filipinolingo_model

This model is a fine-tuned version of youdiniplays/filipinolingo_model on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 3.6597
• Bleu: 11.8044
• Gen Len: 14.75

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: 0.001
• 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: 300
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Gen Len
No log	1.0	4	2.6992	3.5276	13.75
No log	2.0	8	2.3483	6.8767	14.1875
No log	3.0	12	2.2289	8.4749	14.5625
No log	4.0	16	2.2552	8.537	14.375
No log	5.0	20	2.3404	9.3451	13.875
No log	6.0	24	2.5126	15.043	14.0625
No log	7.0	28	2.7072	14.9624	14.125
No log	8.0	32	2.8633	14.8092	14.3125
No log	9.0	36	2.9499	15.0385	14.125
No log	10.0	40	2.9954	9.0751	14.1875
No log	11.0	44	3.0306	8.321	14.125
No log	12.0	48	3.0640	8.5338	14.0625
No log	13.0	52	3.0869	8.5302	14.0625
No log	14.0	56	3.1138	8.3647	14.125
No log	15.0	60	3.1254	8.5765	13.9375
No log	16.0	64	3.1203	8.5302	14.0625
No log	17.0	68	3.1250	12.0182	14.1875
No log	18.0	72	3.1192	12.0182	14.1875
No log	19.0	76	3.1231	8.5338	14.1875
No log	20.0	80	3.1155	11.9388	13.875
No log	21.0	84	3.1176	11.9402	13.875
No log	22.0	88	3.1295	11.9402	13.875
No log	23.0	92	3.1487	11.9402	13.875
No log	24.0	96	3.1673	12.1489	13.875
No log	25.0	100	3.1859	16.2159	13.875
No log	26.0	104	3.2061	15.9711	13.8125
No log	27.0	108	3.2147	15.9711	13.8125
No log	28.0	112	3.2223	15.9711	13.8125
No log	29.0	116	3.2345	16.2159	13.8125
No log	30.0	120	3.2414	16.1289	13.8125
No log	31.0	124	3.2448	16.1261	13.8125
No log	32.0	128	3.2446	16.1261	13.8125
No log	33.0	132	3.2307	15.8836	13.75
No log	34.0	136	3.2247	15.8803	13.75
No log	35.0	140	3.2364	15.8803	13.75
No log	36.0	144	3.2507	16.1261	13.8125
No log	37.0	148	3.2608	16.1261	13.8125
No log	38.0	152	3.2893	16.536	13.8125
No log	39.0	156	3.3026	16.3582	13.8125
No log	40.0	160	3.2786	16.3582	13.9375
No log	41.0	164	3.2607	16.3548	14.0
No log	42.0	168	3.2557	16.4428	14.0
No log	43.0	172	3.2648	16.1734	14.1875
No log	44.0	176	3.2455	12.2013	14.375
No log	45.0	180	3.2444	12.2013	14.375
No log	46.0	184	3.2416	12.2013	14.375
No log	47.0	188	3.2412	11.8127	14.375
No log	48.0	192	3.2656	16.2611	14.3125
No log	49.0	196	3.2998	16.0785	15.1875
No log	50.0	200	3.3196	16.0785	14.6875
No log	51.0	204	3.3304	15.9095	15.0
No log	52.0	208	3.3312	16.0125	15.0
No log	53.0	212	3.3265	16.0956	14.5
No log	54.0	216	3.3282	16.2714	14.0625
No log	55.0	220	3.3316	16.2714	14.0625
No log	56.0	224	3.3312	16.2714	14.0625
No log	57.0	228	3.3262	15.8593	14.5
No log	58.0	232	3.3327	15.8672	14.5
No log	59.0	236	3.3157	15.6948	14.9375
No log	60.0	240	3.2849	15.8593	15.0
No log	61.0	244	3.2707	15.8593	15.0
No log	62.0	248	3.2732	15.8593	15.0625
No log	63.0	252	3.2781	18.4173	15.1875
No log	64.0	256	3.2990	18.6185	15.0
No log	65.0	260	3.3277	18.6185	14.9375
No log	66.0	264	3.3475	15.1975	14.8125
No log	67.0	268	3.3274	15.2762	14.6875
No log	68.0	272	3.3065	15.5165	14.75
No log	69.0	276	3.3111	18.6185	14.625
No log	70.0	280	3.3575	18.2583	14.6875
No log	71.0	284	3.4089	18.5319	14.875
