ArabicNewSplits8_FineTuningAraBERT_noAug_task3_organization

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: 0.5796
• Qwk: 0.3828
• Mse: 0.5796
• Rmse: 0.7613

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Qwk	Mse	Rmse
No log	0.6667	2	3.7564	0.0	3.7564	1.9382
No log	1.3333	4	2.1125	-0.0464	2.1125	1.4535
No log	2.0	6	0.9384	0.0740	0.9384	0.9687
No log	2.6667	8	0.5111	0.2470	0.5111	0.7149
No log	3.3333	10	0.5326	0.1544	0.5326	0.7298
No log	4.0	12	0.5032	0.1842	0.5032	0.7094
No log	4.6667	14	0.5492	0.1597	0.5492	0.7411
No log	5.3333	16	0.5470	0.0653	0.5470	0.7396
No log	6.0	18	0.5204	0.2405	0.5204	0.7214
No log	6.6667	20	0.6030	0.3109	0.6030	0.7765
No log	7.3333	22	0.5992	0.3416	0.5992	0.7741
No log	8.0	24	0.9672	0.1132	0.9672	0.9834
No log	8.6667	26	0.7700	0.2313	0.7700	0.8775
No log	9.3333	28	0.8126	0.2709	0.8126	0.9015
No log	10.0	30	0.6651	0.2674	0.6651	0.8156
No log	10.6667	32	0.9214	0.1527	0.9214	0.9599
No log	11.3333	34	0.8128	0.1633	0.8128	0.9016
No log	12.0	36	0.6493	0.2222	0.6493	0.8058
No log	12.6667	38	0.6881	0.2212	0.6881	0.8295
No log	13.3333	40	0.6895	0.2198	0.6895	0.8304
No log	14.0	42	0.7659	0.3136	0.7659	0.8752
No log	14.6667	44	0.7675	0.3136	0.7675	0.8761
No log	15.3333	46	0.7291	0.1964	0.7291	0.8539
No log	16.0	48	0.7371	0.1918	0.7371	0.8585
No log	16.6667	50	0.6185	0.2923	0.6185	0.7865
No log	17.3333	52	0.6598	0.2093	0.6598	0.8123
No log	18.0	54	0.5984	0.1029	0.5984	0.7736
No log	18.6667	56	0.7167	0.1423	0.7167	0.8466
No log	19.3333	58	0.6259	0.2017	0.6259	0.7911
No log	20.0	60	0.6114	0.3149	0.6114	0.7819
No log	20.6667	62	0.6216	0.2795	0.6216	0.7884
No log	21.3333	64	0.7252	0.1365	0.7252	0.8516
No log	22.0	66	0.8477	0.1620	0.8477	0.9207
No log	22.6667	68	0.6088	0.3655	0.6088	0.7802
No log	23.3333	70	0.6748	0.2151	0.6748	0.8215
No log	24.0	72	0.5741	0.3327	0.5741	0.7577
No log	24.6667	74	0.6138	0.1870	0.6138	0.7834
No log	25.3333	76	0.6094	0.2450	0.6094	0.7806
No log	26.0	78	0.5562	0.3521	0.5562	0.7458
No log	26.6667	80	0.5740	0.3841	0.5740	0.7576
No log	27.3333	82	0.7036	0.1747	0.7036	0.8388
No log	28.0	84	0.6490	0.2295	0.6490	0.8056
No log	28.6667	86	0.6288	0.1889	0.6288	0.7930
No log	29.3333	88	0.6417	0.1940	0.6417	0.8011
No log	30.0	90	0.6549	0.1904	0.6549	0.8092
No log	30.6667	92	0.5789	0.2051	0.5789	0.7609
No log	31.3333	94	0.5853	0.1934	0.5853	0.7651
No log	32.0	96	0.5992	0.3107	0.5992	0.7741
No log	32.6667	98	0.9428	0.1885	0.9428	0.9710
No log	33.3333	100	1.1201	0.1443	1.1201	1.0583
No log	34.0	102	0.7434	0.2101	0.7434	0.8622
No log	34.6667	104	0.6690	0.3275	0.6690	0.8179
No log	35.3333	106	0.6602	0.3868	0.6602	0.8125
No log	36.0	108	0.5896	0.4023	0.5896	0.7679
No log	36.6667	110	0.6188	0.2697	0.6188	0.7867
No log	37.3333	112	0.5646	0.4023	0.5646	0.7514
No log	38.0	114	0.5554	0.3543	0.5554	0.7453
No log	38.6667	116	0.6494	0.2928	0.6494	0.8059
No log	39.3333	118	0.5946	0.2872	0.5946	0.7711
No log	40.0	120	0.5958	0.2847	0.5958	0.7719
No log	40.6667	122	0.5830	0.2805	0.5830	0.7636
No log	41.3333	124	0.5896	0.3256	0.5896	0.7678
No log	42.0	126	0.6523	0.2354	0.6523	0.8076
No log	42.6667	128	0.6406	0.2922	0.6406	0.8004
No log	43.3333	130	0.5813	0.2850	0.5813	0.7624
No log	44.0	132	0.6464	0.1943	0.6464	0.8040
No log	44.6667	134	0.5808	0.3519	0.5808	0.7621
No log	45.3333	136	0.5392	0.2736	0.5392	0.7343
No log	46.0	138	0.5365	0.2250	0.5365	0.7325
No log	46.6667	140	0.5390	0.0553	0.5390	0.7342
No log	47.3333	142	0.5755	0.1019	0.5755	0.7586
No log	48.0	144	0.5715	0.0982	0.5715	0.7560
No log	48.6667	146	0.5471	0.2640	0.5471	0.7397
No log	49.3333	148	0.5900	0.2985	0.5900	0.7681
