README.md

metadata

tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - accuracy
model-index:
  - name: Type_of_relation
    results: []

Type_of_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: 1.0885
• Macro F1: 0.7537
• Precision: 0.7463
• Recall: 0.7783
• Kappa: 0.6636
• Accuracy: 0.7783

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	101	1.2153	0.5786	0.5030	0.6835	0.4719	0.6835
No log	2.0	203	1.0583	0.6615	0.6707	0.7365	0.5699	0.7365
No log	3.0	304	0.9495	0.6925	0.6934	0.7525	0.6069	0.7525
No log	4.0	406	0.8934	0.7325	0.7283	0.7635	0.6400	0.7635
0.976	5.0	507	0.9247	0.7219	0.7166	0.7660	0.6352	0.7660
0.976	6.0	609	0.8751	0.7502	0.7422	0.7685	0.6594	0.7685
0.976	7.0	710	0.9145	0.7510	0.7395	0.7783	0.6640	0.7783
0.976	8.0	812	0.9934	0.7479	0.7423	0.7808	0.6609	0.7808
0.976	9.0	913	0.9641	0.7506	0.7425	0.7734	0.6594	0.7734
0.3286	10.0	1015	0.9702	0.7560	0.7587	0.7746	0.6641	0.7746
0.3286	11.0	1116	1.0610	0.7430	0.7370	0.7746	0.6530	0.7746
0.3286	12.0	1218	1.0251	0.7537	0.7442	0.7722	0.6611	0.7722
0.3286	13.0	1319	1.0703	0.7511	0.7433	0.7771	0.6615	0.7771
0.3286	14.0	1421	1.0767	0.7534	0.7451	0.7771	0.6631	0.7771
0.1456	14.93	1515	1.0885	0.7537	0.7463	0.7783	0.6636	0.7783

Framework versions

• Transformers 4.29.2
• Pytorch 2.0.1+cu118
• Tokenizers 0.13.3