distilbert-base-uncased-finetuned-TF-mottos

This model is a fine-tuned version of distilbert-base-uncased on all Autobot and Decepticon mottos from 1984 - 1986. It achieves the following results on the evaluation set:

• Loss: 1.2582
• Accuracy: 0.7931
• F1: 0.7894

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: 8e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 25

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
0.7033	1.0	11	0.6749	0.5862	0.4333
0.6583	2.0	22	0.6416	0.8621	0.8596
0.5224	3.0	33	0.6009	0.7241	0.7241
0.2193	4.0	44	0.9541	0.7241	0.7261
0.0636	5.0	55	1.0212	0.7586	0.7423
0.0478	6.0	66	1.2582	0.7931	0.7931
0.0059	7.0	77	1.3563	0.7931	0.7894
0.0019	8.0	88	1.6273	0.7586	0.7598
0.0453	9.0	99	1.5211	0.7586	0.7598
0.0009	10.0	110	1.5131	0.7586	0.7512
0.0267	11.0	121	1.5327	0.7586	0.7512
0.0008	12.0	132	1.4968	0.7931	0.7894
0.0007	13.0	143	1.5245	0.7586	0.7568
0.0006	14.0	154	1.5497	0.7586	0.7568
0.0005	15.0	165	1.5626	0.7931	0.7894
0.0006	16.0	176	1.6904	0.7586	0.7512
0.0005	17.0	187	1.7350	0.7586	0.7512
0.0004	18.0	198	1.7474	0.7586	0.7512
0.0004	19.0	209	1.7541	0.7586	0.7512
0.0004	20.0	220	1.7573	0.7586	0.7512
0.0004	21.0	231	1.7584	0.7586	0.7512
0.0004	22.0	242	1.7612	0.7586	0.7512
0.0004	23.0	253	1.7633	0.7586	0.7512
0.0003	24.0	264	1.7643	0.7586	0.7512
0.0003	25.0	275	1.7522	0.7586	0.7512

Framework versions

• Transformers 4.47.1
• Pytorch 2.5.1+cu124
• Datasets 3.2.0
• Tokenizers 0.21.0