metadata
license: apache-2.0
base_model: distilbert-base-uncased
tags:
- generated_from_trainer
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: distilbert-base-uncased-finetuned-custom-ner
results: []
distilbert-base-uncased-finetuned-custom-ner
This model is a fine-tuned version of distilbert-base-uncased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.7017
- Precision: 0.2292
- Recall: 0.275
- F1: 0.25
- Accuracy: 0.8598
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: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| No log | 1.0 | 1 | 0.8583 | 0.1045 | 0.175 | 0.1308 | 0.8084 |
| No log | 2.0 | 2 | 0.7963 | 0.1228 | 0.175 | 0.1443 | 0.8271 |
| No log | 3.0 | 3 | 0.7497 | 0.1837 | 0.225 | 0.2022 | 0.8505 |
| No log | 4.0 | 4 | 0.7179 | 0.1837 | 0.225 | 0.2022 | 0.8505 |
| No log | 5.0 | 5 | 0.7017 | 0.2292 | 0.275 | 0.25 | 0.8598 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.3.1+cu121
- Datasets 2.14.1
- Tokenizers 0.19.1