  • sampling_strategy: oversampling
  • num_iterations: 68
  • body_learning_rate: (1.44030579311381e-05, 1.44030579311381e-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
  • max_length: 512
  • seed: 42
  • eval_max_steps: -1
  • load_best_model_at_end: True

Training Results

Epoch Step Training Loss Validation Loss
0.0002 1 0.2917 -
0.0088 50 0.2434 -
0.0175 100 0.2053 -
0.0263 150 0.1789 -
0.0350 200 0.2249 -
0.0438 250 0.1773 -
0.0525 300 0.1648 -
0.0613 350 0.2617 -
0.0700 400 0.1342 -
0.0788 450 0.1064 -
0.0875 500 0.1273 -
0.0963 550 0.1248 -
0.1050 600 0.2013 -
0.1138 650 0.1979 -
0.1225 700 0.1631 -
0.1313 750 0.1079 -
0.1401 800 0.0858 -
0.1488 850 0.0999 -
0.1576 900 0.0638 -
0.1663 950 0.1287 -
0.1751 1000 0.1408 -
0.1838 1050 0.1902 -
0.1926 1100 0.0648 -
0.2013 1150 0.1383 -
0.2101 1200 0.0609 -
0.2188 1250 0.0865 -
0.2276 1300 0.1069 -
0.2363 1350 0.051 -
0.2451 1400 0.0692 -
0.2539 1450 0.123 -
0.2626 1500 0.0758 -
0.2714 1550 0.0835 -
0.2801 1600 0.0523 -
0.2889 1650 0.0946 -
0.2976 1700 0.0445 -
0.3064 1750 0.0248 -
0.3151 1800 0.0373 -
0.3239 1850 0.0248 -
0.3326 1900 0.0446 -
0.3414 1950 0.0142 -
0.3501 2000 0.023 -
0.3589 2050 0.0119 -
0.3676 2100 0.0383 -
0.3764 2150 0.0188 -
0.3852 2200 0.0204 -
0.3939 2250 0.0109 -
0.4027 2300 0.0273 -
0.4114 2350 0.0216 -
0.4202 2400 0.0073 -
0.4289 2450 0.0338 -
0.4377 2500 0.0047 -
0.4464 2550 0.0096 -
0.4552 2600 0.0069 -
0.4639 2650 0.0078 -
0.4727 2700 0.0122 -
0.4814 2750 0.0578 -
0.4902 2800 0.0074 -
0.4989 2850 0.0103 -
0.5077 2900 0.0092 -
0.5165 2950 0.004 -
0.5252 3000 0.0061 -
0.5340 3050 0.0214 -
0.5427 3100 0.0048 -
0.5515 3150 0.0036 -
0.5602 3200 0.0041 -
0.5690 3250 0.0151 -
0.5777 3300 0.0042 -
0.5865 3350 0.0029 -
0.5952 3400 0.0021 -
0.6040 3450 0.0018 -
0.6127 3500 0.0058 -
0.6215 3550 0.0011 -
0.6303 3600 0.0078 -
0.6390 3650 0.0011 -
0.6478 3700 0.0017 -
0.6565 3750 0.0022 -
0.6653 3800 0.0016 -
0.6740 3850 0.002 -
0.6828 3900 0.0023 -
0.6915 3950 0.0011 -
0.7003 4000 0.0012 -
0.7090 4050 0.0007 -
0.7178 4100 0.0021 -
0.7265 4150 0.0019 -
0.7353 4200 0.002 -
0.7440 4250 0.0018 -
0.7528 4300 0.0029 -
0.7616 4350 0.0015 -
0.7703 4400 0.0022 -
0.7791 4450 0.0012 -
0.7878 4500 0.0007 -
0.7966 4550 0.0015 -
0.8053 4600 0.0011 -
0.8141 4650 0.0016 -
0.8228 4700 0.0009 -
0.8316 4750 0.0007 -
0.8403 4800 0.0011 -
0.8491 4850 0.001 -
0.8578 4900 0.0008 -
0.8666 4950 0.0014 -
0.8754 5000 0.0022 -
0.8841 5050 0.0012 -
0.8929 5100 0.0007 -
0.9016 5150 0.0014 -
0.9104 5200 0.0007 -
0.9191 5250 0.0012 -
0.9279 5300 0.0011 -
0.9366 5350 0.0012 -
0.9454 5400 0.0029 -
0.9541 5450 0.001 -
0.9629 5500 0.0011 -
0.9716 5550 0.0004 -
