ArabicNewSplits6_FineTuningAraBERTFreeze_run1_AugV5_k4_task1_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: 1.0144
• Qwk: 0.6009
• Mse: 1.0144
• Rmse: 1.0072

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: 16
• eval_batch_size: 16
• 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	0.1667	2	7.8504	-0.0474	7.8504	2.8019
No log	0.3333	4	5.5472	-0.0349	5.5472	2.3553
No log	0.5	6	3.8843	0.0371	3.8843	1.9709
No log	0.6667	8	2.9161	0.0896	2.9161	1.7076
No log	0.8333	10	2.1897	0.1953	2.1897	1.4798
No log	1.0	12	1.6691	0.1530	1.6691	1.2919
No log	1.1667	14	1.4154	0.2548	1.4154	1.1897
No log	1.3333	16	1.2624	0.2307	1.2624	1.1236
No log	1.5	18	1.1510	0.2704	1.1510	1.0728
No log	1.6667	20	1.0483	0.3729	1.0483	1.0239
No log	1.8333	22	0.9961	0.4087	0.9961	0.9980
No log	2.0	24	1.0064	0.4737	1.0064	1.0032
No log	2.1667	26	1.0270	0.5002	1.0270	1.0134
No log	2.3333	28	0.9855	0.4939	0.9855	0.9927
No log	2.5	30	0.9698	0.5071	0.9698	0.9848
No log	2.6667	32	0.9489	0.5009	0.9489	0.9741
No log	2.8333	34	0.9485	0.5009	0.9485	0.9739
No log	3.0	36	0.9365	0.4970	0.9365	0.9677
No log	3.1667	38	0.9047	0.5016	0.9047	0.9511
No log	3.3333	40	0.8703	0.5158	0.8703	0.9329
No log	3.5	42	0.8604	0.5219	0.8604	0.9276
No log	3.6667	44	0.8712	0.5425	0.8712	0.9334
No log	3.8333	46	0.9398	0.5129	0.9398	0.9694
No log	4.0	48	1.0637	0.4529	1.0637	1.0314
No log	4.1667	50	1.0818	0.4288	1.0818	1.0401
No log	4.3333	52	1.0229	0.4664	1.0229	1.0114
No log	4.5	54	1.0118	0.4517	1.0118	1.0059
No log	4.6667	56	0.9582	0.5040	0.9582	0.9789
No log	4.8333	58	0.8679	0.5694	0.8679	0.9316
No log	5.0	60	0.8116	0.5972	0.8116	0.9009
No log	5.1667	62	0.8200	0.5901	0.8200	0.9055
No log	5.3333	64	0.9309	0.5276	0.9309	0.9648
No log	5.5	66	0.9445	0.5247	0.9445	0.9719
No log	5.6667	68	0.8464	0.5961	0.8464	0.9200
No log	5.8333	70	0.7321	0.6471	0.7321	0.8556
No log	6.0	72	0.7196	0.6539	0.7196	0.8483
No log	6.1667	74	0.7149	0.6631	0.7149	0.8455
No log	6.3333	76	0.7573	0.6355	0.7573	0.8703
No log	6.5	78	0.9272	0.6024	0.9272	0.9629
No log	6.6667	80	1.0294	0.5398	1.0294	1.0146
No log	6.8333	82	0.9545	0.5822	0.9545	0.9770
No log	7.0	84	0.8433	0.6384	0.8433	0.9183
No log	7.1667	86	0.7439	0.6527	0.7439	0.8625
No log	7.3333	88	0.7019	0.6572	0.7019	0.8378
No log	7.5	90	0.6943	0.6317	0.6943	0.8332
No log	7.6667	92	0.7003	0.6845	0.7003	0.8368
No log	7.8333	94	0.7529	0.6535	0.7529	0.8677
No log	8.0	96	0.7927	0.6505	0.7927	0.8903
