ArabicNewSplits7_FineTuningAraBERT_run3_AugV5_k13_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: 0.7224
• Qwk: 0.7286
• Mse: 0.7224
• Rmse: 0.8499

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	0.0333	2	6.8324	-0.0169	6.8324	2.6139
No log	0.0667	4	4.4510	0.0397	4.4510	2.1097
No log	0.1	6	3.2660	0.0460	3.2660	1.8072
No log	0.1333	8	2.5362	0.0533	2.5362	1.5925
No log	0.1667	10	2.0784	0.1871	2.0784	1.4417
No log	0.2	12	1.9006	0.1081	1.9006	1.3786
No log	0.2333	14	1.9117	0.1455	1.9117	1.3826
No log	0.2667	16	1.5758	0.2545	1.5758	1.2553
No log	0.3	18	1.4433	0.2735	1.4433	1.2014
No log	0.3333	20	2.2928	0.1528	2.2928	1.5142
No log	0.3667	22	2.2678	0.1389	2.2678	1.5059
No log	0.4	24	1.7736	0.3307	1.7736	1.3318
No log	0.4333	26	1.3818	0.2857	1.3818	1.1755
No log	0.4667	28	1.3965	0.2778	1.3965	1.1818
No log	0.5	30	1.4961	0.1905	1.4961	1.2231
No log	0.5333	32	1.4378	0.2778	1.4378	1.1991
No log	0.5667	34	1.4509	0.2759	1.4509	1.2046
No log	0.6	36	1.5056	0.272	1.5056	1.2270
No log	0.6333	38	1.4648	0.2857	1.4648	1.2103
No log	0.6667	40	1.3119	0.4427	1.3119	1.1454
No log	0.7	42	1.1427	0.5203	1.1427	1.0690
No log	0.7333	44	1.2601	0.3929	1.2601	1.1225
No log	0.7667	46	1.3875	0.4655	1.3875	1.1779
No log	0.8	48	1.2001	0.4874	1.2001	1.0955
No log	0.8333	50	1.0388	0.5909	1.0388	1.0192
No log	0.8667	52	1.1659	0.4928	1.1659	1.0798
No log	0.9	54	1.1605	0.5714	1.1605	1.0773
No log	0.9333	56	1.0542	0.6187	1.0542	1.0268
No log	0.9667	58	1.0281	0.6364	1.0281	1.0140
No log	1.0	60	1.2077	0.5156	1.2077	1.0990
No log	1.0333	62	1.6283	0.2833	1.6283	1.2760
No log	1.0667	64	1.3941	0.4531	1.3941	1.1807
No log	1.1	66	1.1115	0.5778	1.1115	1.0543
No log	1.1333	68	1.0983	0.5522	1.0983	1.0480
No log	1.1667	70	1.1648	0.5	1.1648	1.0792
No log	1.2	72	1.2024	0.5	1.2024	1.0965
No log	1.2333	74	1.1550	0.6099	1.1550	1.0747
No log	1.2667	76	1.1232	0.5833	1.1232	1.0598
No log	1.3	78	1.0775	0.6154	1.0775	1.0380
No log	1.3333	80	1.2845	0.5578	1.2845	1.1333
No log	1.3667	82	1.7811	0.3586	1.7811	1.3346
No log	1.4	84	1.7967	0.2979	1.7967	1.3404
No log	1.4333	86	1.5624	0.4110	1.5624	1.2500
No log	1.4667	88	1.3100	0.5306	1.3100	1.1445
No log	1.5	90	1.0614	0.6395	1.0614	1.0302
No log	1.5333	92	1.0308	0.6536	1.0308	1.0153
No log	1.5667	94	1.0704	0.5616	1.0704	1.0346
No log	1.6	96	0.9943	0.6	0.9943	0.9971
No log	1.6333	98	1.0325	0.6259	1.0325	1.0161
No log	1.6667	100	1.2430	0.5479	1.2430	1.1149
No log	1.7	102	1.3292	0.4865	1.3292	1.1529
No log	1.7333	104	1.1971	0.5811	1.1971	1.0941
No log	1.7667	106	1.0068	0.6259	1.0068	1.0034
No log	1.8	108	0.9591	0.6753	0.9591	0.9794
No log	1.8333	110	0.9458	0.6753	0.9458	0.9725
No log	1.8667	112	0.9218	0.6711	0.9218	0.9601
No log	1.9	114	1.1100	0.5753	1.1100	1.0536
No log	1.9333	116	1.4674	0.4861	1.4674	1.2114
No log	1.9667	118	1.5825	0.4795	1.5825	1.2580
No log	2.0	120	1.4614	0.4966	1.4614	1.2089
No log	2.0333	122	1.0916	0.5571	1.0916	1.0448
No log	2.0667	124	0.7959	0.7211	0.7959	0.8921