0.9804 5600 0.0009 -
0.9891 5650 0.0004 -
0.9979 5700 0.003 -
1.0 5712 - 0.0459
1.0067 5750 0.0014 -
1.0154 5800 0.0008 -
1.0242 5850 0.0009 -
1.0329 5900 0.0007 -
1.0417 5950 0.0007 -
1.0504 6000 0.0006 -
1.0592 6050 0.0008 -
1.0679 6100 0.0006 -
1.0767 6150 0.0006 -
1.0854 6200 0.0007 -
1.0942 6250 0.0025 -
1.1029 6300 0.0006 -
1.1117 6350 0.0009 -
1.1204 6400 0.0009 -
1.1292 6450 0.0009 -
1.1380 6500 0.0006 -
1.1467 6550 0.0004 -
1.1555 6600 0.0014 -
1.1642 6650 0.0029 -
1.1730 6700 0.0004 -
1.1817 6750 0.0027 -
1.1905 6800 0.0003 -
1.1992 6850 0.0003 -
1.2080 6900 0.0006 -
1.2167 6950 0.0015 -
1.2255 7000 0.0005 -
1.2342 7050 0.0005 -
1.2430 7100 0.0016 -
1.2518 7150 0.0005 -
1.2605 7200 0.0003 -
1.2693 7250 0.0006 -
1.2780 7300 0.0007 -
1.2868 7350 0.0004 -
1.2955 7400 0.0007 -
1.3043 7450 0.0007 -
1.3130 7500 0.0007 -
1.3218 7550 0.0003 -
1.3305 7600 0.0002 -
1.3393 7650 0.0002 -
1.3480 7700 0.0005 -
1.3568 7750 0.0014 -
1.3655 7800 0.0012 -
1.3743 7850 0.0002 -
1.3831 7900 0.0002 -
1.3918 7950 0.0003 -
1.4006 8000 0.0005 -
1.4093 8050 0.0006 -
1.4181 8100 0.0003 -
1.4268 8150 0.0009 -
1.4356 8200 0.0004 -
1.4443 8250 0.0002 -
1.4531 8300 0.0004 -
1.4618 8350 0.0008 -
1.4706 8400 0.0002 -
1.4793 8450 0.0004 -
1.4881 8500 0.0006 -
1.4968 8550 0.0011 -
1.5056 8600 0.0003 -
1.5144 8650 0.0003 -
1.5231 8700 0.0004 -
1.5319 8750 0.0004 -
1.5406 8800 0.0002 -
1.5494 8850 0.0007 -
1.5581 8900 0.0003 -
1.5669 8950 0.0002 -
1.5756 9000 0.0007 -
1.5844 9050 0.0005 -
1.5931 9100 0.0005 -
1.6019 9150 0.0011 -
1.6106 9200 0.0004 -
1.6194 9250 0.0004 -
1.6282 9300 0.0003 -
1.6369 9350 0.0002 -
1.6457 9400 0.0003 -
1.6544 9450 0.0006 -
1.6632 9500 0.0004 -
1.6719 9550 0.0004 -
1.6807 9600 0.0006 -
1.6894 9650 0.0001 -
1.6982 9700 0.0002 -
1.7069 9750 0.0004 -
1.7157 9800 0.0004 -
1.7244 9850 0.0001 -
1.7332 9900 0.0004 -
1.7419 9950 0.0004 -
1.7507 10000 0.0006 -
1.7595 10050 0.0003 -
1.7682 10100 0.0002 -
1.7770 10150 0.0004 -
1.7857 10200 0.0004 -
1.7945 10250 0.0002 -
1.8032 10300 0.0008 -
1.8120 10350 0.0004 -
1.8207 10400 0.0005 -
1.8295 10450 0.0004 -
1.8382 10500 0.0001 -
1.8470 10550 0.0003 -
1.8557 10600 0.0003 -
1.8645 10650 0.0005 -
1.8732 10700 0.0005 -
1.8820 10750 0.0003 -
1.8908 10800 0.0001 -
1.8995 10850 0.0002 -
1.9083 10900 0.0001 -
1.9170 10950 0.0003 -
1.9258 11000 0.0005 -
1.9345 11050 0.0003 -
1.9433 11100 0.0004 -
1.9520 11150 0.0007 -
1.9608 11200 0.0002 -
1.9695 11250 0.0003 -
1.9783 11300 0.0001 -
1.9870 11350 0.0001 -
1.9958 11400 0.0002 -
2.0 11424 - 0.042
2.0046 11450 0.0003 -
2.0133 11500 0.0002 -
2.0221 11550 0.0002 -
2.0308 11600 0.0002 -
2.0396 11650 0.0003 -
2.0483 11700 0.0003 -
2.0571 11750 0.0002 -