No log	8.1667	98	0.8457	0.6357	0.8457	0.9196
No log	8.3333	100	0.7708	0.6794	0.7708	0.8780
No log	8.5	102	0.7431	0.6650	0.7431	0.8620
No log	8.6667	104	0.7692	0.6618	0.7692	0.8771
No log	8.8333	106	0.7622	0.6794	0.7622	0.8730
No log	9.0	108	0.7156	0.6676	0.7156	0.8460
No log	9.1667	110	0.7099	0.6762	0.7099	0.8426
No log	9.3333	112	0.7114	0.6762	0.7114	0.8434
No log	9.5	114	0.7109	0.7128	0.7109	0.8431
No log	9.6667	116	0.7094	0.6831	0.7094	0.8422
No log	9.8333	118	0.7261	0.6854	0.7261	0.8521
No log	10.0	120	0.7479	0.6863	0.7479	0.8648
No log	10.1667	122	0.7404	0.6896	0.7404	0.8604
No log	10.3333	124	0.6989	0.6858	0.6989	0.8360
No log	10.5	126	0.6992	0.6844	0.6992	0.8362
No log	10.6667	128	0.6894	0.7045	0.6894	0.8303
No log	10.8333	130	0.7026	0.7028	0.7026	0.8382
No log	11.0	132	0.7723	0.6767	0.7723	0.8788
No log	11.1667	134	0.8086	0.6431	0.8086	0.8992
No log	11.3333	136	0.7658	0.6795	0.7658	0.8751
No log	11.5	138	0.7378	0.6914	0.7378	0.8590
No log	11.6667	140	0.7332	0.6914	0.7332	0.8563
No log	11.8333	142	0.7403	0.6914	0.7403	0.8604
No log	12.0	144	0.7334	0.7158	0.7334	0.8564
No log	12.1667	146	0.7430	0.6950	0.7430	0.8620
No log	12.3333	148	0.7463	0.7094	0.7463	0.8639
No log	12.5	150	0.7528	0.6749	0.7528	0.8676
No log	12.6667	152	0.7558	0.6594	0.7558	0.8693
No log	12.8333	154	0.7824	0.6588	0.7824	0.8845
No log	13.0	156	0.7690	0.6570	0.7690	0.8769
No log	13.1667	158	0.7601	0.6640	0.7601	0.8718
No log	13.3333	160	0.7529	0.6941	0.7529	0.8677
No log	13.5	162	0.7552	0.6969	0.7552	0.8690
No log	13.6667	164	0.7606	0.6969	0.7606	0.8721
No log	13.8333	166	0.7672	0.7006	0.7672	0.8759
No log	14.0	168	0.8019	0.6828	0.8019	0.8955
No log	14.1667	170	0.9166	0.5955	0.9166	0.9574
No log	14.3333	172	0.9193	0.5976	0.9193	0.9588
No log	14.5	174	0.8294	0.6671	0.8294	0.9107
No log	14.6667	176	0.7725	0.7006	0.7725	0.8789
No log	14.8333	178	0.8553	0.6393	0.8553	0.9248
No log	15.0	180	0.9336	0.5867	0.9336	0.9662
No log	15.1667	182	0.8723	0.6386	0.8723	0.9340
No log	15.3333	184	0.7773	0.6771	0.7773	0.8817
No log	15.5	186	0.8034	0.6619	0.8034	0.8963
No log	15.6667	188	0.8257	0.6362	0.8257	0.9087
No log	15.8333	190	0.7717	0.6772	0.7717	0.8785
No log	16.0	192	0.7506	0.6965	0.7506	0.8664
No log	16.1667	194	0.7300	0.6911	0.7300	0.8544
No log	16.3333	196	0.7430	0.7130	0.7430	0.8620
No log	16.5	198	0.7989	0.6569	0.7989	0.8938
No log	16.6667	200	0.8038	0.6692	0.8038	0.8966
