metadata

license: mit
base_model: prajjwal1/bert-tiny
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - f1
model-index:
  - name: BP-test4
    results: []

BP-test4

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: 0.1958
Accuracy: 0.95
F1: 0.9499

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: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
No log	0.01	50	0.6942	0.45	0.2793
No log	0.02	100	0.6915	0.57	0.4337
No log	0.03	150	0.6879	0.55	0.3903
No log	0.04	200	0.6904	0.56	0.5607
No log	0.05	250	0.6847	0.56	0.5607
No log	0.06	300	0.6693	0.56	0.5607
No log	0.07	350	0.5499	0.9	0.8998
No log	0.08	400	0.4220	0.93	0.9295
No log	0.09	450	0.3421	0.93	0.9295
0.6127	0.1	500	0.2987	0.93	0.9295
0.6127	0.11	550	0.2704	0.93	0.9295
0.6127	0.12	600	0.2530	0.93	0.9295
0.6127	0.13	650	0.2199	0.93	0.9297
0.6127	0.14	700	0.2204	0.93	0.9295
0.6127	0.15	750	0.1965	0.95	0.9499
0.6127	0.16	800	0.1944	0.95	0.9499
0.6127	0.17	850	0.1942	0.95	0.9499
0.6127	0.18	900	0.1938	0.95	0.9499
0.6127	0.19	950	0.1950	0.95	0.9499
0.2388	0.2	1000	0.1943	0.95	0.9499
0.2388	0.2	1050	0.1939	0.95	0.9499
0.2388	0.21	1100	0.1939	0.95	0.9499
0.2388	0.22	1150	0.1928	0.95	0.9499
0.2388	0.23	1200	0.1937	0.95	0.9499
0.2388	0.24	1250	0.1958	0.95	0.9499

Framework versions

Transformers 4.35.2
Pytorch 2.1.0+cu118
Datasets 2.15.0
Tokenizers 0.15.0