2.0658 11800 0.0002 -
2.0746 11850 0.0002 -
2.0833 11900 0.0002 -
2.0921 11950 0.0001 -
2.1008 12000 0.0003 -
2.1096 12050 0.0005 -
2.1183 12100 0.0002 -
2.1271 12150 0.0003 -
2.1359 12200 0.0002 -
2.1446 12250 0.0003 -
2.1534 12300 0.0003 -
2.1621 12350 0.0001 -
2.1709 12400 0.0002 -
2.1796 12450 0.0002 -
2.1884 12500 0.0002 -
2.1971 12550 0.0002 -
2.2059 12600 0.0001 -
2.2146 12650 0.0002 -
2.2234 12700 0.0003 -
2.2321 12750 0.0003 -
2.2409 12800 0.0004 -
2.2496 12850 0.0002 -
2.2584 12900 0.0002 -
2.2672 12950 0.0003 -
2.2759 13000 0.0002 -
2.2847 13050 0.0002 -
2.2934 13100 0.0002 -
2.3022 13150 0.0001 -
2.3109 13200 0.0002 -
2.3197 13250 0.0001 -
2.3284 13300 0.0002 -
2.3372 13350 0.0003 -
2.3459 13400 0.0002 -
2.3547 13450 0.0001 -
2.3634 13500 0.0002 -
2.3722 13550 0.0001 -
2.3810 13600 0.0006 -
2.3897 13650 0.0001 -
2.3985 13700 0.0002 -
2.4072 13750 0.0002 -
2.4160 13800 0.0004 -
2.4247 13850 0.0001 -
2.4335 13900 0.0003 -
2.4422 13950 0.0001 -
2.4510 14000 0.0001 -
2.4597 14050 0.0001 -
2.4685 14100 0.0005 -
2.4772 14150 0.0002 -
2.4860 14200 0.0001 -
2.4947 14250 0.0003 -
2.5035 14300 0.0005 -
2.5123 14350 0.0002 -
2.5210 14400 0.0002 -
2.5298 14450 0.0003 -
2.5385 14500 0.0001 -
2.5473 14550 0.0001 -
2.5560 14600 0.0002 -
2.5648 14650 0.0002 -
2.5735 14700 0.0001 -
2.5823 14750 0.0001 -
2.5910 14800 0.0001 -
2.5998 14850 0.0003 -
2.6085 14900 0.0002 -
2.6173 14950 0.0001 -
2.6261 15000 0.0001 -
2.6348 15050 0.0001 -
2.6436 15100 0.0001 -
2.6523 15150 0.0002 -
2.6611 15200 0.0001 -
2.6698 15250 0.0002 -
2.6786 15300 0.0002 -
2.6873 15350 0.0002 -
2.6961 15400 0.0002 -
2.7048 15450 0.0002 -
2.7136 15500 0.0001 -
2.7223 15550 0.0002 -
2.7311 15600 0.0002 -
2.7398 15650 0.0003 -
2.7486 15700 0.0002 -
2.7574 15750 0.0001 -
2.7661 15800 0.0002 -
2.7749 15850 0.0002 -
2.7836 15900 0.0003 -
2.7924 15950 0.0004 -
2.8011 16000 0.0007 -
2.8099 16050 0.0001 -
2.8186 16100 0.0001 -
2.8274 16150 0.0002 -
2.8361 16200 0.0002 -
2.8449 16250 0.0001 -
2.8536 16300 0.0001 -
2.8624 16350 0.0002 -
2.8711 16400 0.0002 -
2.8799 16450 0.0001 -
2.8887 16500 0.0002 -
2.8974 16550 0.0002 -
2.9062 16600 0.0001 -
2.9149 16650 0.0001 -
2.9237 16700 0.0001 -
2.9324 16750 0.0003 -
2.9412 16800 0.0002 -
2.9499 16850 0.0003 -
2.9587 16900 0.0001 -
2.9674 16950 0.0002 -
2.9762 17000 0.0001 -
2.9849 17050 0.0001 -
2.9937 17100 0.0001 -
3.0 17136 - 0.0419
3.0025 17150 0.0002 -
3.0112 17200 0.0002 -
3.0200 17250 0.0003 -
3.0287 17300 0.0001 -
3.0375 17350 0.0002 -
3.0462 17400 0.0001 -
3.0550 17450 0.0002 -
3.0637 17500 0.0002 -
3.0725 17550 0.0002 -
3.0812 17600 0.0001 -
3.0900 17650 0.0001 -
3.0987 17700 0.0001 -
3.1075 17750 0.0001 -
