ArabicNewSplits7_OSS_usingWellWrittenEssays_FineTuningAraBERT_run2_AugV5_k1_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: 0.8721
• Qwk: 0.6901
• Mse: 0.8721
• Rmse: 0.9339

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	0.25	2	6.9756	0.0299	6.9756	2.6411
No log	0.5	4	4.9600	0.0519	4.9600	2.2271
No log	0.75	6	2.9865	0.0848	2.9865	1.7281
No log	1.0	8	2.1997	0.1286	2.1997	1.4831
No log	1.25	10	1.8600	0.2124	1.8600	1.3638
No log	1.5	12	1.6690	0.1905	1.6690	1.2919
No log	1.75	14	1.6731	0.1509	1.6731	1.2935
No log	2.0	16	2.0864	0.1311	2.0864	1.4444
No log	2.25	18	2.1462	0.1803	2.1462	1.4650
No log	2.5	20	1.9214	0.1622	1.9214	1.3861
No log	2.75	22	1.4132	0.3091	1.4132	1.1888
No log	3.0	24	1.2415	0.3540	1.2415	1.1142
No log	3.25	26	1.2575	0.3793	1.2575	1.1214
No log	3.5	28	1.2397	0.3717	1.2397	1.1134
No log	3.75	30	1.3851	0.2957	1.3851	1.1769
No log	4.0	32	1.4620	0.2393	1.4620	1.2091
No log	4.25	34	1.3767	0.2521	1.3767	1.1733
No log	4.5	36	1.1201	0.4848	1.1201	1.0584
No log	4.75	38	1.0144	0.6061	1.0144	1.0072
No log	5.0	40	1.0413	0.6	1.0413	1.0204
No log	5.25	42	1.2394	0.4706	1.2394	1.1133
No log	5.5	44	1.3671	0.3833	1.3671	1.1692
No log	5.75	46	1.0641	0.5512	1.0641	1.0316
No log	6.0	48	1.0159	0.6533	1.0159	1.0079
No log	6.25	50	1.1652	0.6154	1.1652	1.0795
No log	6.5	52	1.2106	0.5211	1.2106	1.1003
No log	6.75	54	1.0655	0.6099	1.0655	1.0322
No log	7.0	56	1.0535	0.5857	1.0535	1.0264
No log	7.25	58	0.9915	0.6056	0.9915	0.9958
No log	7.5	60	0.8672	0.6528	0.8672	0.9312
No log	7.75	62	0.8603	0.7123	0.8603	0.9275
No log	8.0	64	0.9053	0.6933	0.9053	0.9515
No log	8.25	66	0.9247	0.7285	0.9247	0.9616
No log	8.5	68	0.9744	0.7020	0.9744	0.9871
No log	8.75	70	0.9792	0.6892	0.9792	0.9895
No log	9.0	72	1.0108	0.6528	1.0108	1.0054
No log	9.25	74	1.0288	0.6111	1.0288	1.0143
No log	9.5	76	0.9550	0.6573	0.9550	0.9772
No log	9.75	78	0.8953	0.6993	0.8953	0.9462
No log	10.0	80	0.8908	0.6892	0.8908	0.9438
No log	10.25	82	0.8809	0.6892	0.8809	0.9386
No log	10.5	84	0.8808	0.75	0.8808	0.9385
No log	10.75	86	1.0151	0.6328	1.0151	1.0075
No log	11.0	88	1.1088	0.6220	1.1088	1.0530
No log	11.25	90	1.0300	0.6309	1.0300	1.0149
No log	11.5	92	0.8771	0.6846	0.8771	0.9366
No log	11.75	94	0.7461	0.7483	0.7461	0.8638
No log	12.0	96	0.7069	0.7034	0.7069	0.8408
No log	12.25	98	0.7536	0.7114	0.7536	0.8681
No log	12.5	100	0.7916	0.72	0.7916	0.8897
No log	12.75	102	0.8244	0.7027	0.8244	0.9080
No log	13.0	104	1.0762	0.6590	1.0762	1.0374
No log	13.25	106	1.2229	0.5988	1.2229	1.1059
No log	13.5	108	1.1069	0.6093	1.1069	1.0521
No log	13.75	110	0.9202	0.6475	0.9202	0.9593
No log	14.0	112	0.7877	0.7042	0.7877	0.8875
No log	14.25	114	0.7905	0.7123	0.7905	0.8891
No log	14.5	116	0.8162	0.7152	0.8162	0.9034
No log	14.75	118	0.8505	0.6974	0.8505	0.9222
No log	15.0	120	0.9227	0.7456	0.9227	0.9606
No log	15.25	122	1.1740	0.6404	1.1740	1.0835
No log	15.5	124	1.2146	0.6102	1.2146	1.1021
No log	15.75	126	0.9490	0.6708	0.9490	0.9741