No log	72.0	288	3.3937	18.6269	14.8125
No log	73.0	292	3.3043	18.6269	14.8125
No log	74.0	296	3.2596	18.7252	14.8125
No log	75.0	300	3.2515	12.9228	15.125
No log	76.0	304	3.2995	13.0338	15.125
No log	77.0	308	3.3457	12.7784	15.25
No log	78.0	312	3.3949	12.5078	15.375
No log	79.0	316	3.4148	12.5862	14.625
No log	80.0	320	3.4307	12.3785	14.75
No log	81.0	324	3.4095	11.6247	14.5
No log	82.0	328	3.3948	11.6247	14.5625
No log	83.0	332	3.3857	11.6247	14.4375
No log	84.0	336	3.3724	11.4452	13.875
No log	85.0	340	3.3688	11.4377	13.8125
No log	86.0	344	3.3656	11.4377	13.8125
No log	87.0	348	3.3839	11.4295	13.8125
No log	88.0	352	3.4168	11.1357	13.8125
No log	89.0	356	3.4694	11.1357	13.8125
No log	90.0	360	3.4992	10.5869	13.8125
No log	91.0	364	3.5087	10.5869	13.8125
No log	92.0	368	3.4923	11.0784	14.125
No log	93.0	372	3.4931	14.544	14.5
No log	94.0	376	3.5046	14.544	14.625
No log	95.0	380	3.5058	14.1526	14.375
No log	96.0	384	3.5057	13.9259	14.8125
No log	97.0	388	3.5107	13.9259	14.75
No log	98.0	392	3.5173	11.0784	14.25
No log	99.0	396	3.5231	11.0887	14.3125
No log	100.0	400	3.5289	11.2541	13.75
No log	101.0	404	3.5357	11.2541	13.75
No log	102.0	408	3.5417	11.1254	14.125
No log	103.0	412	3.5468	11.3608	14.25
No log	104.0	416	3.5430	11.3023	14.625
No log	105.0	420	3.5337	10.9245	14.875
No log	106.0	424	3.5247	10.9783	14.8125
No log	107.0	428	3.5199	10.9783	14.8125
No log	108.0	432	3.5172	10.9783	14.8125
No log	109.0	436	3.5164	11.3128	14.9375
No log	110.0	440	3.5167	11.3128	14.9375
No log	111.0	444	3.5178	11.3128	14.9375
No log	112.0	448	3.5201	11.3128	14.9375
No log	113.0	452	3.5232	11.5924	14.9375
No log	114.0	456	3.5264	11.5924	14.9375
No log	115.0	460	3.5210	11.5924	14.9375
No log	116.0	464	3.5163	11.3128	14.6875
No log	117.0	468	3.5180	11.3706	14.625
No log	118.0	472	3.5237	11.3706	14.625
No log	119.0	476	3.5285	11.6792	14.875
No log	120.0	480	3.5299	11.9509	14.875
No log	121.0	484	3.5301	11.9509	14.875
No log	122.0	488	3.5318	11.9509	14.875
No log	123.0	492	3.5342	11.9509	14.875
No log	124.0	496	3.5355	11.9509	14.875
0.0683	125.0	500	3.5385	11.9509	14.6875
0.0683	126.0	504	3.5422	11.9509	14.6875
0.0683	127.0	508	3.5454	11.9509	14.6875
0.0683	128.0	512	3.5490	11.9509	14.875
0.0683	129.0	516	3.5494	11.9509	14.6875
0.0683	130.0	520	3.5500	11.9509	14.6875
0.0683	131.0	524	3.5513	11.6107	14.6875
0.0683	132.0	528	3.5545	11.8824	14.6875
0.0683	133.0	532	3.5571	11.8202	14.6875
0.0683	134.0	536	3.5597	11.8202	14.875
0.0683	135.0	540	3.5611	11.8824	14.5625
0.0683	136.0	544	3.5629	11.8824	14.5625
0.0683	137.0	548	3.5666	11.8824	14.5625
0.0683	138.0	552	3.5715	11.8824	14.5625
0.0683	139.0	556	3.5762	11.8824	14.5625
0.0683	140.0	560	3.5789	11.8824	14.5625
0.0683	141.0	564	3.5807	11.8824	14.5625
0.0683	142.0	568	3.5858	11.8824	14.5625
0.0683	143.0	572	3.5902	11.8202	14.875
0.0683	144.0	576	3.5886	11.5499	14.875
0.0683	145.0	580	3.5877	11.5499	14.875
0.0683	146.0	584	3.5866	11.6107	14.875
0.0683	147.0	588	3.5875	11.6107	14.875
0.0683	148.0	592	3.5892	11.6107	14.875
0.0683	149.0	596	3.5951	11.6792	14.875