3.1162 17800 0.0001 -
3.125 17850 0.0001 -
3.1338 17900 0.0002 -
3.1425 17950 0.0001 -
3.1513 18000 0.0003 -
3.1600 18050 0.0001 -
3.1688 18100 0.0001 -
3.1775 18150 0.0001 -
3.1863 18200 0.0002 -
3.1950 18250 0.0002 -
3.2038 18300 0.0001 -
3.2125 18350 0.0001 -
3.2213 18400 0.0001 -
3.2300 18450 0.0002 -
3.2388 18500 0.0001 -
3.2475 18550 0.0002 -
3.2563 18600 0.0001 -
3.2651 18650 0.0002 -
3.2738 18700 0.0001 -
3.2826 18750 0.0001 -
3.2913 18800 0.0001 -
3.3001 18850 0.0001 -
3.3088 18900 0.0003 -
3.3176 18950 0.0002 -
3.3263 19000 0.0001 -
3.3351 19050 0.0003 -
3.3438 19100 0.0001 -
3.3526 19150 0.0001 -
3.3613 19200 0.0001 -
3.3701 19250 0.0001 -
3.3789 19300 0.0001 -
3.3876 19350 0.0002 -
3.3964 19400 0.0001 -
3.4051 19450 0.0001 -
3.4139 19500 0.0001 -
3.4226 19550 0.0001 -
3.4314 19600 0.0001 -
3.4401 19650 0.0001 -
3.4489 19700 0.0002 -
3.4576 19750 0.0001 -
3.4664 19800 0.0001 -
3.4751 19850 0.0001 -
3.4839 19900 0.0001 -
3.4926 19950 0.0001 -
3.5014 20000 0.0001 -
3.5102 20050 0.0002 -
3.5189 20100 0.0003 -
3.5277 20150 0.0001 -
3.5364 20200 0.0002 -
3.5452 20250 0.0001 -
3.5539 20300 0.0001 -
3.5627 20350 0.0001 -
3.5714 20400 0.0004 -
3.5802 20450 0.0001 -
3.5889 20500 0.0001 -
3.5977 20550 0.0001 -
3.6064 20600 0.0002 -
3.6152 20650 0.0001 -
3.6239 20700 0.0001 -
3.6327 20750 0.0 -
3.6415 20800 0.0002 -
3.6502 20850 0.0001 -
3.6590 20900 0.0001 -
3.6677 20950 0.0002 -
3.6765 21000 0.0001 -
3.6852 21050 0.0001 -
3.6940 21100 0.0001 -
3.7027 21150 0.0002 -
3.7115 21200 0.0004 -
3.7202 21250 0.0001 -
3.7290 21300 0.0002 -
3.7377 21350 0.0001 -
3.7465 21400 0.0004 -
3.7553 21450 0.0002 -
3.7640 21500 0.0001 -
3.7728 21550 0.0001 -
3.7815 21600 0.0001 -
3.7903 21650 0.0001 -
3.7990 21700 0.0001 -
3.8078 21750 0.0001 -
3.8165 21800 0.0 -
3.8253 21850 0.0 -
3.8340 21900 0.0001 -
3.8428 21950 0.0003 -
3.8515 22000 0.0001 -
3.8603 22050 0.0001 -
3.8690 22100 0.0002 -
3.8778 22150 0.0001 -
3.8866 22200 0.0003 -
3.8953 22250 0.0001 -
3.9041 22300 0.0 -
3.9128 22350 0.0001 -
3.9216 22400 0.0002 -
3.9303 22450 0.0001 -
3.9391 22500 0.0001 -
3.9478 22550 0.0 -
3.9566 22600 0.0003 -
3.9653 22650 0.0001 -
3.9741 22700 0.0001 -
3.9828 22750 0.0001 -
3.9916 22800 0.0002 -
4.0 22848 - 0.0419
4.0004 22850 0.0 -
4.0091 22900 0.0001 -
4.0179 22950 0.0001 -
4.0266 23000 0.0001 -
4.0354 23050 0.0001 -
4.0441 23100 0.0002 -
4.0529 23150 0.0001 -
4.0616 23200 0.0001 -
4.0704 23250 0.0002 -
4.0791 23300 0.0 -
4.0879 23350 0.0001 -
4.0966 23400 0.0001 -
4.1054 23450 0.0001 -
4.1141 23500 0.0001 -
4.1229 23550 0.0002 -
4.1317 23600 0.0001 -
4.1404 23650 0.0001 -
4.1492 23700 0.0001 -
4.1579 23750 0.0002 -
