ArabicNewSplits5_FineTuningAraBERT_run2_AugV5_k3_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: 0.7703
• Qwk: 0.4010
• Mse: 0.7703
• Rmse: 0.8777

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: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Qwk	Mse	Rmse
No log	0.1	2	3.2684	-0.0041	3.2684	1.8079
No log	0.2	4	2.7279	-0.0290	2.7279	1.6516
No log	0.3	6	1.0255	0.0	1.0255	1.0127
No log	0.4	8	0.9475	0.0745	0.9475	0.9734
No log	0.5	10	2.2209	0.0909	2.2209	1.4903
No log	0.6	12	1.7684	0.0255	1.7684	1.3298
No log	0.7	14	0.8458	0.0794	0.8458	0.9196
No log	0.8	16	0.5499	0.0	0.5499	0.7415
No log	0.9	18	0.5856	0.0	0.5856	0.7653
No log	1.0	20	0.5967	-0.0081	0.5967	0.7725
No log	1.1	22	0.9925	0.0388	0.9925	0.9962
No log	1.2	24	1.3507	0.0	1.3507	1.1622
No log	1.3	26	0.8637	-0.0256	0.8637	0.9293
No log	1.4	28	0.6189	-0.0081	0.6189	0.7867
No log	1.5	30	0.6242	-0.0853	0.6242	0.7901
No log	1.6	32	0.6280	-0.0303	0.6280	0.7925
No log	1.7	34	0.8754	0.1579	0.8754	0.9356
No log	1.8	36	1.0114	0.0745	1.0114	1.0057
No log	1.9	38	0.7511	-0.0505	0.7511	0.8667
No log	2.0	40	0.5838	0.0	0.5838	0.7641
No log	2.1	42	0.5937	0.0	0.5937	0.7705
No log	2.2	44	0.5774	0.0	0.5774	0.7599
No log	2.3	46	0.6814	0.1345	0.6814	0.8255
No log	2.4	48	0.7966	0.0222	0.7966	0.8925
No log	2.5	50	0.8042	0.0933	0.8042	0.8968
No log	2.6	52	0.6213	0.0617	0.6213	0.7882
No log	2.7	54	0.5860	0.0145	0.5860	0.7655
No log	2.8	56	0.6004	0.0850	0.6004	0.7749
No log	2.9	58	0.6074	0.1206	0.6074	0.7793
No log	3.0	60	0.6860	0.0391	0.6860	0.8283
No log	3.1	62	0.6894	0.0520	0.6894	0.8303
No log	3.2	64	0.7153	0.1642	0.7153	0.8458
No log	3.3	66	0.7895	0.0	0.7895	0.8885
No log	3.4	68	0.7516	0.0	0.7516	0.8669
No log	3.5	70	0.6517	0.2340	0.6517	0.8073
No log	3.6	72	0.7559	0.1832	0.7559	0.8694
No log	3.7	74	0.7080	0.0703	0.7080	0.8414
No log	3.8	76	0.6558	0.1678	0.6558	0.8098
No log	3.9	78	0.6702	0.2704	0.6702	0.8187
No log	4.0	80	0.7523	0.0955	0.7523	0.8674
No log	4.1	82	1.0699	0.1290	1.0699	1.0344
No log	4.2	84	0.9659	0.1347	0.9659	0.9828
No log	4.3	86	0.6817	0.1230	0.6817	0.8257
No log	4.4	88	0.6650	0.25	0.6650	0.8155
No log	4.5	90	0.8105	0.1005	0.8105	0.9003
No log	4.6	92	0.8501	0.0508	0.8501	0.9220
No log	4.7	94	0.7500	0.3398	0.7500	0.8660
No log	4.8	96	0.7629	0.2762	0.7629	0.8734
No log	4.9	98	0.8192	0.2821	0.8192	0.9051
No log	5.0	100	0.8622	0.2188	0.8622	0.9285
No log	5.1	102	0.9911	0.1280	0.9911	0.9955
No log	5.2	104	0.9229	0.1714	0.9229	0.9607
No log	5.3	106	0.7325	0.2464	0.7325	0.8558
No log	5.4	108	0.7969	0.3462	0.7969	0.8927
No log	5.5	110	0.7850	0.3524	0.7850	0.8860
No log	5.6	112	0.8498	0.2711	0.8498	0.9218
No log	5.7	114	0.9648	0.1769	0.9648	0.9823
No log	5.8	116	1.2418	0.0909	1.2418	1.1143
No log	5.9	118	1.1157	0.0790	1.1157	1.0563
No log	6.0	120	0.8325	0.3394	0.8325	0.9124
No log	6.1	122	0.7181	0.2287	0.7181	0.8474
No log	6.2	124	0.7104	0.2364	0.7104	0.8428
No log	6.3	126	0.8357	0.3455	0.8357	0.9142
No log	6.4	128	1.1899	0.0831	1.1899	1.0908
No log	6.5	130	1.4617	0.0960	1.4617	1.2090
No log	6.6	132	1.3192	0.1182	1.3192	1.1486
No log	6.7	134	0.8820	0.3448	0.8820	0.9391
No log	6.8	136	0.6527	0.2709	0.6527	0.8079
No log	6.9	138	0.6527	0.1852	0.6527	0.8079
No log	7.0	140	0.6468	0.1923	0.6468	0.8042
No log	7.1	142	0.6507	0.2709	0.6507	0.8067
No log	7.2	144	0.8202	0.4010	0.8202	0.9057
No log	7.3	146	1.1856	0.0604	1.1856	1.0888
No log	7.4	148	1.2708	0.0604	1.2708	1.1273
No log	7.5	150	1.0710	0.1828	1.0710	1.0349
No log	7.6	152	0.7601	0.3462	0.7601	0.8718
No log	7.7	154	0.6820	0.1928	0.6820	0.8259
No log	7.8	156	0.6878	0.1928	0.6878	0.8294
No log	7.9	158	0.7371	0.3514	0.7371	0.8585
No log	8.0	160	0.8966	0.3480	0.8966	0.9469
No log	8.1	162	1.0636	0.1292	1.0636	1.0313
No log	8.2	164	1.0710	0.1304	1.0710	1.0349
No log	8.3	166	0.9187	0.2381	0.9187	0.9585
No log	8.4	168	0.7188	0.3561	0.7188	0.8478
No log	8.5	170	0.6598	0.3398	0.6598	0.8123
No log	8.6	172	0.6798	0.3200	0.6798	0.8245
No log	8.7	174	0.7844	0.4010	0.7844	0.8857
No log	8.8	176	0.8623	0.3080	0.8623	0.9286
No log	8.9	178	0.8512	0.3128	0.8512	0.9226
No log	9.0	180	0.7702	0.4010	0.7702	0.8776
No log	9.1	182	0.6968	0.3200	0.6968	0.8348
No log	9.2	184	0.6588	0.3200	0.6588	0.8117
No log	9.3	186	0.6622	0.3200	0.6622	0.8137
No log	9.4	188	0.6756	0.3200	0.6756	0.8220
No log	9.5	190	0.6910	0.3200	0.6910	0.8313
No log	9.6	192	0.6999	0.3200	0.6999	0.8366
No log	9.7	194	0.7224	0.3663	0.7224	0.8499
No log	9.8	196	0.7505	0.3663	0.7505	0.8663
No log	9.9	198	0.7663	0.4010	0.7663	0.8754
No log	10.0	200	0.7703	0.4010	0.7703	0.8777

Framework versions

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