BARTModel_for_Ecommerce

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: 0.6400
• Rouge1: 0.3515
• Rouge2: 0.2381
• Rougel: 0.3187
• Rougelsum: 0.3187
• Gen Len: 20.0

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: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	1.0	27	3.5206	0.3072	0.1595	0.26	0.26	20.0
No log	2.0	54	2.3786	0.3139	0.1747	0.268	0.2681	20.0
No log	3.0	81	1.8443	0.3328	0.2038	0.2924	0.2932	20.0
No log	4.0	108	1.4537	0.3276	0.2076	0.2887	0.2892	20.0
No log	5.0	135	1.1480	0.3301	0.212	0.292	0.2924	20.0
No log	6.0	162	0.9457	0.3465	0.2292	0.3084	0.3091	20.0
No log	7.0	189	0.8317	0.345	0.2253	0.3074	0.3078	20.0
No log	8.0	216	0.7544	0.3456	0.2293	0.3121	0.3124	20.0
No log	9.0	243	0.7076	0.3601	0.246	0.3278	0.3276	20.0
No log	10.0	270	0.6817	0.3464	0.2358	0.3139	0.3139	20.0
No log	11.0	297	0.6609	0.3586	0.2407	0.3235	0.3241	20.0
No log	12.0	324	0.6557	0.3563	0.2432	0.3226	0.3227	20.0
No log	13.0	351	0.6451	0.3511	0.238	0.3192	0.3195	20.0
No log	14.0	378	0.6430	0.3516	0.2385	0.3182	0.3183	20.0
No log	15.0	405	0.6400	0.3515	0.2381	0.3187	0.3187	20.0

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu121
• Datasets 2.16.0
• Tokenizers 0.15.0