No log	16.0	128	0.6853	0.7534	0.6853	0.8278
No log	16.25	130	0.7250	0.6901	0.7250	0.8515
No log	16.5	132	0.7346	0.7083	0.7346	0.8571
No log	16.75	134	0.6563	0.7397	0.6563	0.8101
No log	17.0	136	0.7895	0.7237	0.7895	0.8885
No log	17.25	138	1.1174	0.6108	1.1174	1.0571
No log	17.5	140	1.1610	0.6012	1.1610	1.0775
No log	17.75	142	0.9747	0.6667	0.9747	0.9872
No log	18.0	144	0.9051	0.6667	0.9051	0.9514
No log	18.25	146	0.8716	0.6573	0.8716	0.9336
No log	18.5	148	0.7804	0.6950	0.7804	0.8834
No log	18.75	150	0.7638	0.7114	0.7638	0.8739
No log	19.0	152	0.7469	0.7730	0.7469	0.8642
No log	19.25	154	0.7300	0.7215	0.7300	0.8544
No log	19.5	156	0.7453	0.7329	0.7453	0.8633
No log	19.75	158	0.6881	0.7692	0.6881	0.8295
No log	20.0	160	0.6113	0.7733	0.6113	0.7818
No log	20.25	162	0.6071	0.7550	0.6071	0.7791
No log	20.5	164	0.6632	0.7625	0.6632	0.8144
No log	20.75	166	0.7326	0.7719	0.7326	0.8559
No log	21.0	168	0.8309	0.7425	0.8309	0.9115
No log	21.25	170	0.7534	0.7421	0.7534	0.8680
No log	21.5	172	0.7135	0.7285	0.7135	0.8447
No log	21.75	174	0.7754	0.7285	0.7755	0.8806
No log	22.0	176	0.9541	0.7126	0.9541	0.9768
No log	22.25	178	0.9315	0.7368	0.9315	0.9651
No log	22.5	180	0.7876	0.7226	0.7876	0.8874
No log	22.75	182	0.6750	0.7368	0.6750	0.8216
No log	23.0	184	0.6328	0.7564	0.6328	0.7955
No log	23.25	186	0.6396	0.7904	0.6396	0.7998
No log	23.5	188	0.7182	0.7719	0.7182	0.8475
No log	23.75	190	0.7450	0.7624	0.7450	0.8631
No log	24.0	192	0.6891	0.7470	0.6891	0.8301
No log	24.25	194	0.6789	0.7190	0.6789	0.8240
No log	24.5	196	0.7257	0.7044	0.7257	0.8519
No log	24.75	198	0.7577	0.7170	0.7577	0.8705
No log	25.0	200	0.8112	0.7349	0.8112	0.9007
No log	25.25	202	0.8974	0.7159	0.8974	0.9473
No log	25.5	204	0.8455	0.7239	0.8455	0.9195
No log	25.75	206	0.7418	0.7133	0.7418	0.8613
No log	26.0	208	0.7058	0.7034	0.7058	0.8401
No log	26.25	210	0.6901	0.7285	0.6901	0.8307
No log	26.5	212	0.6847	0.7448	0.6847	0.8274
No log	26.75	214	0.7672	0.7417	0.7672	0.8759
No log	27.0	216	0.9528	0.6711	0.9528	0.9761
No log	27.25	218	1.1034	0.6867	1.1034	1.0504
No log	27.5	220	1.0527	0.6790	1.0527	1.0260
No log	27.75	222	0.8446	0.6887	0.8446	0.9190
No log	28.0	224	0.6662	0.7568	0.6662	0.8162
No log	28.25	226	0.6467	0.7417	0.6467	0.8042
No log	28.5	228	0.6783	0.7285	0.6783	0.8236
No log	28.75	230	0.7425	0.7296	0.7425	0.8617
No log	29.0	232	0.8201	0.7273	0.8201	0.9056
No log	29.25	234	0.8693	0.6871	0.8693	0.9324
No log	29.5	236	0.8668	0.6933	0.8668	0.9310
No log	29.75	238	0.8399	0.6857	0.8399	0.9164
No log	30.0	240	0.8376	0.6667	0.8376	0.9152
No log	30.25	242	0.8430	0.6765	0.8430	0.9182
No log	30.5	244	0.8484	0.6567	0.8484	0.9211
No log	30.75	246	0.9014	0.6571	0.9014	0.9494
No log	31.0	248	0.9510	0.6216	0.9510	0.9752
No log	31.25	250	0.9296	0.6621	0.9296	0.9642
No log	31.5	252	0.8746	0.6906	0.8746	0.9352
No log	31.75	254	0.8408	0.7042	0.8408	0.9170
No log	32.0	256	0.8395	0.7042	0.8395	0.9162
No log	32.25	258	0.8648	0.7133	0.8648	0.9300
