alex-miller/nyt-cat

DistilBERT base classify news topics - Devinit

This model is a fine-tuned version of distilbert-base-uncased on the New York Times Topics dataset. It achieves the following results on the evaluation set:

• Loss: 0.2871
• Accuracy: 0.9135

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: 1e-05
• train_batch_size: 128
• eval_batch_size: 128
• 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
0.386	1.0	1340	0.3275	0.8921
0.2833	2.0	2680	0.2840	0.9033
0.2411	3.0	4020	0.2694	0.9102
0.2069	4.0	5360	0.2665	0.9114
0.1796	5.0	6700	0.2657	0.9128
0.1636	6.0	8040	0.2674	0.9142
0.144	7.0	9380	0.2761	0.9129
0.1277	8.0	10720	0.2820	0.9125
0.1201	9.0	12060	0.2853	0.9136
0.1104	10.0	13400	0.2871	0.9135

Framework versions

• Transformers 4.36.2
• Pytorch 2.1.0+cu121
• Datasets 2.16.1
• Tokenizers 0.15.0