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: 0.6537
• Rouge1: 0.3618
• Rouge2: 0.2634
• Rougel: 0.3348
• Rougelsum: 0.336
• 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	3.3018	0.2994	0.1537	0.2528	0.2525	20.0
No log	2.0	54	2.2697	0.3287	0.1959	0.286	0.2866	20.0
No log	3.0	81	1.7739	0.3265	0.2103	0.2947	0.2954	20.0
No log	4.0	108	1.4085	0.3257	0.2128	0.2931	0.2937	20.0
No log	5.0	135	1.1230	0.3458	0.2307	0.3116	0.3124	20.0
No log	6.0	162	0.9408	0.3448	0.2371	0.3129	0.3139	20.0
No log	7.0	189	0.8269	0.3502	0.2479	0.3201	0.3212	20.0
No log	8.0	216	0.7584	0.3442	0.2434	0.3145	0.3157	20.0
No log	9.0	243	0.7075	0.3606	0.2619	0.3329	0.3339	20.0
No log	10.0	270	0.6890	0.3508	0.2517	0.3208	0.3219	20.0
No log	11.0	297	0.6761	0.3501	0.2483	0.3215	0.3228	20.0
No log	12.0	324	0.6631	0.3532	0.2522	0.3237	0.3245	20.0
No log	13.0	351	0.6573	0.3625	0.2661	0.3358	0.3372	20.0
No log	14.0	378	0.6528	0.3594	0.2608	0.3322	0.3333	20.0
No log	15.0	405	0.6537	0.3618	0.2634	0.3348	0.336	20.0

Framework versions

• Transformers 4.40.2
• Pytorch 2.2.1+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1