No log	32.5	260	0.8904	0.7013	0.8904	0.9436
No log	32.75	262	0.8704	0.7215	0.8704	0.9329
No log	33.0	264	0.8434	0.7125	0.8434	0.9184
No log	33.25	266	0.7901	0.7355	0.7901	0.8889
No log	33.5	268	0.7730	0.7059	0.7730	0.8792
No log	33.75	270	0.7916	0.7558	0.7916	0.8897
No log	34.0	272	0.7453	0.7614	0.7453	0.8633
No log	34.25	274	0.6839	0.7531	0.6839	0.8270
No log	34.5	276	0.6484	0.7595	0.6484	0.8052
No log	34.75	278	0.6428	0.7792	0.6428	0.8017
No log	35.0	280	0.6590	0.7792	0.6590	0.8118
No log	35.25	282	0.6854	0.75	0.6854	0.8279
No log	35.5	284	0.7148	0.7308	0.7148	0.8454
No log	35.75	286	0.7181	0.72	0.7181	0.8474
No log	36.0	288	0.7316	0.7413	0.7316	0.8553
No log	36.25	290	0.7589	0.7194	0.7589	0.8712
No log	36.5	292	0.7878	0.6667	0.7878	0.8876
No log	36.75	294	0.8333	0.6383	0.8333	0.9129
No log	37.0	296	0.8739	0.6667	0.8739	0.9348
No log	37.25	298	0.8382	0.6622	0.8382	0.9155
No log	37.5	300	0.8292	0.6667	0.8292	0.9106
No log	37.75	302	0.8757	0.6797	0.8757	0.9358
No log	38.0	304	0.8931	0.7152	0.8931	0.9450
No log	38.25	306	0.8570	0.6962	0.8570	0.9257
No log	38.5	308	0.8004	0.7067	0.8004	0.8947
No log	38.75	310	0.7791	0.7347	0.7791	0.8826
No log	39.0	312	0.7691	0.7234	0.7691	0.8770
No log	39.25	314	0.7870	0.7	0.7870	0.8871
No log	39.5	316	0.8627	0.6389	0.8627	0.9288
No log	39.75	318	1.0011	0.6463	1.0011	1.0005
No log	40.0	320	1.0803	0.6279	1.0803	1.0394
No log	40.25	322	1.0541	0.6429	1.0541	1.0267
No log	40.5	324	0.9738	0.6883	0.9738	0.9868
No log	40.75	326	0.9081	0.6573	0.9081	0.9529
No log	41.0	328	0.8634	0.6571	0.8634	0.9292
No log	41.25	330	0.8323	0.6667	0.8323	0.9123
No log	41.5	332	0.8529	0.6667	0.8529	0.9235
No log	41.75	334	0.9344	0.6710	0.9344	0.9667
No log	42.0	336	0.9701	0.6744	0.9701	0.9849
No log	42.25	338	0.9330	0.6936	0.9330	0.9659
No log	42.5	340	0.8401	0.6982	0.8401	0.9166
No log	42.75	342	0.7564	0.7152	0.7564	0.8697
No log	43.0	344	0.7163	0.7324	0.7163	0.8463
No log	43.25	346	0.6923	0.7397	0.6923	0.8320
No log	43.5	348	0.6946	0.7397	0.6946	0.8334
No log	43.75	350	0.7274	0.7432	0.7274	0.8529
No log	44.0	352	0.7840	0.6755	0.7840	0.8855
No log	44.25	354	0.8335	0.6835	0.8335	0.9130
No log	44.5	356	0.8199	0.6842	0.8199	0.9055
No log	44.75	358	0.7618	0.6757	0.7618	0.8728
No log	45.0	360	0.7104	0.7310	0.7104	0.8429
No log	45.25	362	0.7011	0.7310	0.7011	0.8373
No log	45.5	364	0.7156	0.7310	0.7156	0.8459
No log	45.75	366	0.7226	0.7383	0.7226	0.8501
No log	46.0	368	0.7422	0.7260	0.7422	0.8615
No log	46.25	370	0.7725	0.7260	0.7725	0.8789
No log	46.5	372	0.7951	0.7183	0.7951	0.8917
No log	46.75	374	0.8282	0.6434	0.8282	0.9101
No log	47.0	376	0.8098	0.7183	0.8098	0.8999
No log	47.25	378	0.8133	0.7183	0.8133	0.9019
No log	47.5	380	0.8545	0.625	0.8545	0.9244
No log	47.75	382	0.8564	0.6483	0.8564	0.9254
No log	48.0	384	0.8308	0.6714	0.8308	0.9115
No log	48.25	386	0.8471	0.6713	0.8471	0.9204
No log	48.5	388	0.8963	0.6755	0.8963	0.9467
No log	48.75	390	0.8905	0.6974	0.8905	0.9437
