ArabicNewSplits5_FineTuningAraBERT_run1_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: 0.7328
• Qwk: 0.3744
• Mse: 0.7328
• Rmse: 0.8560

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	2.9128	-0.0016	2.9128	1.7067
No log	0.2	4	1.8052	0.0649	1.8052	1.3436
No log	0.3	6	1.0602	0.0345	1.0602	1.0297
No log	0.4	8	1.0795	0.0038	1.0795	1.0390
No log	0.5	10	1.1156	0.0	1.1156	1.0562
No log	0.6	12	1.4115	0.0	1.4115	1.1881
No log	0.7	14	1.2417	0.0038	1.2417	1.1143
No log	0.8	16	0.9592	0.0345	0.9592	0.9794
No log	0.9	18	0.7833	0.1373	0.7833	0.8850
No log	1.0	20	0.8955	0.1861	0.8955	0.9463
No log	1.1	22	0.7795	0.1443	0.7795	0.8829
No log	1.2	24	0.7637	0.1515	0.7637	0.8739
No log	1.3	26	1.0076	-0.0233	1.0076	1.0038
No log	1.4	28	1.4845	0.0	1.4845	1.2184
No log	1.5	30	1.5404	0.0	1.5404	1.2411
No log	1.6	32	1.2194	0.0	1.2194	1.1043
No log	1.7	34	0.9036	0.0431	0.9036	0.9506
No log	1.8	36	0.6705	0.2281	0.6705	0.8188
No log	1.9	38	0.6151	0.0071	0.6151	0.7843
No log	2.0	40	0.6087	-0.0909	0.6087	0.7802
No log	2.1	42	0.6171	-0.0370	0.6171	0.7855
No log	2.2	44	0.6692	-0.0556	0.6692	0.8180
No log	2.3	46	0.8371	0.2157	0.8371	0.9149
No log	2.4	48	1.1354	0.0569	1.1354	1.0656
No log	2.5	50	1.5015	-0.0070	1.5015	1.2254
No log	2.6	52	1.4014	0.0204	1.4014	1.1838
No log	2.7	54	1.0092	0.0901	1.0092	1.0046
No log	2.8	56	0.7952	0.1030	0.7952	0.8918
No log	2.9	58	0.7255	0.0	0.7255	0.8518
No log	3.0	60	0.6910	0.0	0.6910	0.8313
No log	3.1	62	0.6584	0.0	0.6584	0.8114
No log	3.2	64	0.6370	0.0	0.6370	0.7981
No log	3.3	66	0.6568	0.2099	0.6568	0.8105
No log	3.4	68	0.6678	0.2444	0.6678	0.8172
No log	3.5	70	0.6276	0.0933	0.6276	0.7922
No log	3.6	72	0.6004	0.0604	0.6004	0.7748
No log	3.7	74	0.5987	0.0728	0.5987	0.7738
No log	3.8	76	0.6033	0.0728	0.6033	0.7767
No log	3.9	78	0.6536	0.2575	0.6536	0.8085
No log	4.0	80	0.7201	0.2165	0.7201	0.8486
No log	4.1	82	0.6928	0.2766	0.6928	0.8323
No log	4.2	84	0.5941	0.1169	0.5941	0.7708
No log	4.3	86	0.7461	0.2258	0.7461	0.8638
No log	4.4	88	0.7634	0.1475	0.7634	0.8737
No log	4.5	90	0.6465	0.2273	0.6465	0.8040
No log	4.6	92	0.5657	0.3121	0.5657	0.7521
No log	4.7	94	0.6294	0.2593	0.6294	0.7933
No log	4.8	96	0.6491	0.3149	0.6491	0.8057
No log	4.9	98	0.5743	0.3953	0.5743	0.7578
No log	5.0	100	0.6186	0.3535	0.6186	0.7865
No log	5.1	102	0.6630	0.2670	0.6630	0.8143
No log	5.2	104	0.6243	0.2917	0.6243	0.7901
No log	5.3	106	0.5793	0.3966	0.5793	0.7611
No log	5.4	108	0.6483	0.2000	0.6483	0.8052
No log	5.5	110	0.7670	0.1923	0.7670	0.8758
No log	5.6	112	0.6958	0.1915	0.6958	0.8341
No log	5.7	114	0.6523	0.3439	0.6523	0.8076
No log	5.8	116	0.7403	0.2780	0.7403	0.8604
No log	5.9	118	0.7447	0.2762	0.7447	0.8629
No log	6.0	120	0.7169	0.2637	0.7169	0.8467
No log	6.1	122	0.7285	0.3010	0.7285	0.8535
No log	6.2	124	0.7232	0.3171	0.7232	0.8504
No log	6.3	126	0.7369	0.3367	0.7369	0.8584
No log	6.4	128	0.7702	0.2661	0.7702	0.8776
No log	6.5	130	0.7793	0.2727	0.7793	0.8828
No log	6.6	132	0.7457	0.3394	0.7457	0.8635
No log	6.7	134	0.7378	0.3427	0.7378	0.8590
No log	6.8	136	0.7435	0.3427	0.7435	0.8623
No log	6.9	138	0.7777	0.3067	0.7777	0.8819
No log	7.0	140	0.8127	0.2775	0.8127	0.9015
No log	7.1	142	0.7980	0.3091	0.7980	0.8933
No log	7.2	144	0.8139	0.3091	0.8139	0.9021
No log	7.3	146	0.7664	0.3394	0.7664	0.8754
No log	7.4	148	0.7690	0.2877	0.7690	0.8769
No log	7.5	150	0.7934	0.3004	0.7934	0.8907
No log	7.6	152	0.8239	0.3067	0.8239	0.9077
No log	7.7	154	0.7904	0.3028	0.7904	0.8890
No log	7.8	156	0.7641	0.2621	0.7641	0.8741
No log	7.9	158	0.7994	0.2632	0.7994	0.8941
No log	8.0	160	0.8528	0.3021	0.8528	0.9235
No log	8.1	162	0.8214	0.2618	0.8214	0.9063
No log	8.2	164	0.7635	0.3271	0.7635	0.8738
No log	8.3	166	0.7507	0.2487	0.7507	0.8664
No log	8.4	168	0.7535	0.2941	0.7535	0.8681
No log	8.5	170	0.7505	0.3301	0.7505	0.8663
No log	8.6	172	0.7673	0.3010	0.7673	0.8759
No log	8.7	174	0.7681	0.2579	0.7681	0.8764
No log	8.8	176	0.7713	0.2372	0.7713	0.8782
No log	8.9	178	0.7683	0.2372	0.7683	0.8765
No log	9.0	180	0.7599	0.2372	0.7599	0.8717
No log	9.1	182	0.7513	0.2372	0.7513	0.8668
No log	9.2	184	0.7440	0.3427	0.7440	0.8625
No log	9.3	186	0.7516	0.2381	0.7516	0.8670
No log	9.4	188	0.7506	0.2381	0.7506	0.8664
No log	9.5	190	0.7459	0.2381	0.7459	0.8637
No log	9.6	192	0.7426	0.2381	0.7426	0.8618
No log	9.7	194	0.7378	0.2780	0.7378	0.8590
No log	9.8	196	0.7365	0.3462	0.7365	0.8582
No log	9.9	198	0.7341	0.3744	0.7341	0.8568
No log	10.0	200	0.7328	0.3744	0.7328	0.8560

Framework versions

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