README.md

metadata

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

TestForColab

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: 0.6591
• Accuracy: 0.62
• F1: 0.6205

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	0.6891	0.54	0.3787
No log	0.01	100	0.6893	0.54	0.3787
No log	0.02	150	0.6876	0.54	0.3787
No log	0.03	200	0.6978	0.48	0.4231
No log	0.03	250	0.6899	0.5	0.4878
No log	0.04	300	0.6825	0.57	0.5577
No log	0.04	350	0.6782	0.62	0.6205
No log	0.05	400	0.6692	0.6	0.5981
No log	0.06	450	0.6688	0.58	0.5664
0.6773	0.06	500	0.6692	0.6	0.5966
0.6773	0.07	550	0.6642	0.62	0.62
0.6773	0.08	600	0.6577	0.65	0.6505
0.6773	0.08	650	0.6618	0.6	0.5992
0.6773	0.09	700	0.6617	0.62	0.62
0.6773	0.09	750	0.6641	0.62	0.6205
0.6773	0.1	800	0.6573	0.62	0.62
0.6773	0.11	850	0.6625	0.61	0.6096
0.6773	0.11	900	0.6625	0.63	0.6303
0.6773	0.12	950	0.6632	0.62	0.6181
0.6414	0.13	1000	0.6613	0.62	0.6206
0.6414	0.13	1050	0.6594	0.62	0.6206
0.6414	0.14	1100	0.6607	0.62	0.6206
0.6414	0.14	1150	0.6580	0.62	0.6205
0.6414	0.15	1200	0.6628	0.62	0.6205
0.6414	0.16	1250	0.6591	0.62	0.6205

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0