metadata

library_name: transformers
license: apache-2.0
base_model: distilbert/distilbert-base-uncased
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: bert-question-ner
    results: []

bert-question-ner

This model is a fine-tuned version of distilbert/distilbert-base-uncased on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.2044
Precision: 0.7342
Recall: 0.7964
F1: 0.7640
Accuracy: 0.9338

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: 9e-06
train_batch_size: 4
eval_batch_size: 4
seed: 42
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.1
num_epochs: 6

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	0.3311	100	1.1156	0.0	0.0	0.0	0.6528
No log	0.6623	200	0.6775	0.3169	0.4012	0.3541	0.7815
No log	0.9934	300	0.4010	0.5	0.6310	0.5579	0.8771
No log	1.3245	400	0.2844	0.6344	0.6996	0.6654	0.9046
0.7464	1.6556	500	0.2394	0.6404	0.7036	0.6705	0.9163
0.7464	1.9868	600	0.2204	0.6774	0.7661	0.7190	0.9241
0.7464	2.3179	700	0.2080	0.7143	0.7460	0.7298	0.9288
0.7464	2.6490	800	0.2044	0.7342	0.7964	0.7640	0.9338
0.7464	2.9801	900	0.2055	0.7227	0.7883	0.7541	0.9346
0.2123	3.3113	1000	0.2030	0.7361	0.7762	0.7556	0.9353

Framework versions

Transformers 4.48.1
Pytorch 2.5.1+cu124
Datasets 3.2.0
Tokenizers 0.21.0