Hugging Face's logo

bobbyw
/
copilot_relex_v1_with_context

Text Classification
Transformers
TensorBoard
Safetensors
deberta-v2
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
copilot_relex_v1_with_context / README.md
bobbyw's picture
bobbyw
End of training
7c17eec verified
preview code
|
raw
history blame
11.6 kB
metadata
license: mit
base_model: microsoft/deberta-v3-small
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - f1
  - precision
  - recall
model-index:
  - name: copilot_relex_v1_with_context
    results: []

copilot_relex_v1_with_context

This model is a fine-tuned version of microsoft/deberta-v3-small on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.0124
  • Accuracy: 0.0036
  • F1: 0.0059
  • Precision: 0.0030
  • Recall: 0.5938
  • Learning Rate: 0.0

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 2e-05
  • train_batch_size: 32
  • eval_batch_size: 32
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 100

Training results

Training Loss Epoch Step Validation Loss Accuracy F1 Precision Recall Rate
No log 1.0 20 0.5600 0.1031 0.0104 0.0052 0.9375 0.0000
No log 2.0 40 0.3965 0.0171 0.0101 0.0051 1.0 0.0000
No log 3.0 60 0.2675 0.0050 0.0100 0.0050 1.0 0.0000
No log 4.0 80 0.1762 0.0050 0.0100 0.0050 1.0 0.0000
No log 5.0 100 0.1157 0.0050 0.0100 0.0050 1.0 0.0000
No log 6.0 120 0.0800 0.0050 0.0100 0.0050 1.0 0.0000
No log 7.0 140 0.0600 0.0050 0.0100 0.0050 1.0 0.0000
No log 8.0 160 0.0485 0.0050 0.0100 0.0050 1.0 0.0000
No log 9.0 180 0.0418 0.0050 0.0100 0.0050 1.0 0.0000
No log 10.0 200 0.0375 0.0050 0.0100 0.0050 1.0 0.0000
No log 11.0 220 0.0348 0.0050 0.0100 0.0050 1.0 0.0000
No log 12.0 240 0.0330 0.0050 0.0100 0.0050 1.0 0.0000
No log 13.0 260 0.0317 0.0050 0.0100 0.0050 1.0 0.0000
No log 14.0 280 0.0307 0.0050 0.0100 0.0050 1.0 0.0000
No log 15.0 300 0.0299 0.0050 0.0100 0.0050 1.0 0.0000
No log 16.0 320 0.0293 0.0050 0.0100 0.0050 1.0 0.0000
No log 17.0 340 0.0288 0.0050 0.0100 0.0050 1.0 0.0000
No log 18.0 360 0.0279 0.0050 0.0100 0.0050 1.0 0.0000
No log 19.0 380 0.0262 0.0050 0.0100 0.0050 1.0 0.0000
No log 20.0 400 0.0268 0.0050 0.0100 0.0050 1.0 0.0000
No log 21.0 420 0.0253 0.0050 0.0100 0.0050 1.0 0.0000
No log 22.0 440 0.0245 0.0050 0.0100 0.0050 1.0 0.0000
No log 23.0 460 0.0236 0.0050 0.0100 0.0050 1.0 0.0000
No log 24.0 480 0.0234 0.0050 0.0100 0.0050 1.0 0.0000
0.1128 25.0 500 0.0229 0.0050 0.0100 0.0050 1.0 0.0000
0.1128 26.0 520 0.0221 0.0050 0.0100 0.0050 1.0 0.0000
0.1128 27.0 540 0.0216 0.0050 0.0100 0.0050 1.0 0.0000
0.1128 28.0 560 0.0219 0.0045 0.0090 0.0045 0.9062 0.0000
0.1128 29.0 580 0.0206 0.0045 0.0090 0.0045 0.9062 0.0000
0.1128 30.0 600 0.0200 0.0045 0.0090 0.0045 0.9062 0.0000
0.1128 31.0 620 0.0198 0.0045 0.0090 0.0045 0.9062 0.0000
0.1128 32.0 640 0.0195 0.0045 0.0090 0.0045 0.9062 0.0000
0.1128 33.0 660 0.0189 0.0045 0.0090 0.0045 0.9062 0.0000
0.1128 34.0 680 0.0192 0.0045 0.0090 0.0045 0.9062 0.0000
0.1128 35.0 700 0.0184 0.0045 0.0090 0.0045 0.9062 0.0000
0.1128 36.0 720 0.0180 0.0044 0.0087 0.0044 0.875 0.0000
0.1128 37.0 740 0.0174 0.0044 0.0087 0.0044 0.875 0.0000
0.1128 38.0 760 0.0172 0.0042 0.0084 0.0042 0.8438 0.0000
0.1128 39.0 780 0.0167 0.0044 0.0087 0.0044 0.875 0.0000
0.1128 40.0 800 0.0168 0.0045 0.0090 0.0045 0.9062 0.0000
0.1128 41.0 820 0.0168 0.0041 0.0081 0.0041 0.8125 0.0000
0.1128 42.0 840 0.0160 0.0041 0.0081 0.0041 0.8125 0.0000
0.1128 43.0 860 0.0156 0.0042 0.0084 0.0042 0.8438 0.0000
