ArabicNewSplits_FineTuningAraBERT_AugV5_k5_task1_organization_fold1

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: 1.0006
• Qwk: 0.5207
• Mse: 1.0006
• Rmse: 1.0003

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	0.0833	2	4.9606	0.0312	4.9606	2.2272
No log	0.1667	4	2.9276	0.0707	2.9276	1.7110
No log	0.25	6	2.3044	0.1556	2.3044	1.5180
No log	0.3333	8	1.9036	0.2258	1.9036	1.3797
No log	0.4167	10	1.3857	0.1806	1.3857	1.1772
No log	0.5	12	1.7253	0.2586	1.7253	1.3135
No log	0.5833	14	1.4305	0.2301	1.4305	1.1960
No log	0.6667	16	1.2784	0.3296	1.2784	1.1307
No log	0.75	18	1.2347	0.3414	1.2347	1.1112
No log	0.8333	20	1.1342	0.3349	1.1342	1.0650
No log	0.9167	22	1.1429	0.3582	1.1429	1.0691
No log	1.0	24	0.9705	0.4283	0.9705	0.9852
No log	1.0833	26	0.8889	0.4653	0.8889	0.9428
No log	1.1667	28	1.0111	0.4112	1.0111	1.0055
No log	1.25	30	1.0631	0.4670	1.0631	1.0311
No log	1.3333	32	0.8982	0.4424	0.8982	0.9477
No log	1.4167	34	0.8290	0.4501	0.8290	0.9105
No log	1.5	36	0.8134	0.4488	0.8134	0.9019
No log	1.5833	38	0.9258	0.5245	0.9258	0.9622
No log	1.6667	40	1.0375	0.5199	1.0375	1.0186
No log	1.75	42	1.2331	0.3474	1.2331	1.1105
No log	1.8333	44	1.0971	0.3779	1.0971	1.0474
No log	1.9167	46	0.8753	0.5472	0.8753	0.9356
No log	2.0	48	0.8362	0.4553	0.8362	0.9145
No log	2.0833	50	0.9431	0.4013	0.9431	0.9711
No log	2.1667	52	0.9346	0.5030	0.9346	0.9667
No log	2.25	54	0.7745	0.5417	0.7745	0.8801
No log	2.3333	56	0.8652	0.5494	0.8652	0.9302
No log	2.4167	58	1.1866	0.3628	1.1866	1.0893
No log	2.5	60	1.2865	0.3171	1.2865	1.1342
No log	2.5833	62	0.9594	0.5215	0.9594	0.9795
No log	2.6667	64	0.7031	0.5379	0.7031	0.8385
No log	2.75	66	0.8242	0.5606	0.8242	0.9079
No log	2.8333	68	0.8559	0.5606	0.8559	0.9252
No log	2.9167	70	0.7720	0.5549	0.7720	0.8786
No log	3.0	72	0.7964	0.5065	0.7964	0.8924
No log	3.0833	74	0.8629	0.5210	0.8629	0.9289
No log	3.1667	76	0.8525	0.5305	0.8525	0.9233
No log	3.25	78	0.8231	0.5984	0.8231	0.9073
No log	3.3333	80	0.8945	0.5049	0.8945	0.9458
No log	3.4167	82	1.0252	0.4463	1.0252	1.0125
No log	3.5	84	1.1201	0.4533	1.1201	1.0583
No log	3.5833	86	1.0473	0.4231	1.0473	1.0234
No log	3.6667	88	0.9130	0.5687	0.9130	0.9555
No log	3.75	90	0.8636	0.5882	0.8636	0.9293
No log	3.8333	92	0.8323	0.6106	0.8323	0.9123
No log	3.9167	94	0.8001	0.6291	0.8001	0.8945
No log	4.0	96	0.8142	0.5526	0.8142	0.9023
No log	4.0833	98	0.8312	0.5453	0.8312	0.9117
No log	4.1667	100	0.8237	0.5704	0.8237	0.9076
No log	4.25	102	0.8202	0.5704	0.8202	0.9056
No log	4.3333	104	0.7796	0.6149	0.7796	0.8829
No log	4.4167	106	0.7856	0.6201	0.7856	0.8863
No log	4.5	108	0.8377	0.5753	0.8377	0.9153
No log	4.5833	110	0.9201	0.4766	0.9201	0.9592
No log	4.6667	112	1.0205	0.4865	1.0205	1.0102
No log	4.75	114	0.9134	0.5408	0.9134	0.9557
No log	4.8333	116	0.7209	0.6580	0.7209	0.8491
No log	4.9167	118	0.6856	0.6267	0.6856	0.8280
