dimitarpg13/roberta-finetuned-wines-resampled-val-ds

roberta-finetuned-wines

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

• Loss: 1.4708
• Accuracy: 0.7826
• F1: 0.6998
• Precision: 0.8330
• Recall: 0.7875

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: 1e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 5
• num_epochs: 200
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
7.8604	1.0	405	7.8021	0.0046	0.0001	0.9916	0.0025
7.7552	2.0	810	7.6534	0.0084	0.0005	0.9847	0.0053
7.6208	3.0	1215	7.5104	0.0096	0.0004	0.9728	0.0075
7.4786	4.0	1620	7.3656	0.0192	0.0027	0.9642	0.0137
7.3399	5.0	2025	7.2197	0.0235	0.0032	0.9571	0.0175
7.2026	6.0	2430	7.0715	0.0260	0.0044	0.9587	0.0186
7.0616	7.0	2835	6.9360	0.0322	0.0082	0.9483	0.0232
6.9247	8.0	3240	6.8037	0.0328	0.0089	0.9517	0.0231
6.7897	9.0	3645	6.6682	0.0455	0.0129	0.9493	0.0311
6.6593	10.0	4050	6.5377	0.0473	0.0137	0.9381	0.0343
6.5321	11.0	4455	6.4207	0.0517	0.0153	0.9425	0.0340
6.4102	12.0	4860	6.2955	0.0569	0.0205	0.9339	0.0414
6.291	13.0	5265	6.1805	0.0656	0.0270	0.9252	0.0498
6.1789	14.0	5670	6.0748	0.0755	0.0311	0.9258	0.0577
6.0653	15.0	6075	5.9658	0.0795	0.0313	0.9216	0.0597
5.9595	16.0	6480	5.8629	0.0885	0.0388	0.9137	0.0697
5.8568	17.0	6885	5.7689	0.0928	0.0396	0.9113	0.0706
5.7597	18.0	7290	5.6795	0.0965	0.0410	0.9092	0.0731
5.6659	19.0	7695	5.5862	0.1049	0.0461	0.9029	0.0802
5.5714	20.0	8100	5.4965	0.1135	0.0506	0.8985	0.0877
5.484	21.0	8505	5.4127	0.1166	0.0503	0.8954	0.0901
5.3961	22.0	8910	5.3354	0.1250	0.0579	0.8887	0.0987
5.3137	23.0	9315	5.2551	0.1259	0.0571	0.8888	0.0997
5.2296	24.0	9720	5.1792	0.1349	0.0645	0.8856	0.1067
5.1527	25.0	10125	5.1053	0.1457	0.0742	0.8755	0.1203
5.0725	26.0	10530	5.0318	0.1519	0.0750	0.8735	0.1246
4.9995	27.0	10935	4.9612	0.1581	0.0829	0.8711	0.1322
4.9284	28.0	11340	4.8977	0.1661	0.0857	0.8721	0.1347
4.8567	29.0	11745	4.8218	0.1735	0.0937	0.8683	0.1418
4.783	30.0	12150	4.7584	0.1819	0.0971	0.8637	0.1510
4.7193	31.0	12555	4.6947	0.1896	0.1045	0.8617	0.1586
4.6478	32.0	12960	4.6337	0.2020	0.1148	0.8592	0.1727
4.5815	33.0	13365	4.5725	0.2051	0.1148	0.8541	0.1743
4.523	34.0	13770	4.5180	0.2106	0.1214	0.8551	0.1838
4.4537	35.0	14175	4.4538	0.2215	0.1263	0.8512	0.1897
4.3924	36.0	14580	4.4020	0.2230	0.1297	0.8494	0.1935
4.3313	37.0	14985	4.3394	0.2363	0.1408	0.8389	0.2107
4.2718	38.0	15390	4.2840	0.2484	0.1537	0.8427	0.2200
4.2099	39.0	15795	4.2260	0.2614	0.1625	0.8352	0.2358
4.1507	40.0	16200	4.1730	0.2645	0.1676	0.8324	0.2399
4.0893	41.0	16605	4.1179	0.2827	0.1817	0.8301	0.2552
4.0296	42.0	17010	4.0695	0.2836	0.1852	0.8292	0.2610
3.9781	43.0	17415	4.0107	0.3053	0.2016	0.8286	0.2768
3.9233	44.0	17820	3.9595	0.3071	0.2072	0.8179	0.2846
3.8604	45.0	18225	3.9078	0.3195	0.2186	0.8172	0.2980
3.8038	46.0	18630	3.8599	0.3341	0.2277	0.8140	0.3101