No log	16.8333	202	0.7632	0.7233	0.7632	0.8736
No log	17.0	204	0.7660	0.7275	0.7660	0.8752
No log	17.1667	206	0.7964	0.6997	0.7964	0.8924
No log	17.3333	208	0.7789	0.7280	0.7789	0.8826
No log	17.5	210	0.7872	0.7217	0.7872	0.8872
No log	17.6667	212	0.8683	0.6560	0.8683	0.9318
No log	17.8333	214	0.9608	0.6009	0.9608	0.9802
No log	18.0	216	1.0572	0.5561	1.0572	1.0282
No log	18.1667	218	1.0341	0.5755	1.0341	1.0169
No log	18.3333	220	0.9318	0.6042	0.9318	0.9653
No log	18.5	222	0.8377	0.7057	0.8377	0.9152
No log	18.6667	224	0.8381	0.6602	0.8381	0.9155
No log	18.8333	226	0.8380	0.6559	0.8380	0.9154
No log	19.0	228	0.8527	0.7086	0.8527	0.9234
No log	19.1667	230	0.9368	0.6152	0.9368	0.9679
No log	19.3333	232	0.9913	0.5883	0.9913	0.9956
No log	19.5	234	0.9740	0.5909	0.9740	0.9869
No log	19.6667	236	0.8883	0.6758	0.8883	0.9425
No log	19.8333	238	0.8567	0.6940	0.8567	0.9256
No log	20.0	240	0.8475	0.6894	0.8475	0.9206
No log	20.1667	242	0.8497	0.6807	0.8497	0.9218
No log	20.3333	244	0.8734	0.6569	0.8734	0.9346
No log	20.5	246	0.9395	0.6391	0.9395	0.9693
No log	20.6667	248	0.9702	0.6116	0.9702	0.9850
No log	20.8333	250	0.9639	0.6050	0.9639	0.9818
No log	21.0	252	1.0380	0.5742	1.0380	1.0188
No log	21.1667	254	1.0504	0.5937	1.0504	1.0249
No log	21.3333	256	0.9729	0.6050	0.9729	0.9864
No log	21.5	258	0.8878	0.6594	0.8878	0.9422
No log	21.6667	260	0.8540	0.6837	0.8540	0.9241
No log	21.8333	262	0.8779	0.6594	0.8779	0.9370
No log	22.0	264	0.9245	0.6377	0.9245	0.9615
No log	22.1667	266	0.9589	0.6019	0.9589	0.9792
No log	22.3333	268	0.9094	0.6491	0.9094	0.9536
No log	22.5	270	0.9176	0.6407	0.9176	0.9579
No log	22.6667	272	0.9467	0.6279	0.9467	0.9730
No log	22.8333	274	0.9305	0.6317	0.9305	0.9646
No log	23.0	276	0.9104	0.6449	0.9104	0.9541
No log	23.1667	278	0.9366	0.6386	0.9366	0.9678
No log	23.3333	280	1.0228	0.5926	1.0228	1.0113
No log	23.5	282	1.0441	0.5874	1.0441	1.0218
No log	23.6667	284	0.9990	0.6108	0.9990	0.9995
No log	23.8333	286	0.9145	0.6318	0.9145	0.9563
No log	24.0	288	0.8827	0.6689	0.8827	0.9395
No log	24.1667	290	0.9189	0.6332	0.9189	0.9586
No log	24.3333	292	0.9915	0.6083	0.9915	0.9957
No log	24.5	294	1.0247	0.5759	1.0247	1.0123
No log	24.6667	296	1.0487	0.5667	1.0487	1.0241
No log	24.8333	298	1.0137	0.6054	1.0137	1.0068
No log	25.0	300	0.9287	0.6165	0.9287	0.9637
No log	25.1667	302	0.9136	0.6522	0.9136	0.9558
No log	25.3333	304	0.9243	0.6466	0.9243	0.9614
No log	25.5	306	0.9190	0.6406	0.9190	0.9586
