metadata

license: mit
base_model: prajjwal1/bert-tiny
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - f1
model-index:
  - name: MM-MM03
    results: []

MM-MM03

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: 0.6872
Accuracy: 0.57
F1: 0.5203

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	0.0	50	0.6915	0.53	0.3672
No log	0.01	100	0.6923	0.53	0.3672
No log	0.01	150	0.6919	0.53	0.3672
No log	0.01	200	0.6923	0.54	0.4195
No log	0.02	250	0.6924	0.6	0.5777
No log	0.02	300	0.6908	0.54	0.3892
No log	0.03	350	0.6893	0.55	0.4104
No log	0.03	400	0.6911	0.57	0.5437
No log	0.03	450	0.6904	0.57	0.5482
0.6938	0.04	500	0.6981	0.47	0.3005
0.6938	0.04	550	0.6953	0.47	0.3005
0.6938	0.04	600	0.6891	0.56	0.5252
0.6938	0.05	650	0.6871	0.55	0.4817
0.6938	0.05	700	0.6898	0.56	0.4975
0.6938	0.06	750	0.6899	0.55	0.4817
0.6938	0.06	800	0.6906	0.51	0.5102
0.6938	0.06	850	0.6934	0.48	0.48
0.6938	0.07	900	0.6889	0.55	0.4817
0.6938	0.07	950	0.6996	0.47	0.3175
0.6929	0.07	1000	0.6894	0.59	0.5601
0.6929	0.08	1050	0.6941	0.5	0.4777
0.6929	0.08	1100	0.6927	0.49	0.4896
0.6929	0.08	1150	0.6916	0.49	0.4903
0.6929	0.09	1200	0.6874	0.56	0.5252
0.6929	0.09	1250	0.6872	0.57	0.5203

Framework versions

Transformers 4.35.2
Pytorch 2.1.0+cu118
Datasets 2.15.0
Tokenizers 0.15.0