No log	5.0	120	0.6909	0.6407	0.6909	0.8312
No log	5.0833	122	0.7800	0.5626	0.7800	0.8832
No log	5.1667	124	0.9435	0.5373	0.9435	0.9713
No log	5.25	126	0.9317	0.5373	0.9317	0.9653
No log	5.3333	128	0.9342	0.5373	0.9342	0.9665
No log	5.4167	130	0.8115	0.5741	0.8115	0.9008
No log	5.5	132	0.7211	0.6442	0.7211	0.8492
No log	5.5833	134	0.7367	0.6227	0.7367	0.8583
No log	5.6667	136	0.7765	0.6142	0.7765	0.8812
No log	5.75	138	0.8685	0.5503	0.8685	0.9319
No log	5.8333	140	0.9554	0.4966	0.9554	0.9774
No log	5.9167	142	0.9164	0.5110	0.9164	0.9573
No log	6.0	144	0.8545	0.5390	0.8545	0.9244
No log	6.0833	146	0.8004	0.5540	0.8004	0.8946
No log	6.1667	148	0.8009	0.5710	0.8009	0.8950
No log	6.25	150	0.8017	0.5565	0.8017	0.8954
No log	6.3333	152	0.8578	0.5281	0.8578	0.9262
No log	6.4167	154	0.8561	0.5281	0.8561	0.9253
No log	6.5	156	0.7718	0.5822	0.7718	0.8785
No log	6.5833	158	0.7471	0.5497	0.7471	0.8644
No log	6.6667	160	0.7597	0.5839	0.7597	0.8716
No log	6.75	162	0.8287	0.5593	0.8287	0.9103
No log	6.8333	164	0.8824	0.5456	0.8824	0.9394
No log	6.9167	166	0.8811	0.5363	0.8811	0.9387
No log	7.0	168	0.8439	0.5423	0.8439	0.9187
No log	7.0833	170	0.8256	0.5423	0.8256	0.9086
No log	7.1667	172	0.8283	0.5481	0.8283	0.9101
No log	7.25	174	0.8390	0.5389	0.8390	0.9160
No log	7.3333	176	0.9273	0.5180	0.9273	0.9630
No log	7.4167	178	1.0427	0.5352	1.0427	1.0211
No log	7.5	180	1.1114	0.5070	1.1114	1.0542
No log	7.5833	182	1.0839	0.4925	1.0839	1.0411
No log	7.6667	184	1.0742	0.5352	1.0742	1.0364
No log	7.75	186	1.0050	0.5055	1.0050	1.0025
No log	7.8333	188	0.9105	0.5525	0.9105	0.9542
No log	7.9167	190	0.8345	0.5525	0.8345	0.9135
No log	8.0	192	0.8002	0.5541	0.8002	0.8946
No log	8.0833	194	0.8248	0.5412	0.8248	0.9082
No log	8.1667	196	0.8705	0.5504	0.8705	0.9330
No log	8.25	198	0.9169	0.5363	0.9169	0.9575
No log	8.3333	200	0.9332	0.5363	0.9332	0.9660
No log	8.4167	202	0.8985	0.5363	0.8985	0.9479
No log	8.5	204	0.8451	0.5423	0.8451	0.9193
No log	8.5833	206	0.8209	0.5423	0.8209	0.9060
No log	8.6667	208	0.8142	0.5341	0.8142	0.9023
No log	8.75	210	0.8258	0.5423	0.8258	0.9087
No log	8.8333	212	0.8403	0.5423	0.8403	0.9167
No log	8.9167	214	0.8550	0.5423	0.8550	0.9247
No log	9.0	216	0.8761	0.5423	0.8761	0.9360
No log	9.0833	218	0.9001	0.5504	0.9001	0.9487
No log	9.1667	220	0.9271	0.5352	0.9271	0.9629
No log	9.25	222	0.9672	0.5352	0.9672	0.9834
No log	9.3333	224	1.0165	0.5285	1.0165	1.0082
No log	9.4167	226	1.0515	0.5429	1.0515	1.0254
No log	9.5	228	1.0585	0.5429	1.0585	1.0288
No log	9.5833	230	1.0517	0.5429	1.0517	1.0255
No log	9.6667	232	1.0456	0.5429	1.0456	1.0226
No log	9.75	234	1.0345	0.5285	1.0345	1.0171
No log	9.8333	236	1.0201	0.5207	1.0201	1.0100
No log	9.9167	238	1.0067	0.5207	1.0067	1.0034
No log	10.0	240	1.0006	0.5207	1.0006	1.0003

Framework versions

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