No log	2.1	126	0.7568	0.7692	0.7568	0.8699
No log	2.1333	128	0.7565	0.7643	0.7565	0.8698
No log	2.1667	130	0.9433	0.6069	0.9433	0.9713
No log	2.2	132	1.2585	0.5306	1.2585	1.1218
No log	2.2333	134	1.3233	0.5135	1.3233	1.1504
No log	2.2667	136	1.1540	0.5772	1.1540	1.0742
No log	2.3	138	0.9523	0.6207	0.9523	0.9759
No log	2.3333	140	0.9395	0.6620	0.9395	0.9693
No log	2.3667	142	1.0000	0.5882	1.0000	1.0000
No log	2.4	144	0.9856	0.5899	0.9856	0.9928
No log	2.4333	146	1.0112	0.6154	1.0112	1.0056
No log	2.4667	148	0.9667	0.6207	0.9667	0.9832
No log	2.5	150	0.9558	0.6621	0.9558	0.9777
No log	2.5333	152	1.0174	0.6207	1.0174	1.0087
No log	2.5667	154	1.0125	0.625	1.0125	1.0062
No log	2.6	156	0.9263	0.6712	0.9263	0.9624
No log	2.6333	158	0.8871	0.6712	0.8871	0.9419
No log	2.6667	160	0.9379	0.6395	0.9379	0.9684
No log	2.7	162	0.8881	0.6667	0.8881	0.9424
No log	2.7333	164	0.8443	0.6846	0.8443	0.9189
No log	2.7667	166	0.8830	0.6711	0.8830	0.9397
No log	2.8	168	0.9047	0.6993	0.9047	0.9512
No log	2.8333	170	0.9689	0.6667	0.9689	0.9843
No log	2.8667	172	0.9990	0.6569	0.9990	0.9995
No log	2.9	174	1.0167	0.6713	1.0167	1.0083
No log	2.9333	176	0.9758	0.6901	0.9758	0.9878
No log	2.9667	178	0.9069	0.6887	0.9069	0.9523
No log	3.0	180	0.8712	0.7067	0.8712	0.9334
No log	3.0333	182	0.8269	0.7013	0.8269	0.9094
No log	3.0667	184	0.8173	0.7051	0.8173	0.9041
No log	3.1	186	0.8515	0.7134	0.8514	0.9227
No log	3.1333	188	0.7380	0.7436	0.7380	0.8591
No log	3.1667	190	0.7505	0.7451	0.7505	0.8663
No log	3.2	192	0.7213	0.7368	0.7213	0.8493
No log	3.2333	194	0.8384	0.6933	0.8384	0.9156
No log	3.2667	196	0.8644	0.6803	0.8644	0.9298
No log	3.3	198	0.7679	0.7308	0.7679	0.8763
No log	3.3333	200	0.7660	0.7451	0.7660	0.8752
No log	3.3667	202	0.7693	0.7451	0.7693	0.8771
No log	3.4	204	0.8064	0.6986	0.8064	0.8980
No log	3.4333	206	0.8433	0.6806	0.8433	0.9183
No log	3.4667	208	1.0028	0.6301	1.0028	1.0014
No log	3.5	210	1.2523	0.5467	1.2523	1.1191
No log	3.5333	212	1.2305	0.5467	1.2305	1.1093
No log	3.5667	214	0.9985	0.6301	0.9985	0.9993
No log	3.6	216	0.8276	0.7237	0.8276	0.9097
No log	3.6333	218	0.8569	0.6928	0.8569	0.9257
No log	3.6667	220	0.8285	0.7013	0.8285	0.9102
No log	3.7	222	0.7665	0.7432	0.7665	0.8755
No log	3.7333	224	0.8373	0.6621	0.8373	0.9151
No log	3.7667	226	0.9904	0.6443	0.9904	0.9952
No log	3.8	228	1.0247	0.6081	1.0247	1.0123
No log	3.8333	230	0.8323	0.6533	0.8323	0.9123
No log	3.8667	232	0.7157	0.7237	0.7157	0.8460
No log	3.9	234	0.7395	0.7237	0.7395	0.8599
No log	3.9333	236	0.7913	0.7211	0.7913	0.8895
No log	3.9667	238	0.8125	0.7448	0.8125	0.9014
No log	4.0	240	0.8178	0.75	0.8178	0.9043
No log	4.0333	242	0.8373	0.7123	0.8373	0.9151
No log	4.0667	244	0.8286	0.75	0.8286	0.9103
No log	4.1	246	0.8270	0.7222	0.8270	0.9094
No log	4.1333	248	0.8157	0.7432	0.8157	0.9032
No log	4.1667	250	0.7898	0.7211	0.7898	0.8887
No log	4.2	252	0.7385	0.7550	0.7385	0.8593
No log	4.2333	254	0.6709	0.7451	0.6709	0.8191
