Edit model card

deberta-v3-small_v1_no_entities_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: 0.0315
  • Accuracy: 0.0062
  • F1: 0.0086
  • Precision: 0.0043
  • Recall: 0.9070
  • 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: 8
  • eval_batch_size: 8
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 50

Training results

Training Loss Epoch Step Validation Loss Accuracy F1 Precision Recall Rate
No log 1.0 191 0.0385 0.0047 0.0094 0.0047 1.0 0.0000
No log 2.0 382 0.0282 0.0047 0.0094 0.0047 1.0 0.0000
0.1139 3.0 573 0.0274 0.0047 0.0094 0.0047 1.0 0.0000
0.1139 4.0 764 0.0270 0.0047 0.0094 0.0047 1.0 0.0000
0.1139 5.0 955 0.0271 0.0047 0.0094 0.0047 1.0 0.0000
0.0317 6.0 1146 0.0269 0.0047 0.0094 0.0047 1.0 0.0000
0.0317 7.0 1337 0.0271 0.0047 0.0094 0.0047 1.0 0.0000
0.0316 8.0 1528 0.0264 0.0047 0.0094 0.0047 1.0 0.0000
0.0316 9.0 1719 0.0261 0.0047 0.0094 0.0047 1.0 0.0000
0.0316 10.0 1910 0.0261 0.0047 0.0094 0.0047 1.0 0.0000
0.0299 11.0 2101 0.0263 0.0047 0.0094 0.0047 1.0 0.0000
0.0299 12.0 2292 0.0262 0.0047 0.0094 0.0047 1.0 0.0000
0.0299 13.0 2483 0.0260 0.0047 0.0094 0.0047 1.0 0.0000
0.0294 14.0 2674 0.0263 0.0047 0.0094 0.0047 1.0 0.0000
0.0294 15.0 2865 0.0259 0.0047 0.0094 0.0047 1.0 0.0000
0.026 16.0 3056 0.0262 0.0050 0.0094 0.0047 1.0 0.0000
0.026 17.0 3247 0.0265 0.0050 0.0092 0.0046 0.9767 0.0000
0.026 18.0 3438 0.0270 0.0048 0.0093 0.0047 0.9884 0.0000
0.0224 19.0 3629 0.0272 0.0056 0.0090 0.0045 0.9535 0.0000
0.0224 20.0 3820 0.0271 0.0055 0.0091 0.0046 0.9651 0.0000
0.0197 21.0 4011 0.0271 0.0052 0.0090 0.0045 0.9535 0.0000
0.0197 22.0 4202 0.0270 0.0050 0.0090 0.0045 0.9535 0.0000
0.0197 23.0 4393 0.0271 0.0056 0.0090 0.0045 0.9535 0.0000
0.0172 24.0 4584 0.0275 0.0053 0.0089 0.0045 0.9419 0.0000
0.0172 25.0 4775 0.0273 0.0053 0.0089 0.0045 0.9419 1e-05
0.0172 26.0 4966 0.0282 0.0061 0.0087 0.0044 0.9186 0.0000
0.0152 27.0 5157 0.0281 0.0060 0.0088 0.0044 0.9302 0.0000
0.0152 28.0 5348 0.0281 0.0058 0.0088 0.0044 0.9302 0.0000
0.0138 29.0 5539 0.0277 0.0059 0.0088 0.0044 0.9302 0.0000
0.0138 30.0 5730 0.0292 0.0056 0.0089 0.0045 0.9419 0.0000
0.0138 31.0 5921 0.0287 0.0061 0.0088 0.0044 0.9302 0.0000
0.0124 32.0 6112 0.0289 0.0059 0.0087 0.0044 0.9186 0.0000
0.0124 33.0 6303 0.0300 0.0062 0.0086 0.0043 0.9070 0.0000
0.0124 34.0 6494 0.0293 0.0057 0.0087 0.0044 0.9186 0.0000
0.0113 35.0 6685 0.0297 0.0059 0.0089 0.0045 0.9419 6e-06
0.0113 36.0 6876 0.0293 0.0060 0.0086 0.0043 0.9070 0.0000
0.0106 37.0 7067 0.0295 0.0060 0.0085 0.0043 0.8953 0.0000
0.0106 38.0 7258 0.0301 0.0063 0.0086 0.0043 0.9070 0.0000
0.0106 39.0 7449 0.0300 0.0063 0.0085 0.0043 0.8953 0.0000
0.0092 40.0 7640 0.0297 0.0057 0.0086 0.0043 0.9070 0.0000
0.0092 41.0 7831 0.0299 0.0061 0.0086 0.0043 0.9070 0.0000
0.0091 42.0 8022 0.0298 0.0064 0.0086 0.0043 0.9070 0.0000
0.0091 43.0 8213 0.0302 0.0061 0.0087 0.0044 0.9186 0.0000
0.0091 44.0 8404 0.0307 0.0062 0.0086 0.0043 0.9070 0.0000
0.0082 45.0 8595 0.0310 0.0062 0.0086 0.0043 0.9070 0.0000
0.0082 46.0 8786 0.0308 0.0062 0.0087 0.0044 0.9186 0.0000
0.0082 47.0 8977 0.0314 0.0062 0.0087 0.0044 0.9186 0.0000
0.0081 48.0 9168 0.0314 0.0064 0.0087 0.0044 0.9186 0.0000
0.0081 49.0 9359 0.0315 0.0062 0.0086 0.0043 0.9070 0.0000
0.0077 50.0 9550 0.0315 0.0062 0.0086 0.0043 0.9070 0.0

Framework versions

  • Transformers 4.40.1
  • Pytorch 2.2.1+cu121
  • Datasets 2.19.0
  • Tokenizers 0.19.1
Downloads last month
7
Safetensors
Model size
142M params
Tensor type
F32
·
Inference Examples
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 bobbyw/deberta-v3-small_v1_no_entities_with_context

Finetuned
(105)
this model