Is_there_relation

This model is a fine-tuned version of aubmindlab/bert-base-arabertv02 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.8855
• Macro F1: 0.7979
• Precision: 0.8002
• Recall: 0.7995
• Kappa: 0.5894
• Accuracy: 0.7995

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: 128
• seed: 25
• gradient_accumulation_steps: 2
• total_train_batch_size: 32
• 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	Macro F1	Precision	Recall	Kappa	Accuracy
No log	1.0	218	0.5160	0.7251	0.7659	0.7398	0.4511	0.7398
No log	2.0	437	0.4608	0.8014	0.8108	0.8049	0.5970	0.8049
0.4812	3.0	655	0.5087	0.7864	0.7900	0.7886	0.5661	0.7886
0.4812	4.0	874	0.5219	0.8107	0.8118	0.8103	0.6177	0.8103
0.2407	5.0	1092	0.5657	0.8319	0.8416	0.8347	0.6588	0.8347
0.2407	6.0	1311	0.6980	0.7988	0.8074	0.8022	0.5917	0.8022
0.1383	7.0	1529	0.7769	0.7933	0.8017	0.7967	0.5805	0.7967
0.1383	8.0	1748	0.7336	0.8059	0.8087	0.8076	0.6058	0.8076
0.1383	9.0	1966	0.7426	0.7988	0.8074	0.8022	0.5917	0.8022
0.0878	10.0	2185	0.8211	0.8017	0.8098	0.8049	0.5975	0.8049
0.0878	11.0	2403	0.8737	0.7955	0.7969	0.7967	0.5846	0.7967
0.0573	12.0	2622	0.9043	0.7900	0.7914	0.7913	0.5735	0.7913
0.0573	13.0	2840	0.8937	0.7906	0.7909	0.7913	0.5751	0.7913
0.0423	14.0	3059	0.9004	0.8013	0.8019	0.8022	0.5967	0.8022
0.0423	14.97	3270	0.8855	0.7979	0.8002	0.7995	0.5894	0.7995

Framework versions

• Transformers 4.29.2
• Pytorch 2.0.1+cu118
• Tokenizers 0.13.3