3.7492	47.0	19035	3.8101	0.3387	0.2347	0.8154	0.3158
3.6949	48.0	19440	3.7631	0.3542	0.2510	0.8153	0.3319
3.6347	49.0	19845	3.7169	0.3610	0.2535	0.8138	0.3365
3.5873	50.0	20250	3.6666	0.3721	0.2624	0.8083	0.3478
3.5366	51.0	20655	3.6167	0.3848	0.2773	0.8078	0.3634
3.4761	52.0	21060	3.5746	0.3885	0.2799	0.8113	0.3638
3.4254	53.0	21465	3.5332	0.3934	0.2839	0.8057	0.3692
3.3765	54.0	21870	3.4823	0.4105	0.3037	0.8037	0.3910
3.3245	55.0	22275	3.4392	0.4253	0.3176	0.8054	0.4059
3.2754	56.0	22680	3.3974	0.4337	0.3239	0.8034	0.4124
3.2224	57.0	23085	3.3565	0.4408	0.3293	0.7991	0.4177
3.1768	58.0	23490	3.3120	0.4550	0.3429	0.7966	0.4339
3.1202	59.0	23895	3.2699	0.4596	0.3497	0.7995	0.4376
3.0742	60.0	24300	3.2277	0.4692	0.3574	0.7977	0.4497
3.0225	61.0	24705	3.1866	0.4751	0.3627	0.7967	0.4543
2.9801	62.0	25110	3.1455	0.4872	0.3748	0.7943	0.4683
2.9287	63.0	25515	3.1078	0.5014	0.3916	0.7893	0.4848
2.8847	64.0	25920	3.0645	0.5032	0.3909	0.7918	0.4833
2.8325	65.0	26325	3.0316	0.5097	0.3963	0.7925	0.4908
2.7866	66.0	26730	2.9929	0.5280	0.4163	0.7906	0.5092
2.7484	67.0	27135	2.9514	0.5410	0.4302	0.7948	0.5228
2.6964	68.0	27540	2.9139	0.5444	0.4320	0.7966	0.5251
2.653	69.0	27945	2.8757	0.5583	0.4459	0.7941	0.5417
2.6088	70.0	28350	2.8426	0.5605	0.4464	0.7953	0.5415
2.564	71.0	28755	2.8091	0.5725	0.4575	0.7981	0.5481
2.5236	72.0	29160	2.7726	0.5806	0.4725	0.7971	0.5623
2.4776	73.0	29565	2.7376	0.5849	0.4735	0.7967	0.5652
2.4381	74.0	29970	2.7033	0.5939	0.4833	0.7983	0.5764
2.3904	75.0	30375	2.6753	0.6053	0.4958	0.7992	0.5886
2.3502	76.0	30780	2.6403	0.6121	0.5020	0.8015	0.5939
2.3087	77.0	31185	2.6105	0.6174	0.5064	0.8044	0.6001
2.2701	78.0	31590	2.5760	0.6294	0.5233	0.8102	0.6120
2.2276	79.0	31995	2.5457	0.6397	0.5358	0.8074	0.6253
2.1939	80.0	32400	2.5160	0.6443	0.5394	0.8074	0.6302
2.1468	81.0	32805	2.4892	0.6452	0.5398	0.8070	0.6298
2.1113	82.0	33210	2.4566	0.6508	0.5456	0.8080	0.6381
2.0773	83.0	33615	2.4327	0.6539	0.5498	0.8109	0.6382
2.0369	84.0	34020	2.4050	0.6641	0.5625	0.8124	0.6513
1.9986	85.0	34425	2.3793	0.6666	0.5628	0.8116	0.6521
1.9656	86.0	34830	2.3470	0.6709	0.5689	0.8106	0.6579
1.9329	87.0	35235	2.3257	0.6740	0.5718	0.8107	0.6623
1.8962	88.0	35640	2.2960	0.6823	0.5810	0.8145	0.6697
1.8577	89.0	36045	2.2736	0.6898	0.5873	0.8150	0.6791
1.8192	90.0	36450	2.2484	0.6910	0.5887	0.8149	0.6808
1.7895	91.0	36855	2.2235	0.6966	0.5961	0.8153	0.6873
1.7584	92.0	37260	2.2028	0.6944	0.5986	0.8180	0.6850
1.7229	93.0	37665	2.1772	0.7031	0.6063	0.8165	0.6949
1.6962	94.0	38070	2.1558	0.7065	0.6074	0.8195	0.6969
1.6626	95.0	38475	2.1385	0.7077	0.6104	0.8230	0.7005
1.6304	96.0	38880	2.1168	0.7148	0.6172	0.8210	0.7075
1.597	97.0	39285	2.0962	0.7176	0.6237	0.8235	0.7114
1.5736	98.0	39690	2.0748	0.7210	0.6266	0.8215	0.7160