0.1128 44.0 880 0.0159 0.0038 0.0072 0.0036 0.7188 0.0000
0.1128 45.0 900 0.0153 0.0038 0.0075 0.0038 0.75 0.0000
0.1128 46.0 920 0.0155 0.0039 0.0078 0.0039 0.7812 0.0000
0.1128 47.0 940 0.0149 0.0038 0.0072 0.0036 0.7188 0.0000
0.1128 48.0 960 0.0148 0.0038 0.0075 0.0038 0.75 0.0000
0.1128 49.0 980 0.0149 0.0038 0.0072 0.0036 0.7188 0.0000
0.0237 50.0 1000 0.0146 0.0036 0.0072 0.0036 0.7188 1e-05
0.0237 51.0 1020 0.0143 0.0038 0.0075 0.0038 0.75 0.0000
0.0237 52.0 1040 0.0143 0.0038 0.0069 0.0034 0.6875 0.0000
0.0237 53.0 1060 0.0145 0.0038 0.0072 0.0036 0.7188 0.0000
0.0237 54.0 1080 0.0144 0.0036 0.0066 0.0033 0.6562 0.0000
0.0237 55.0 1100 0.0140 0.0039 0.0066 0.0033 0.6562 9e-06
0.0237 56.0 1120 0.0144 0.0036 0.0062 0.0031 0.625 0.0000
0.0237 57.0 1140 0.0142 0.0038 0.0069 0.0034 0.6875 0.0000
0.0237 58.0 1160 0.0140 0.0034 0.0062 0.0031 0.625 0.0000
0.0237 59.0 1180 0.0138 0.0038 0.0066 0.0033 0.6562 0.0000
0.0237 60.0 1200 0.0137 0.0039 0.0069 0.0034 0.6875 0.0000
0.0237 61.0 1220 0.0139 0.0038 0.0066 0.0033 0.6562 0.0000
0.0237 62.0 1240 0.0139 0.0038 0.0066 0.0033 0.6562 0.0000
0.0237 63.0 1260 0.0136 0.0036 0.0062 0.0031 0.625 0.0000
0.0237 64.0 1280 0.0135 0.0039 0.0072 0.0036 0.7188 0.0000
0.0237 65.0 1300 0.0133 0.0038 0.0066 0.0033 0.6562 7e-06
0.0237 66.0 1320 0.0136 0.0038 0.0066 0.0033 0.6562 0.0000
0.0237 67.0 1340 0.0137 0.0034 0.0062 0.0031 0.625 0.0000
0.0237 68.0 1360 0.0134 0.0034 0.0059 0.0030 0.5938 0.0000
0.0237 69.0 1380 0.0131 0.0036 0.0062 0.0031 0.625 0.0000
0.0237 70.0 1400 0.0128 0.0036 0.0062 0.0031 0.625 6e-06
0.0237 71.0 1420 0.0129 0.0036 0.0062 0.0031 0.625 0.0000
0.0237 72.0 1440 0.0131 0.0034 0.0059 0.0030 0.5938 0.0000
0.0237 73.0 1460 0.0129 0.0039 0.0069 0.0034 0.6875 0.0000
0.0237 74.0 1480 0.0127 0.0036 0.0066 0.0033 0.6562 0.0000
0.0179 75.0 1500 0.0129 0.0036 0.0059 0.0030 0.5938 5e-06
0.0179 76.0 1520 0.0130 0.0038 0.0062 0.0031 0.625 0.0000
0.0179 77.0 1540 0.0128 0.0033 0.0059 0.0030 0.5938 0.0000
0.0179 78.0 1560 0.0125 0.0036 0.0062 0.0031 0.625 0.0000
0.0179 79.0 1580 0.0127 0.0034 0.0059 0.0030 0.5938 0.0000
0.0179 80.0 1600 0.0126 0.0036 0.0062 0.0031 0.625 0.0000
0.0179 81.0 1620 0.0124 0.0034 0.0059 0.0030 0.5938 0.0000
0.0179 82.0 1640 0.0124 0.0034 0.0059 0.0030 0.5938 0.0000
0.0179 83.0 1660 0.0126 0.0034 0.0059 0.0030 0.5938 0.0000
0.0179 84.0 1680 0.0127 0.0036 0.0059 0.0030 0.5938 0.0000
0.0179 85.0 1700 0.0124 0.0036 0.0062 0.0031 0.625 3e-06
0.0179 86.0 1720 0.0125 0.0036 0.0059 0.0030 0.5938 0.0000
0.0179 87.0 1740 0.0124 0.0036 0.0059 0.0030 0.5938 0.0000
0.0179 88.0 1760 0.0127 0.0036 0.0059 0.0030 0.5938 0.0000
0.0179 89.0 1780 0.0123 0.0034 0.0059 0.0030 0.5938 0.0000
0.0179 90.0 1800 0.0124 0.0034 0.0059 0.0030 0.5938 0.0000
0.0179 91.0 1820 0.0124 0.0036 0.0059 0.0030 0.5938 0.0000
0.0179 92.0 1840 0.0125 0.0036 0.0059 0.0030 0.5938 0.0000
0.0179 93.0 1860 0.0124 0.0036 0.0059 0.0030 0.5938 0.0000
0.0179 94.0 1880 0.0123 0.0034 0.0059 0.0030 0.5938 0.0000
0.0179 95.0 1900 0.0124 0.0036 0.0059 0.0030 0.5938 0.0000
0.0179 96.0 1920 0.0124 0.0036 0.0059 0.0030 0.5938 0.0000
0.0179 97.0 1940 0.0124 0.0034 0.0059 0.0030 0.5938 0.0000
0.0179 98.0 1960 0.0124 0.0034 0.0059 0.0030 0.5938 0.0000
0.0179 99.0 1980 0.0124 0.0036 0.0059 0.0030 0.5938 0.0000
0.0152 100.0 2000 0.0124 0.0036 0.0059 0.0030 0.5938 0.0

Framework versions

  • Transformers 4.40.1
  • Pytorch 2.2.1+cu121
  • Datasets 2.19.0
  • Tokenizers 0.19.1