0.0683	150.0	600	3.6008	11.6792	14.875
0.0683	151.0	604	3.6067	11.6792	14.875
0.0683	152.0	608	3.5964	11.6107	14.875
0.0683	153.0	612	3.5930	11.6107	14.875
0.0683	154.0	616	3.5945	11.5499	15.125
0.0683	155.0	620	3.5948	11.5499	15.125
0.0683	156.0	624	3.5953	11.6107	14.875
0.0683	157.0	628	3.5990	11.6107	14.875
0.0683	158.0	632	3.6028	11.6107	14.875
0.0683	159.0	636	3.6059	11.6026	14.875
0.0683	160.0	640	3.6090	11.6026	14.875
0.0683	161.0	644	3.6104	11.6026	14.875
0.0683	162.0	648	3.6114	11.6026	14.875
0.0683	163.0	652	3.6129	11.6026	14.875
0.0683	164.0	656	3.6135	11.6026	14.875
0.0683	165.0	660	3.6145	11.6026	14.875
0.0683	166.0	664	3.6152	11.6026	14.875
0.0683	167.0	668	3.6175	11.6026	14.875
0.0683	168.0	672	3.6140	11.6026	14.875
0.0683	169.0	676	3.6140	11.6026	14.875
0.0683	170.0	680	3.6159	11.3715	14.875
0.0683	171.0	684	3.6162	11.3715	14.875
0.0683	172.0	688	3.6174	11.3715	14.875
0.0683	173.0	692	3.6192	11.3715	14.875
0.0683	174.0	696	3.6209	11.3715	14.875
0.0683	175.0	700	3.6219	11.3715	14.875
0.0683	176.0	704	3.6239	11.3715	14.875
0.0683	177.0	708	3.6266	11.3715	14.875
0.0683	178.0	712	3.6308	11.3715	14.875
0.0683	179.0	716	3.6316	11.3715	14.875
0.0683	180.0	720	3.6321	11.6026	14.875
0.0683	181.0	724	3.6322	11.6026	14.875
0.0683	182.0	728	3.6319	11.8757	14.875
0.0683	183.0	732	3.6319	11.6577	14.875
0.0683	184.0	736	3.6293	11.8757	14.875
0.0683	185.0	740	3.6229	11.8757	14.875
0.0683	186.0	744	3.6186	11.8757	14.875
0.0683	187.0	748	3.6166	11.8757	14.875
0.0683	188.0	752	3.6165	11.8757	14.875
0.0683	189.0	756	3.6193	11.8757	14.875
0.0683	190.0	760	3.6216	11.8757	14.875
0.0683	191.0	764	3.6239	11.8757	14.875
0.0683	192.0	768	3.6265	11.8757	14.875
0.0683	193.0	772	3.6284	11.8757	14.875
0.0683	194.0	776	3.6301	11.8684	14.8125
0.0683	195.0	780	3.6319	11.8684	14.8125
0.0683	196.0	784	3.6341	11.8684	14.8125
0.0683	197.0	788	3.6364	11.8684	14.8125
0.0683	198.0	792	3.6386	11.8684	14.8125
0.0683	199.0	796	3.6418	11.8757	14.8125
0.0683	200.0	800	3.6447	11.8757	14.8125
0.0683	201.0	804	3.6463	12.1401	14.8125
0.0683	202.0	808	3.6476	12.1401	14.8125
0.0683	203.0	812	3.6496	11.9402	14.5625
0.0683	204.0	816	3.6518	12.0061	14.1875
0.0683	205.0	820	3.6544	12.0061	14.1875
0.0683	206.0	824	3.6561	12.0061	14.1875
0.0683	207.0	828	3.6574	12.206	14.3125
0.0683	208.0	832	3.6588	12.1401	14.6875
0.0683	209.0	836	3.6603	12.1401	14.6875
0.0683	210.0	840	3.6612	12.1401	14.6875
0.0683	211.0	844	3.6620	12.1401	14.6875
0.0683	212.0	848	3.6628	12.1401	14.6875
0.0683	213.0	852	3.6628	12.1401	14.6875
0.0683	214.0	856	3.6633	11.8757	14.6875
0.0683	215.0	860	3.6648	11.8757	14.6875
0.0683	216.0	864	3.6665	11.8757	14.6875
0.0683	217.0	868	3.6678	11.8044	14.75
0.0683	218.0	872	3.6690	11.8044	14.75
0.0683	219.0	876	3.6699	11.8044	14.75
0.0683	220.0	880	3.6693	11.8044	14.75
0.0683	221.0	884	3.6689	11.8757	14.6875
0.0683	222.0	888	3.6687	11.8757	14.8125
0.0683	223.0	892	3.6687	11.8757	14.8125
0.0683	224.0	896	3.6690	11.8757	14.8125
0.0683	225.0	900	3.6662	11.8757	14.8125
0.0683	226.0	904	3.6609	11.8757	14.8125