1.5383	99.0	40095	2.0569	0.7250	0.6321	0.8251	0.7197
1.5096	100.0	40500	2.0302	0.7294	0.6350	0.8228	0.7243
1.4847	101.0	40905	2.0159	0.7315	0.6380	0.8261	0.7244
1.4569	102.0	41310	2.0005	0.7318	0.6419	0.8261	0.7274
1.426	103.0	41715	1.9833	0.7340	0.6448	0.8291	0.7291
1.401	104.0	42120	1.9616	0.7408	0.6512	0.8283	0.7369
1.3718	105.0	42525	1.9425	0.7427	0.6502	0.8281	0.7391
1.3496	106.0	42930	1.9275	0.7442	0.6533	0.8273	0.7411
1.3248	107.0	43335	1.9164	0.7420	0.6533	0.8296	0.7389
1.2936	108.0	43740	1.8988	0.7485	0.6554	0.8257	0.7461
1.2709	109.0	44145	1.8812	0.7488	0.6614	0.8268	0.7472
1.2543	110.0	44550	1.8682	0.7504	0.6617	0.8276	0.7499
1.2273	111.0	44955	1.8524	0.7532	0.6612	0.8276	0.7524
1.2041	112.0	45360	1.8357	0.7535	0.6641	0.8307	0.7512
1.1821	113.0	45765	1.8244	0.7563	0.6678	0.8299	0.7546
1.1605	114.0	46170	1.8137	0.7550	0.6658	0.8269	0.7533
1.138	115.0	46575	1.7954	0.7581	0.6714	0.8317	0.7563
1.1139	116.0	46980	1.7860	0.7590	0.6711	0.8292	0.7586
1.095	117.0	47385	1.7778	0.7609	0.6716	0.8295	0.7593
1.0755	118.0	47790	1.7636	0.7615	0.6726	0.8320	0.7597
1.0568	119.0	48195	1.7544	0.7615	0.6715	0.8304	0.7594
1.0381	120.0	48600	1.7403	0.7624	0.6730	0.8293	0.7606
1.0219	121.0	49005	1.7280	0.7621	0.6725	0.8293	0.7615
1.0027	122.0	49410	1.7173	0.7637	0.6751	0.8309	0.7619
0.982	123.0	49815	1.7084	0.7631	0.6738	0.8286	0.7625
0.9653	124.0	50220	1.7028	0.7637	0.6766	0.8305	0.7643
0.9514	125.0	50625	1.6931	0.7637	0.6747	0.8267	0.7638
0.9339	126.0	51030	1.6833	0.7649	0.6759	0.8266	0.7652
0.9179	127.0	51435	1.6746	0.7655	0.6780	0.8293	0.7652
0.9026	128.0	51840	1.6650	0.7643	0.6767	0.8264	0.7650
0.8864	129.0	52245	1.6575	0.7652	0.6764	0.8286	0.7653
0.8695	130.0	52650	1.6465	0.7674	0.6781	0.8250	0.7671
0.8548	131.0	53055	1.6428	0.7674	0.6799	0.8277	0.7692
0.8407	132.0	53460	1.6348	0.7662	0.6791	0.8269	0.7681
0.8302	133.0	53865	1.6304	0.7665	0.6776	0.8260	0.7678
0.816	134.0	54270	1.6201	0.7696	0.6812	0.8271	0.7718
0.8042	135.0	54675	1.6137	0.7699	0.6841	0.8280	0.7725
0.7889	136.0	55080	1.6066	0.7714	0.6839	0.8275	0.7753
0.7737	137.0	55485	1.6014	0.7736	0.6907	0.8285	0.7768
0.7639	138.0	55890	1.5958	0.7708	0.6879	0.8285	0.7742
0.7531	139.0	56295	1.5896	0.7720	0.6875	0.8302	0.7751
0.7447	140.0	56700	1.5849	0.7761	0.6931	0.8301	0.7794
0.7321	141.0	57105	1.5799	0.7717	0.6870	0.8301	0.7755
0.7172	142.0	57510	1.5744	0.7717	0.6865	0.8288	0.7753
0.7069	143.0	57915	1.5718	0.7736	0.6914	0.8315	0.7776
0.7006	144.0	58320	1.5667	0.7754	0.6889	0.8274	0.7805
0.6878	145.0	58725	1.5621	0.7758	0.6909	0.8285	0.7797
0.6781	146.0	59130	1.5597	0.7770	0.6917	0.8304	0.7807
0.6685	147.0	59535	1.5567	0.7751	0.6914	0.8280	0.7807
0.659	148.0	59940	1.5520	0.7758	0.6908	0.8300	0.7809
0.6563	149.0	60345	1.5482	0.7767	0.6912	0.8304	0.7815
0.6433	150.0	60750	1.5440	0.7745	0.6894	0.8287	0.7803
