metadata

license: mit
base_model: prajjwal1/bert-tiny
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - f1
model-index:
  - name: MM05-PC
    results: []

MM05-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: 0.3358
Accuracy: 0.99
F1: 0.9950

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: 4e-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	0.6889	0.58	0.4258
No log	0.01	100	0.6984	0.42	0.2485
No log	0.01	150	0.6948	0.42	0.2485
No log	0.02	200	0.6867	0.58	0.4258
No log	0.02	250	0.6859	0.75	0.7192
No log	0.02	300	0.6674	0.67	0.6705
No log	0.03	350	0.6093	0.65	0.6514
No log	0.03	400	0.5726	0.71	0.7118
No log	0.04	450	0.5586	0.73	0.7311
0.6576	0.04	500	0.5368	0.67	0.6680
0.6576	0.04	550	0.5532	0.73	0.7311
0.6576	0.05	600	0.5290	0.73	0.7311
0.6576	0.05	650	0.5073	0.73	0.7311
0.6576	0.06	700	0.4731	0.76	0.7489
0.6576	0.06	750	0.4519	0.81	0.7956
0.6576	0.06	800	0.4411	0.81	0.7956
0.6576	0.07	850	0.4401	0.81	0.7956
0.6576	0.07	900	0.4382	0.81	0.7956
0.6576	0.08	950	0.4351	0.81	0.7956
0.4885	0.08	1000	0.4348	0.81	0.7956
0.4885	0.08	1050	0.4289	0.81	0.7956
0.4885	0.09	1100	0.4130	0.81	0.7956
0.4885	0.09	1150	0.4087	0.81	0.7956
0.4885	0.1	1200	0.4165	0.81	0.7956
0.4885	0.1	1250	0.4088	0.81	0.7956

Framework versions

Transformers 4.35.2
Pytorch 2.1.0+cu121
Datasets 2.16.0
Tokenizers 0.15.0