No log	4.2667	256	0.6444	0.7564	0.6444	0.8027
No log	4.3	258	0.6735	0.7485	0.6735	0.8207
No log	4.3333	260	0.6634	0.7485	0.6634	0.8145
No log	4.3667	262	0.6279	0.7871	0.6279	0.7924
No log	4.4	264	0.6406	0.7792	0.6406	0.8004
No log	4.4333	266	0.6432	0.7792	0.6432	0.8020
No log	4.4667	268	0.6576	0.7682	0.6576	0.8109
No log	4.5	270	0.6357	0.7682	0.6357	0.7973
No log	4.5333	272	0.6581	0.7632	0.6581	0.8112
No log	4.5667	274	0.7118	0.7143	0.7118	0.8437
No log	4.6	276	0.7620	0.6928	0.7620	0.8729
No log	4.6333	278	0.7217	0.7143	0.7217	0.8495
No log	4.6667	280	0.7091	0.7226	0.7091	0.8421
No log	4.7	282	0.6793	0.7582	0.6793	0.8242
No log	4.7333	284	0.6749	0.7712	0.6749	0.8215
No log	4.7667	286	0.6905	0.76	0.6905	0.8310
No log	4.8	288	0.7061	0.7632	0.7061	0.8403
No log	4.8333	290	0.6908	0.7550	0.6908	0.8311
No log	4.8667	292	0.6635	0.7925	0.6635	0.8145
No log	4.9	294	0.6801	0.7436	0.6801	0.8247
No log	4.9333	296	0.6873	0.7643	0.6873	0.8291
No log	4.9667	298	0.6812	0.7771	0.6812	0.8254
No log	5.0	300	0.7098	0.7703	0.7098	0.8425
No log	5.0333	302	0.7608	0.72	0.7608	0.8722
No log	5.0667	304	0.7611	0.7190	0.7611	0.8724
No log	5.1	306	0.6619	0.7403	0.6619	0.8136
No log	5.1333	308	0.6085	0.7821	0.6085	0.7801
No log	5.1667	310	0.6553	0.7901	0.6553	0.8095
No log	5.2	312	0.6146	0.7826	0.6146	0.7839
No log	5.2333	314	0.5518	0.8049	0.5518	0.7429
No log	5.2667	316	0.6238	0.7772	0.6238	0.7898
No log	5.3	318	0.6885	0.7821	0.6885	0.8298
No log	5.3333	320	0.6350	0.7485	0.6350	0.7969
No log	5.3667	322	0.6418	0.7564	0.6418	0.8011
No log	5.4	324	0.6263	0.7949	0.6263	0.7914
No log	5.4333	326	0.6684	0.7871	0.6684	0.8176
No log	5.4667	328	0.6821	0.7582	0.6821	0.8259
No log	5.5	330	0.6821	0.7763	0.6821	0.8259
No log	5.5333	332	0.6703	0.8025	0.6703	0.8187
No log	5.5667	334	0.6519	0.8025	0.6519	0.8074
No log	5.6	336	0.6432	0.8	0.6432	0.8020
No log	5.6333	338	0.6553	0.7763	0.6553	0.8095
No log	5.6667	340	0.7328	0.6980	0.7328	0.8560
No log	5.7	342	0.8113	0.6709	0.8113	0.9007
No log	5.7333	344	0.7661	0.7067	0.7661	0.8753
No log	5.7667	346	0.6702	0.75	0.6702	0.8186
No log	5.8	348	0.6337	0.7949	0.6337	0.7961
No log	5.8333	350	0.6422	0.8	0.6422	0.8014
No log	5.8667	352	0.6313	0.8	0.6313	0.7946
No log	5.9	354	0.6032	0.7949	0.6032	0.7766
No log	5.9333	356	0.5969	0.8	0.5969	0.7726
No log	5.9667	358	0.5738	0.8025	0.5738	0.7575
No log	6.0	360	0.6970	0.7436	0.6970	0.8348
No log	6.0333	362	0.8364	0.6968	0.8364	0.9146
No log	6.0667	364	0.7388	0.7436	0.7388	0.8595
No log	6.1	366	0.7055	0.7273	0.7055	0.8399
No log	6.1333	368	0.7155	0.7712	0.7155	0.8459
No log	6.1667	370	0.7615	0.72	0.7615	0.8726
No log	6.2	372	0.8303	0.6351	0.8303	0.9112
No log	6.2333	374	0.8467	0.6309	0.8467	0.9202
No log	6.2667	376	0.8513	0.6490	0.8513	0.9226
No log	6.3	378	0.7153	0.7643	0.7153	0.8457
No log	6.3333	380	0.6406	0.7826	0.6406	0.8004
No log	6.3667	382	0.6362	0.7901	0.6362	0.7977
No log	6.4	384	0.6484	0.7826	0.6484	0.8052