4.1667 23800 0.0002 -
4.1754 23850 0.0001 -
4.1842 23900 0.0001 -
4.1929 23950 0.0001 -
4.2017 24000 0.0001 -
4.2104 24050 0.0001 -
4.2192 24100 0.0001 -
4.2279 24150 0.0 -
4.2367 24200 0.0001 -
4.2454 24250 0.0001 -
4.2542 24300 0.0003 -
4.2630 24350 0.0 -
4.2717 24400 0.0001 -
4.2805 24450 0.0 -
4.2892 24500 0.0001 -
4.2980 24550 0.0001 -
4.3067 24600 0.0002 -
4.3155 24650 0.0 -
4.3242 24700 0.0001 -
4.3330 24750 0.0001 -
4.3417 24800 0.0001 -
4.3505 24850 0.0001 -
4.3592 24900 0.0001 -
4.3680 24950 0.0 -
4.3768 25000 0.0002 -
4.3855 25050 0.0001 -
4.3943 25100 0.0001 -
4.4030 25150 0.0001 -
4.4118 25200 0.0001 -
4.4205 25250 0.0001 -
4.4293 25300 0.0002 -
4.4380 25350 0.0002 -
4.4468 25400 0.0001 -
4.4555 25450 0.0001 -
4.4643 25500 0.0001 -
4.4730 25550 0.0001 -
4.4818 25600 0.0001 -
4.4905 25650 0.0001 -
4.4993 25700 0.0001 -
4.5081 25750 0.0001 -
4.5168 25800 0.0001 -
4.5256 25850 0.0001 -
4.5343 25900 0.0001 -
4.5431 25950 0.0001 -
4.5518 26000 0.0 -
4.5606 26050 0.0 -
4.5693 26100 0.0001 -
4.5781 26150 0.0001 -
4.5868 26200 0.0001 -
4.5956 26250 0.0001 -
4.6043 26300 0.0001 -
4.6131 26350 0.0001 -
4.6218 26400 0.0002 -
4.6306 26450 0.0001 -
4.6394 26500 0.0001 -
4.6481 26550 0.0001 -
4.6569 26600 0.0001 -
4.6656 26650 0.0 -
4.6744 26700 0.0002 -
4.6831 26750 0.0 -
4.6919 26800 0.0001 -
4.7006 26850 0.0002 -
4.7094 26900 0.0002 -
4.7181 26950 0.0001 -
4.7269 27000 0.0001 -
4.7356 27050 0.0001 -
4.7444 27100 0.0 -
4.7532 27150 0.0001 -
4.7619 27200 0.0001 -
4.7707 27250 0.0001 -
4.7794 27300 0.0 -
4.7882 27350 0.0001 -
4.7969 27400 0.0001 -
4.8057 27450 0.0002 -
4.8144 27500 0.0 -
4.8232 27550 0.0001 -
4.8319 27600 0.0001 -
4.8407 27650 0.0001 -
4.8494 27700 0.0 -
4.8582 27750 0.0001 -
4.8669 27800 0.0001 -
4.8757 27850 0.0001 -
4.8845 27900 0.0001 -
4.8932 27950 0.0001 -
4.9020 28000 0.0001 -
4.9107 28050 0.0001 -
4.9195 28100 0.0 -
4.9282 28150 0.0001 -
4.9370 28200 0.0001 -
4.9457 28250 0.0001 -
4.9545 28300 0.0001 -
4.9632 28350 0.0001 -
4.9720 28400 0.0001 -
4.9807 28450 0.0001 -
4.9895 28500 0.0002 -
4.9982 28550 0.0 -
5.0 28560 - 0.0425
  • The bold row denotes the saved checkpoint.

Framework Versions

  • Python: 3.10.12
  • SetFit: 1.1.0.dev0
  • Sentence Transformers: 3.0.1
  • Transformers: 4.42.4
  • PyTorch: 2.3.1+cu121
  • Datasets: 2.20.0
  • Tokenizers: 0.19.1

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}
}
Downloads last month
3
Safetensors
Model size
22.7M params
Tensor type
F32
·
Inference Examples
Text Classification
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for mann2107/BCMPIIRAB_MiniLM_ALLNewV2

Base model

sentence-transformers/all-MiniLM-L6-v2
Finetuned
(162)
this model