bart_tr_dependencyparsing

This model is a fine-tuned version of facebook/bart-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0136

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: 5e-05
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss
0.3664	0.2187	1000	0.0402
0.0437	0.4374	2000	0.0287
0.0358	0.6562	3000	0.0250
0.0304	0.8749	4000	0.0223
0.0276	1.0936	5000	0.0203
0.0246	1.3123	6000	0.0195
0.0234	1.5311	7000	0.0181
0.0227	1.7498	8000	0.0167
0.0213	1.9685	9000	0.0157
0.0193	2.1872	10000	0.0149
0.0184	2.4059	11000	0.0144
0.0177	2.6247	12000	0.0140
0.017	2.8434	13000	0.0136

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.0+cu121
• Datasets 3.0.0
• Tokenizers 0.19.1