No log	50.0	150	0.5944	0.3380	0.5944	0.7710
No log	50.6667	152	0.5782	0.3202	0.5782	0.7604
No log	51.3333	154	0.5748	0.3543	0.5748	0.7582
No log	52.0	156	0.6213	0.1585	0.6213	0.7882
No log	52.6667	158	0.5877	0.2748	0.5877	0.7666
No log	53.3333	160	0.5677	0.2423	0.5677	0.7535
No log	54.0	162	0.5553	0.2564	0.5553	0.7452
No log	54.6667	164	0.5583	0.1520	0.5583	0.7472
No log	55.3333	166	0.6110	0.0793	0.6110	0.7817
No log	56.0	168	0.6696	0.2138	0.6696	0.8183
No log	56.6667	170	0.6429	0.1148	0.6429	0.8018
No log	57.3333	172	0.5756	0.3696	0.5756	0.7587
No log	58.0	174	0.6037	0.3052	0.6037	0.7770
No log	58.6667	176	0.6507	0.2999	0.6507	0.8067
No log	59.3333	178	0.5970	0.3052	0.5970	0.7727
No log	60.0	180	0.5701	0.2956	0.5701	0.7550
No log	60.6667	182	0.5676	0.2524	0.5676	0.7534
No log	61.3333	184	0.5744	0.3092	0.5744	0.7579
No log	62.0	186	0.5831	0.3007	0.5831	0.7636
No log	62.6667	188	0.5900	0.2540	0.5900	0.7681
No log	63.3333	190	0.5928	0.3007	0.5928	0.7700
No log	64.0	192	0.5925	0.1570	0.5925	0.7697
No log	64.6667	194	0.5855	0.1061	0.5855	0.7652
No log	65.3333	196	0.5739	0.1101	0.5739	0.7575
No log	66.0	198	0.5610	0.1622	0.5610	0.7490
No log	66.6667	200	0.5536	0.1622	0.5536	0.7440
No log	67.3333	202	0.5490	0.1622	0.5490	0.7409
No log	68.0	204	0.5494	0.2564	0.5494	0.7412
No log	68.6667	206	0.5558	0.2821	0.5558	0.7455
No log	69.3333	208	0.5531	0.2220	0.5531	0.7437
No log	70.0	210	0.5397	0.3415	0.5397	0.7347
No log	70.6667	212	0.5334	0.3915	0.5334	0.7303
No log	71.3333	214	0.5390	0.3521	0.5390	0.7341
No log	72.0	216	0.5409	0.3521	0.5409	0.7355
No log	72.6667	218	0.5429	0.3521	0.5429	0.7368
No log	73.3333	220	0.5443	0.3521	0.5443	0.7378
No log	74.0	222	0.5499	0.2492	0.5499	0.7415
No log	74.6667	224	0.5622	0.2034	0.5622	0.7498
No log	75.3333	226	0.5809	0.2806	0.5809	0.7622
No log	76.0	228	0.5955	0.2205	0.5955	0.7717
No log	76.6667	230	0.5906	0.2205	0.5906	0.7685
No log	77.3333	232	0.5853	0.2806	0.5853	0.7651
No log	78.0	234	0.5749	0.2978	0.5749	0.7582
No log	78.6667	236	0.5635	0.2927	0.5635	0.7507
No log	79.3333	238	0.5759	0.2540	0.5759	0.7589
No log	80.0	240	0.6189	0.3052	0.6189	0.7867
No log	80.6667	242	0.6669	0.2450	0.6669	0.8166
No log	81.3333	244	0.6621	0.2450	0.6621	0.8137
No log	82.0	246	0.6159	0.3052	0.6159	0.7848
No log	82.6667	248	0.5722	0.2612	0.5722	0.7564
No log	83.3333	250	0.5602	0.2114	0.5602	0.7485
No log	84.0	252	0.5704	0.2034	0.5704	0.7552
No log	84.6667	254	0.5751	0.2034	0.5751	0.7583
No log	85.3333	256	0.5732	0.2034	0.5732	0.7571
No log	86.0	258	0.5750	0.2993	0.5750	0.7583
No log	86.6667	260	0.5745	0.3816	0.5745	0.7580
No log	87.3333	262	0.5710	0.3816	0.5710	0.7556
No log	88.0	264	0.5743	0.2956	0.5743	0.7578
No log	88.6667	266	0.5887	0.3052	0.5887	0.7673
No log	89.3333	268	0.5982	0.3052	0.5982	0.7734
No log	90.0	270	0.5983	0.3052	0.5983	0.7735
No log	90.6667	272	0.5956	0.3052	0.5956	0.7718
No log	91.3333	274	0.5864	0.3052	0.5864	0.7658
No log	92.0	276	0.5801	0.3052	0.5801	0.7617
No log	92.6667	278	0.5776	0.3052	0.5776	0.7600
No log	93.3333	280	0.5798	0.3052	0.5798	0.7614
No log	94.0	282	0.5814	0.3052	0.5814	0.7625
No log	94.6667	284	0.5811	0.2971	0.5811	0.7623
No log	95.3333	286	0.5799	0.3442	0.5799	0.7615
No log	96.0	288	0.5803	0.3442	0.5803	0.7618
No log	96.6667	290	0.5803	0.3442	0.5803	0.7618
No log	97.3333	292	0.5803	0.3828	0.5803	0.7618
No log	98.0	294	0.5800	0.3828	0.5800	0.7616
No log	98.6667	296	0.5799	0.3828	0.5799	0.7615
No log	99.3333	298	0.5797	0.3828	0.5797	0.7614
No log	100.0	300	0.5796	0.3828	0.5796	0.7613

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.0+cu118
• Datasets 2.21.0
• Tokenizers 0.19.1