No log	25.6667	308	0.9488	0.6220	0.9488	0.9740
No log	25.8333	310	0.9993	0.5914	0.9993	0.9997
No log	26.0	312	1.0849	0.5881	1.0849	1.0416
No log	26.1667	314	1.1688	0.5648	1.1688	1.0811
No log	26.3333	316	1.1585	0.5815	1.1585	1.0763
No log	26.5	318	1.1304	0.5756	1.1304	1.0632
No log	26.6667	320	1.0927	0.5749	1.0927	1.0453
No log	26.8333	322	1.0293	0.5844	1.0293	1.0146
No log	27.0	324	0.9985	0.5980	0.9985	0.9993
No log	27.1667	326	1.0041	0.5882	1.0041	1.0021
No log	27.3333	328	1.0892	0.5812	1.0892	1.0436
No log	27.5	330	1.1745	0.5663	1.1745	1.0837
No log	27.6667	332	1.1666	0.5736	1.1666	1.0801
No log	27.8333	334	1.0871	0.5846	1.0871	1.0427
No log	28.0	336	1.0268	0.5837	1.0268	1.0133
No log	28.1667	338	0.9589	0.6139	0.9589	0.9792
No log	28.3333	340	0.9547	0.5923	0.9547	0.9771
No log	28.5	342	1.0117	0.5845	1.0117	1.0058
No log	28.6667	344	1.0778	0.5759	1.0778	1.0382
No log	28.8333	346	1.0613	0.5694	1.0613	1.0302
No log	29.0	348	1.0266	0.5776	1.0266	1.0132
No log	29.1667	350	0.9931	0.5965	0.9931	0.9966
No log	29.3333	352	1.0163	0.5864	1.0163	1.0081
No log	29.5	354	1.0049	0.5985	1.0049	1.0024
No log	29.6667	356	1.0099	0.5898	1.0099	1.0049
No log	29.8333	358	0.9809	0.6036	0.9809	0.9904
No log	30.0	360	0.9587	0.6094	0.9587	0.9791
No log	30.1667	362	0.9736	0.6062	0.9736	0.9867
No log	30.3333	364	1.0287	0.5791	1.0287	1.0142
No log	30.5	366	1.0234	0.5789	1.0234	1.0116
No log	30.6667	368	1.0098	0.5997	1.0098	1.0049
No log	30.8333	370	1.0298	0.5939	1.0298	1.0148
No log	31.0	372	1.0276	0.5947	1.0276	1.0137
No log	31.1667	374	0.9883	0.6224	0.9883	0.9941
No log	31.3333	376	0.9708	0.6427	0.9708	0.9853
No log	31.5	378	1.0116	0.6202	1.0116	1.0058
No log	31.6667	380	1.0674	0.6126	1.0674	1.0331
No log	31.8333	382	1.0675	0.6305	1.0675	1.0332
No log	32.0	384	1.0276	0.6305	1.0276	1.0137
No log	32.1667	386	0.9722	0.6342	0.9722	0.9860
No log	32.3333	388	0.9381	0.6427	0.9381	0.9685
No log	32.5	390	0.9103	0.6476	0.9103	0.9541
No log	32.6667	392	0.9077	0.6476	0.9077	0.9527
No log	32.8333	394	0.9469	0.6441	0.9469	0.9731
No log	33.0	396	1.0310	0.6340	1.0310	1.0154
No log	33.1667	398	1.0983	0.6228	1.0983	1.0480
No log	33.3333	400	1.0869	0.5791	1.0869	1.0426
No log	33.5	402	1.0386	0.5824	1.0386	1.0191
No log	33.6667	404	1.0210	0.5863	1.0210	1.0104
No log	33.8333	406	1.0384	0.5910	1.0384	1.0190
No log	34.0	408	1.0808	0.5769	1.0808	1.0396
No log	34.1667	410	1.0982	0.5707	1.0982	1.0479
