metadata

license: mit
base_model: prajjwal1/bert-tiny
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - f1
model-index:
  - name: I04-PC
    results: []

I04-PC

This model is a fine-tuned version of prajjwal1/bert-tiny on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.0916
Accuracy: 0.98
F1: 0.9899

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: 5e-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: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
No log	0.01	50	0.6723	0.66	0.6376
No log	0.02	100	0.5802	0.71	0.7065
No log	0.03	150	0.4534	0.87	0.8701
No log	0.04	200	0.3370	0.88	0.88
No log	0.05	250	0.2473	0.94	0.9394
No log	0.06	300	0.2231	0.93	0.9295
No log	0.07	350	0.1813	0.94	0.9394
No log	0.08	400	0.1519	0.96	0.9599
No log	0.09	450	0.1526	0.96	0.9599
0.3852	0.1	500	0.1554	0.96	0.9599
0.3852	0.11	550	0.1495	0.96	0.9599
0.3852	0.12	600	0.1206	0.96	0.9599
0.3852	0.13	650	0.1013	0.96	0.9599
0.3852	0.14	700	0.0592	0.99	0.9900
0.3852	0.15	750	0.0583	0.99	0.9900
0.3852	0.16	800	0.0551	0.99	0.9900
0.3852	0.17	850	0.0540	0.99	0.9900
0.3852	0.18	900	0.0539	0.99	0.9900
0.3852	0.19	950	0.0537	0.99	0.9900
0.1428	0.2	1000	0.0533	0.99	0.9900
0.1428	0.2	1050	0.0528	0.99	0.9900
0.1428	0.21	1100	0.0510	0.99	0.9900
0.1428	0.22	1150	0.0524	0.99	0.9900
0.1428	0.23	1200	0.0524	0.99	0.9900
0.1428	0.24	1250	0.0510	0.99	0.9900

Framework versions

Transformers 4.35.2
Pytorch 2.1.0+cu121
Datasets 2.16.0
Tokenizers 0.15.0