arabert_cross_development_task1_fold0

This model is a fine-tuned version of aubmindlab/bert-base-arabertv02 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.8306
• Qwk: 0.2565
• Mse: 0.8306

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: 64
• eval_batch_size: 64
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Qwk	Mse
No log	0.1333	2	4.0034	-0.0039	4.0034
No log	0.2667	4	1.7572	0.0385	1.7572
No log	0.4	6	0.8795	0.1253	0.8795
No log	0.5333	8	1.0722	0.1468	1.0722
No log	0.6667	10	1.9981	0.0739	1.9981
No log	0.8	12	1.4228	0.1298	1.4228
No log	0.9333	14	0.7120	0.1720	0.7120
No log	1.0667	16	0.5182	0.2951	0.5182
No log	1.2	18	0.6133	0.2736	0.6133
No log	1.3333	20	1.1336	0.1776	1.1336
No log	1.4667	22	1.2656	0.1714	1.2656
No log	1.6	24	0.8675	0.2380	0.8675
No log	1.7333	26	0.5730	0.3050	0.5730
No log	1.8667	28	0.5924	0.2784	0.5924
No log	2.0	30	0.6861	0.2120	0.6861
No log	2.1333	32	0.8867	0.2029	0.8867
No log	2.2667	34	0.9172	0.2121	0.9172
No log	2.4	36	0.7721	0.2159	0.7721
No log	2.5333	38	0.8001	0.2277	0.8001
No log	2.6667	40	0.8684	0.2498	0.8684
No log	2.8	42	0.9570	0.2264	0.9570
No log	2.9333	44	0.8803	0.2439	0.8803
No log	3.0667	46	0.7435	0.2799	0.7435
No log	3.2	48	0.6805	0.3082	0.6805
No log	3.3333	50	0.8424	0.2730	0.8424
No log	3.4667	52	0.8402	0.2670	0.8402
No log	3.6	54	0.8115	0.2861	0.8115
No log	3.7333	56	0.8179	0.2776	0.8179
No log	3.8667	58	0.7692	0.2822	0.7692
No log	4.0	60	0.6605	0.2944	0.6605
No log	4.1333	62	0.6724	0.3033	0.6724
No log	4.2667	64	0.7918	0.2700	0.7918
No log	4.4	66	0.9373	0.2649	0.9373
No log	4.5333	68	0.8734	0.2482	0.8734
No log	4.6667	70	0.6994	0.2866	0.6994
No log	4.8	72	0.5761	0.3307	0.5761
No log	4.9333	74	0.6497	0.2879	0.6497
No log	5.0667	76	0.7333	0.2628	0.7333
No log	5.2	78	1.0341	0.2345	1.0341
No log	5.3333	80	1.1378	0.2214	1.1378
No log	5.4667	82	0.9255	0.2527	0.9255
No log	5.6	84	0.6589	0.3108	0.6589
No log	5.7333	86	0.5709	0.3576	0.5709
No log	5.8667	88	0.6392	0.2982	0.6392
No log	6.0	90	0.7919	0.2436	0.7919
No log	6.1333	92	0.7855	0.2412	0.7855
No log	6.2667	94	0.6800	0.2919	0.6800
No log	6.4	96	0.6850	0.2998	0.6850
No log	6.5333	98	0.8473	0.2545	0.8473
No log	6.6667	100	0.9780	0.2417	0.9780
No log	6.8	102	0.9663	0.2331	0.9663
No log	6.9333	104	0.8213	0.2337	0.8213
No log	7.0667	106	0.6942	0.2789	0.6942
No log	7.2	108	0.7119	0.2738	0.7119
No log	7.3333	110	0.8067	0.2517	0.8067
No log	7.4667	112	0.8998	0.2447	0.8998
No log	7.6	114	0.9907	0.2199	0.9907
No log	7.7333	116	0.9290	0.2364	0.9290
No log	7.8667	118	0.7925	0.2551	0.7925
No log	8.0	120	0.7137	0.2738	0.7137
No log	8.1333	122	0.7202	0.2753	0.7202
No log	8.2667	124	0.7738	0.2663	0.7738
No log	8.4	126	0.8078	0.2539	0.8078
No log	8.5333	128	0.8794	0.2516	0.8794
No log	8.6667	130	0.9809	0.2362	0.9809
No log	8.8	132	1.0066	0.2362	1.0066
No log	8.9333	134	0.9524	0.2397	0.9524
No log	9.0667	136	0.8849	0.2464	0.8849
No log	9.2	138	0.8285	0.2565	0.8285
No log	9.3333	140	0.7806	0.2545	0.7806
No log	9.4667	142	0.7707	0.2569	0.7707
No log	9.6	144	0.7836	0.2545	0.7836
No log	9.7333	146	0.8063	0.2550	0.8063
No log	9.8667	148	0.8258	0.2565	0.8258
No log	10.0	150	0.8306	0.2565	0.8306

Framework versions

• Transformers 4.44.0
• Pytorch 2.4.0
• Datasets 2.21.0
• Tokenizers 0.19.1