No log	34.3333	412	1.0495	0.6001	1.0495	1.0244
No log	34.5	414	0.9700	0.6062	0.9700	0.9849
No log	34.6667	416	0.9136	0.6485	0.9136	0.9558
No log	34.8333	418	0.9095	0.6269	0.9095	0.9537
No log	35.0	420	0.9446	0.6271	0.9446	0.9719
No log	35.1667	422	1.0100	0.6079	1.0100	1.0050
No log	35.3333	424	1.0145	0.6168	1.0145	1.0072
No log	35.5	426	0.9669	0.6270	0.9669	0.9833
No log	35.6667	428	0.9403	0.6277	0.9403	0.9697
No log	35.8333	430	0.9673	0.6167	0.9673	0.9835
No log	36.0	432	1.0043	0.6263	1.0043	1.0021
No log	36.1667	434	1.0058	0.6263	1.0058	1.0029
No log	36.3333	436	1.0113	0.6263	1.0113	1.0057
No log	36.5	438	1.0395	0.6178	1.0395	1.0195
No log	36.6667	440	1.0593	0.6084	1.0593	1.0292
No log	36.8333	442	1.0344	0.5906	1.0344	1.0171
No log	37.0	444	0.9968	0.6074	0.9968	0.9984
No log	37.1667	446	1.0180	0.5960	1.0180	1.0089
No log	37.3333	448	1.0105	0.5994	1.0105	1.0052
No log	37.5	450	1.0228	0.6028	1.0228	1.0113
No log	37.6667	452	1.0286	0.6016	1.0286	1.0142
No log	37.8333	454	1.0282	0.6297	1.0282	1.0140
No log	38.0	456	0.9694	0.6512	0.9694	0.9846
No log	38.1667	458	0.9205	0.6487	0.9205	0.9594
No log	38.3333	460	0.9298	0.6364	0.9298	0.9643
No log	38.5	462	0.9661	0.6512	0.9661	0.9829
No log	38.6667	464	0.9964	0.5951	0.9964	0.9982
No log	38.8333	466	0.9923	0.6172	0.9923	0.9962
No log	39.0	468	0.9416	0.6186	0.9416	0.9703
No log	39.1667	470	0.9335	0.6303	0.9335	0.9662
No log	39.3333	472	0.9717	0.6085	0.9717	0.9858
No log	39.5	474	1.0309	0.5845	1.0309	1.0153
No log	39.6667	476	1.0192	0.5932	1.0192	1.0096
No log	39.8333	478	0.9833	0.6194	0.9833	0.9916
No log	40.0	480	0.9695	0.6290	0.9695	0.9846
No log	40.1667	482	0.9723	0.6290	0.9723	0.9860
No log	40.3333	484	0.9289	0.6548	0.9289	0.9638
No log	40.5	486	0.9170	0.6485	0.9170	0.9576
No log	40.6667	488	0.9191	0.6464	0.9191	0.9587
No log	40.8333	490	0.9463	0.6548	0.9463	0.9728
No log	41.0	492	1.0111	0.6361	1.0111	1.0055
No log	41.1667	494	1.0747	0.5996	1.0747	1.0367
No log	41.3333	496	1.0871	0.5840	1.0871	1.0427
No log	41.5	498	1.0342	0.6008	1.0342	1.0170
0.5439	41.6667	500	1.0019	0.6070	1.0019	1.0010
0.5439	41.8333	502	1.0256	0.5982	1.0256	1.0127
0.5439	42.0	504	1.0783	0.5920	1.0783	1.0384
0.5439	42.1667	506	1.0997	0.5836	1.0997	1.0487
0.5439	42.3333	508	1.0857	0.5836	1.0857	1.0420
0.5439	42.5	510	1.0144	0.6009	1.0144	1.0072

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.1+cu121
• Datasets 3.2.0
• Tokenizers 0.19.1