0.6343	151.0	61155	1.5383	0.7764	0.6907	0.8307	0.7804
0.6251	152.0	61560	1.5365	0.7773	0.6920	0.8310	0.7817
0.6171	153.0	61965	1.5329	0.7776	0.6934	0.8313	0.7823
0.6112	154.0	62370	1.5278	0.7792	0.6929	0.8304	0.7832
0.6017	155.0	62775	1.5239	0.7795	0.6941	0.8308	0.7843
0.5956	156.0	63180	1.5230	0.7795	0.6955	0.8313	0.7852
0.5865	157.0	63585	1.5181	0.7795	0.6968	0.8318	0.7855
0.5826	158.0	63990	1.5172	0.7779	0.6951	0.8314	0.7840
0.5755	159.0	64395	1.5160	0.7779	0.6956	0.8315	0.7828
0.5672	160.0	64800	1.5119	0.7782	0.6927	0.8294	0.7827
0.5625	161.0	65205	1.5083	0.7798	0.6962	0.8314	0.7843
0.5587	162.0	65610	1.5078	0.7792	0.6954	0.8288	0.7860
0.5496	163.0	66015	1.5044	0.7792	0.6952	0.8315	0.7846
0.5456	164.0	66420	1.5009	0.7801	0.6965	0.8313	0.7853
0.5368	165.0	66825	1.5002	0.7801	0.6967	0.8316	0.7845
0.5374	166.0	67230	1.4984	0.7801	0.6965	0.8312	0.7848
0.5277	167.0	67635	1.4951	0.7822	0.7000	0.8342	0.7868
0.5273	168.0	68040	1.4969	0.7813	0.6992	0.8317	0.7860
0.5184	169.0	68445	1.4934	0.7813	0.6977	0.8319	0.7864
0.5143	170.0	68850	1.4894	0.7819	0.6988	0.8324	0.7876
0.5116	171.0	69255	1.4901	0.7829	0.7000	0.8346	0.7879
0.5059	172.0	69660	1.4893	0.7822	0.7003	0.8331	0.7880
0.5043	173.0	70065	1.4877	0.7816	0.6974	0.8312	0.7870
0.4958	174.0	70470	1.4858	0.7807	0.6968	0.8317	0.7865
0.4936	175.0	70875	1.4854	0.7826	0.6991	0.8324	0.7870
0.4923	176.0	71280	1.4841	0.7810	0.6977	0.8322	0.7862
0.4866	177.0	71685	1.4830	0.7807	0.7004	0.8339	0.7861
0.4851	178.0	72090	1.4826	0.7798	0.6974	0.8319	0.7854
0.4795	179.0	72495	1.4816	0.7816	0.6985	0.8317	0.7870
0.4772	180.0	72900	1.4789	0.7798	0.6985	0.8327	0.7855
0.4765	181.0	73305	1.4805	0.7810	0.6975	0.8320	0.7858
0.4734	182.0	73710	1.4777	0.7822	0.6997	0.8337	0.7867
0.4702	183.0	74115	1.4776	0.7822	0.6988	0.8316	0.7879
0.4717	184.0	74520	1.4765	0.7822	0.6986	0.8332	0.7875
0.4658	185.0	74925	1.4764	0.7816	0.6982	0.8329	0.7866
0.4642	186.0	75330	1.4758	0.7826	0.6992	0.8331	0.7877
0.462	187.0	75735	1.4746	0.7829	0.7001	0.8322	0.7888
0.4599	188.0	76140	1.4740	0.7822	0.6986	0.8308	0.7881
0.4586	189.0	76545	1.4744	0.7832	0.7002	0.8324	0.7887
0.4554	190.0	76950	1.4738	0.7826	0.7001	0.8328	0.7882
0.4539	191.0	77355	1.4731	0.7822	0.6988	0.8303	0.7881
0.4525	192.0	77760	1.4724	0.7829	0.7010	0.8328	0.7879
0.4483	193.0	78165	1.4722	0.7829	0.6989	0.8315	0.7888
0.4469	194.0	78570	1.4715	0.7829	0.6995	0.8313	0.7885
0.447	195.0	78975	1.4716	0.7835	0.6997	0.8322	0.7886
0.4484	196.0	79380	1.4712	0.7826	0.6994	0.8324	0.7878
0.4459	197.0	79785	1.4707	0.7832	0.7003	0.8329	0.7882
0.4455	198.0	80190	1.4710	0.7826	0.7001	0.8333	0.7878
0.4457	199.0	80595	1.4709	0.7829	0.7000	0.8326	0.7882
0.4422	200.0	81000	1.4708	0.7826	0.6998	0.8330	0.7875

Framework versions

• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 3.5.0
• Tokenizers 0.21.1