No log	6.4333	386	0.6413	0.7826	0.6413	0.8008
No log	6.4667	388	0.6560	0.8075	0.6560	0.8099
No log	6.5	390	0.7055	0.7468	0.7055	0.8400
No log	6.5333	392	0.7255	0.7320	0.7255	0.8518
No log	6.5667	394	0.6949	0.7922	0.6949	0.8336
No log	6.6	396	0.6823	0.7792	0.6823	0.8260
No log	6.6333	398	0.6843	0.7613	0.6843	0.8272
No log	6.6667	400	0.6294	0.8025	0.6294	0.7933
No log	6.7	402	0.6003	0.8075	0.6003	0.7748
No log	6.7333	404	0.5737	0.7925	0.5737	0.7575
No log	6.7667	406	0.5853	0.8025	0.5853	0.7651
No log	6.8	408	0.6106	0.8026	0.6106	0.7814
No log	6.8333	410	0.6332	0.8026	0.6332	0.7957
No log	6.8667	412	0.6627	0.8026	0.6627	0.8140
No log	6.9	414	0.6677	0.7838	0.6677	0.8171
No log	6.9333	416	0.6535	0.7771	0.6535	0.8084
No log	6.9667	418	0.7176	0.7558	0.7176	0.8471
No log	7.0	420	0.8058	0.6905	0.8058	0.8977
No log	7.0333	422	0.8238	0.6790	0.8238	0.9076
No log	7.0667	424	0.8111	0.6575	0.8111	0.9006
No log	7.1	426	0.8081	0.6714	0.8081	0.8990
No log	7.1333	428	0.8007	0.7153	0.8007	0.8948
No log	7.1667	430	0.7918	0.7286	0.7918	0.8899
No log	7.2	432	0.8085	0.6761	0.8085	0.8992
No log	7.2333	434	0.7387	0.7042	0.7387	0.8595
No log	7.2667	436	0.6940	0.7639	0.6940	0.8331
No log	7.3	438	0.7081	0.7703	0.7081	0.8415
No log	7.3333	440	0.7049	0.7606	0.7049	0.8396
No log	7.3667	442	0.7393	0.7042	0.7393	0.8598
No log	7.4	444	0.8061	0.6525	0.8061	0.8978
No log	7.4333	446	0.8094	0.6525	0.8094	0.8997
No log	7.4667	448	0.7833	0.6912	0.7833	0.8850
No log	7.5	450	0.8039	0.6815	0.8039	0.8966
No log	7.5333	452	0.8806	0.5581	0.8806	0.9384
No log	7.5667	454	0.8347	0.6212	0.8347	0.9136
No log	7.6	456	0.7359	0.7482	0.7359	0.8578
No log	7.6333	458	0.7711	0.6620	0.7711	0.8781
No log	7.6667	460	0.7717	0.6713	0.7717	0.8785
No log	7.7	462	0.6997	0.7273	0.6997	0.8365
No log	7.7333	464	0.6816	0.7724	0.6816	0.8256
No log	7.7667	466	0.6775	0.7671	0.6775	0.8231
No log	7.8	468	0.6674	0.7815	0.6674	0.8169
No log	7.8333	470	0.6564	0.7534	0.6564	0.8102
No log	7.8667	472	0.7262	0.6806	0.7262	0.8522
No log	7.9	474	0.7347	0.6806	0.7347	0.8572
No log	7.9333	476	0.6853	0.7183	0.6853	0.8278
No log	7.9667	478	0.6674	0.7518	0.6674	0.8169
No log	8.0	480	0.6383	0.8077	0.6383	0.7989
No log	8.0333	482	0.6160	0.8025	0.6160	0.7848
No log	8.0667	484	0.6074	0.8077	0.6074	0.7794
No log	8.1	486	0.6210	0.7871	0.6210	0.7880
No log	8.1333	488	0.6449	0.7733	0.6449	0.8031
No log	8.1667	490	0.6536	0.7682	0.6536	0.8085
No log	8.2	492	0.6697	0.7682	0.6697	0.8183
No log	8.2333	494	0.6918	0.7368	0.6918	0.8317
No log	8.2667	496	0.7155	0.7260	0.7155	0.8459
No log	8.3	498	0.7127	0.7467	0.7127	0.8442
0.4158	8.3333	500	0.7142	0.7550	0.7142	0.8451
0.4158	8.3667	502	0.7176	0.7467	0.7176	0.8471
0.4158	8.4	504	0.7500	0.7260	0.7500	0.8660
0.4158	8.4333	506	0.7438	0.6897	0.7438	0.8625
0.4158	8.4667	508	0.7229	0.7361	0.7229	0.8503
0.4158	8.5	510	0.7224	0.7286	0.7224	0.8499

Framework versions

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