0.0683	227.0	908	3.6561	11.8757	14.8125
0.0683	228.0	912	3.6536	11.8757	14.8125
0.0683	229.0	916	3.6522	11.8757	14.8125
0.0683	230.0	920	3.6515	11.8757	14.8125
0.0683	231.0	924	3.6526	11.8757	14.8125
0.0683	232.0	928	3.6532	11.8757	14.8125
0.0683	233.0	932	3.6537	11.8757	14.8125
0.0683	234.0	936	3.6536	11.8757	14.8125
0.0683	235.0	940	3.6540	11.8757	14.8125
0.0683	236.0	944	3.6540	11.8757	14.8125
0.0683	237.0	948	3.6540	11.8757	14.8125
0.0683	238.0	952	3.6545	11.8757	14.8125
0.0683	239.0	956	3.6553	11.8757	14.8125
0.0683	240.0	960	3.6557	11.8757	14.8125
0.0683	241.0	964	3.6563	11.8757	14.8125
0.0683	242.0	968	3.6573	11.8757	14.8125
0.0683	243.0	972	3.6579	11.8757	14.8125
0.0683	244.0	976	3.6583	11.8757	14.8125
0.0683	245.0	980	3.6594	11.8757	14.8125
0.0683	246.0	984	3.6599	11.8757	14.8125
0.0683	247.0	988	3.6606	11.8757	14.8125
0.0683	248.0	992	3.6513	11.8757	14.8125
0.0683	249.0	996	3.6454	11.8757	14.8125
0.0005	250.0	1000	3.6429	11.8757	14.8125
0.0005	251.0	1004	3.6415	11.8757	14.8125
0.0005	252.0	1008	3.6403	11.8757	14.8125
0.0005	253.0	1012	3.6400	11.8757	14.8125
0.0005	254.0	1016	3.6410	11.8757	14.8125
0.0005	255.0	1020	3.6418	11.8757	14.8125
0.0005	256.0	1024	3.6430	11.8044	14.75
0.0005	257.0	1028	3.6441	11.8044	14.75
0.0005	258.0	1032	3.6455	11.8044	14.75
0.0005	259.0	1036	3.6463	11.8044	14.75
0.0005	260.0	1040	3.6471	11.8044	14.75
0.0005	261.0	1044	3.6478	11.8044	14.75
0.0005	262.0	1048	3.6487	11.8044	14.75
0.0005	263.0	1052	3.6499	11.8044	14.75
0.0005	264.0	1056	3.6509	11.8044	14.75
0.0005	265.0	1060	3.6516	11.8044	14.75
0.0005	266.0	1064	3.6518	11.8044	14.75
0.0005	267.0	1068	3.6522	11.8044	14.75
0.0005	268.0	1072	3.6524	11.8044	14.75
0.0005	269.0	1076	3.6533	11.8044	14.75
0.0005	270.0	1080	3.6535	11.8044	14.75
0.0005	271.0	1084	3.6543	11.8044	14.75
0.0005	272.0	1088	3.6551	11.8044	14.75
0.0005	273.0	1092	3.6554	11.8044	14.75
0.0005	274.0	1096	3.6559	11.8044	14.75
0.0005	275.0	1100	3.6558	11.8044	14.75
0.0005	276.0	1104	3.6563	11.8044	14.75
0.0005	277.0	1108	3.6567	11.8044	14.75
0.0005	278.0	1112	3.6568	11.8044	14.75
0.0005	279.0	1116	3.6570	11.8044	14.75
0.0005	280.0	1120	3.6573	11.8044	14.75
0.0005	281.0	1124	3.6575	11.8044	14.75
0.0005	282.0	1128	3.6575	11.8044	14.75
0.0005	283.0	1132	3.6574	11.8044	14.75
0.0005	284.0	1136	3.6574	11.8044	14.75
0.0005	285.0	1140	3.6580	11.8044	14.75
0.0005	286.0	1144	3.6579	11.8044	14.75
0.0005	287.0	1148	3.6583	11.8044	14.75
0.0005	288.0	1152	3.6583	11.8044	14.75
0.0005	289.0	1156	3.6589	11.8044	14.75
0.0005	290.0	1160	3.6588	11.8044	14.75
0.0005	291.0	1164	3.6587	11.8044	14.75
0.0005	292.0	1168	3.6588	11.8044	14.75
0.0005	293.0	1172	3.6592	11.8044	14.75
0.0005	294.0	1176	3.6590	11.8044	14.75
0.0005	295.0	1180	3.6592	11.8044	14.75
0.0005	296.0	1184	3.6593	11.8044	14.75
0.0005	297.0	1188	3.6593	11.8044	14.75
0.0005	298.0	1192	3.6598	11.8044	14.75
0.0005	299.0	1196	3.6597	11.8044	14.75
0.0005	300.0	1200	3.6597	11.8044	14.75

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu121
• Datasets 2.16.1
• Tokenizers 0.15.0