No log	49.0	392	0.8342	0.6763	0.8342	0.9134
No log	49.25	394	0.7968	0.6809	0.7968	0.8926
No log	49.5	396	0.7416	0.6857	0.7416	0.8612
No log	49.75	398	0.7071	0.6857	0.7071	0.8409
No log	50.0	400	0.7058	0.7333	0.7058	0.8401
No log	50.25	402	0.7424	0.7590	0.7424	0.8617
No log	50.5	404	0.8203	0.7251	0.8203	0.9057
No log	50.75	406	0.8369	0.7159	0.8369	0.9148
No log	51.0	408	0.7750	0.7296	0.7750	0.8803
No log	51.25	410	0.6908	0.7083	0.6908	0.8311
No log	51.5	412	0.6735	0.7123	0.6735	0.8207
No log	51.75	414	0.6765	0.7310	0.6765	0.8225
No log	52.0	416	0.6957	0.7123	0.6957	0.8341
No log	52.25	418	0.7181	0.7034	0.7181	0.8474
No log	52.5	420	0.7545	0.6809	0.7545	0.8686
No log	52.75	422	0.7685	0.6809	0.7685	0.8766
No log	53.0	424	0.7685	0.6809	0.7685	0.8767
No log	53.25	426	0.7834	0.6809	0.7834	0.8851
No log	53.5	428	0.7866	0.6809	0.7866	0.8869
No log	53.75	430	0.7943	0.6809	0.7943	0.8913
No log	54.0	432	0.7823	0.6809	0.7823	0.8845
No log	54.25	434	0.7731	0.6809	0.7731	0.8793
No log	54.5	436	0.7726	0.6809	0.7726	0.8790
No log	54.75	438	0.7837	0.6906	0.7837	0.8853
No log	55.0	440	0.7969	0.6906	0.7969	0.8927
No log	55.25	442	0.7869	0.6906	0.7869	0.8870
No log	55.5	444	0.7651	0.7075	0.7651	0.8747
No log	55.75	446	0.7658	0.7075	0.7658	0.8751
No log	56.0	448	0.7854	0.7179	0.7854	0.8862
No log	56.25	450	0.7961	0.75	0.7961	0.8922
No log	56.5	452	0.7878	0.7179	0.7878	0.8876
No log	56.75	454	0.7823	0.7179	0.7823	0.8845
No log	57.0	456	0.7646	0.7333	0.7646	0.8744
No log	57.25	458	0.7533	0.7333	0.7533	0.8679
No log	57.5	460	0.7424	0.7123	0.7424	0.8616
No log	57.75	462	0.7439	0.7075	0.7439	0.8625
No log	58.0	464	0.7773	0.7260	0.7773	0.8816
No log	58.25	466	0.8506	0.6939	0.8506	0.9223
No log	58.5	468	0.8940	0.6667	0.8940	0.9455
No log	58.75	470	0.8821	0.6803	0.8821	0.9392
No log	59.0	472	0.8346	0.6713	0.8346	0.9136
No log	59.25	474	0.7894	0.6901	0.7894	0.8885
No log	59.5	476	0.7870	0.6901	0.7870	0.8871
No log	59.75	478	0.7662	0.7297	0.7662	0.8753
No log	60.0	480	0.7498	0.7432	0.7498	0.8659
No log	60.25	482	0.7496	0.7432	0.7496	0.8658
No log	60.5	484	0.7560	0.7226	0.7560	0.8695
No log	60.75	486	0.7568	0.7226	0.7568	0.8699
No log	61.0	488	0.8036	0.725	0.8036	0.8965
No log	61.25	490	0.8549	0.7251	0.8549	0.9246
No log	61.5	492	0.8585	0.7251	0.8585	0.9266
No log	61.75	494	0.8543	0.7368	0.8543	0.9243
No log	62.0	496	0.8234	0.7425	0.8234	0.9074
No log	62.25	498	0.7927	0.7215	0.7927	0.8903
0.2908	62.5	500	0.7788	0.7105	0.7788	0.8825
0.2908	62.75	502	0.7619	0.7432	0.7619	0.8729
0.2908	63.0	504	0.7531	0.7413	0.7531	0.8678
0.2908	63.25	506	0.7644	0.7222	0.7644	0.8743
0.2908	63.5	508	0.7947	0.7172	0.7947	0.8915
0.2908	63.75	510	0.8359	0.6944	0.8359	0.9143
0.2908	64.0	512	0.8878	0.6667	0.8878	0.9422
0.2908	64.25	514	0.9180	0.6438	0.9180	0.9581
0.2908	64.5	516	0.8971	0.6667	0.8971	0.9471
0.2908	64.75	518	0.8721	0.6901	0.8721	0.9339

Framework versions

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