flanT5_large_Fact_U

This model is a fine-tuned version of google/flan-t5-large on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.0731
• Accuracy: 0.7788
• Precision: 0.8159
• Recall: 0.7421
• F1 score: 0.7773

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: 0.0001
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1 score
1.2434	0.0314	200	0.7545	0.6059	0.5805	0.8733	0.6974
1.4323	0.0628	400	1.7199	0.6	0.575	0.8846	0.6970
1.253	0.0941	600	1.3828	0.6059	0.5812	0.8665	0.6957
1.2986	0.1255	800	2.4072	0.48	0.0	0.0	0.0
1.2843	0.1569	1000	1.2474	0.6553	0.7690	0.4819	0.5925
1.2245	0.1883	1200	1.3841	0.5435	0.95	0.1290	0.2271
1.2389	0.2197	1400	0.8993	0.6706	0.6731	0.7127	0.6923
1.132	0.2511	1600	0.6845	0.6271	0.6110	0.7783	0.6846
1.0836	0.2824	1800	1.1694	0.6824	0.7486	0.5860	0.6574
1.2434	0.3138	2000	1.4787	0.6788	0.7632	0.5543	0.6422
1.1196	0.3452	2200	1.5004	0.6694	0.7018	0.6335	0.6659
1.5791	0.3766	2400	1.1289	0.6376	0.8454	0.3710	0.5157
1.3035	0.4080	2600	1.0136	0.6859	0.7119	0.6652	0.6877
1.1401	0.4394	2800	1.2340	0.6753	0.7485	0.5656	0.6443
0.9518	0.4707	3000	1.2197	0.7024	0.7692	0.6109	0.6810
1.1623	0.5021	3200	1.2827	0.6788	0.7046	0.6584	0.6807
1.1316	0.5335	3400	1.5077	0.6659	0.6396	0.8190	0.7183
1.2599	0.5649	3600	0.8272	0.6341	0.8466	0.3620	0.5071
0.9866	0.5963	3800	1.4574	0.6647	0.6605	0.7308	0.6939
1.147	0.6276	4000	1.2933	0.6824	0.7792	0.5430	0.64
1.0307	0.6590	4200	1.1586	0.6482	0.7658	0.4661	0.5795
1.0616	0.6904	4400	1.2668	0.6976	0.7428	0.6403	0.6877
1.0724	0.7218	4600	1.1130	0.6447	0.6955	0.5633	0.6225
0.9499	0.7532	4800	1.1635	0.7188	0.7766	0.6448	0.7046
1.1302	0.7846	5000	1.2608	0.7118	0.7031	0.7715	0.7357
1.1921	0.8159	5200	1.1742	0.7094	0.8056	0.5814	0.6754
0.9532	0.8473	5400	1.1589	0.7071	0.7749	0.6154	0.6860
0.783	0.8787	5600	1.3256	0.7	0.8086	0.5543	0.6577
0.9835	0.9101	5800	1.1383	0.7282	0.7828	0.6606	0.7166
0.9898	0.9415	6000	1.0662	0.7141	0.7409	0.6923	0.7158
0.9768	0.9729	6200	1.1941	0.7059	0.8019	0.5769	0.6711
1.043	1.0042	6400	1.2302	0.6729	0.8628	0.4412	0.5838
0.9531	1.0356	6600	1.1304	0.7106	0.7593	0.6493	0.7
1.0585	1.0670	6800	1.0234	0.7294	0.7944	0.6471	0.7132
0.8862	1.0984	7000	1.1941	0.6953	0.8735	0.4842	0.6230
0.8721	1.1298	7200	0.9352	0.7376	0.792	0.6719	0.7271
0.8678	1.1611	7400	1.0473	0.7388	0.7402	0.7670	0.7533
0.7617	1.1925	7600	1.3020	0.7294	0.7181	0.7896	0.7522
1.0394	1.2239	7800	1.0322	0.7212	0.7904	0.6312	0.7019
0.822	1.2553	8000	1.0980	0.7388	0.7973	0.6674	0.7266
0.8406	1.2867	8200	1.4589	0.7118	0.7031	0.7715	0.7357
0.7059	1.3181	8400	1.0655	0.7306	0.8318	0.6041	0.6999
0.8649	1.3494	8600	0.9708	0.7424	0.8106	0.6584	0.7266
0.7142	1.3808	8800	1.1603	0.7553	0.8214	0.6765	0.7419
0.9057	1.4122	9000	0.9389	0.76	0.8381	0.6674	0.7431
0.9312	1.4436	9200	1.0568	0.7553	0.7721	0.7511	0.7615
0.8459	1.4750	9400	1.1646	0.7459	0.7974	0.6855	0.7372
0.8427	1.5064	9600	1.0133	0.7459	0.8174	0.6584	0.7293
0.7245	1.5377	9800	1.1397	0.7341	0.8885	0.5588	0.6861
0.6386	1.5691	10000	1.1112	0.7294	0.9015	0.5385	0.6742
0.7513	1.6005	10200	0.9403	0.7671	0.805	0.7285	0.7648
0.828	1.6319	10400	0.9412	0.76	0.7820	0.7466	0.7639
0.8393	1.6633	10600	0.9359	0.7553	0.8824	0.6109	0.7219
0.8679	1.6946	10800	0.8979	0.7588	0.8415	0.6606	0.7402
0.6735	1.7260	11000	1.0666	0.7588	0.8786	0.6222	0.7285
0.8702	1.7574	11200	0.9554	0.7576	0.795	0.7195	0.7553
0.7435	1.7888	11400	1.0937	0.7588	0.8143	0.6946	0.7497
0.8796	1.8202	11600	0.9257	0.7824	0.8320	0.7285	0.7768
0.6257	1.8516	11800	0.9606	0.7659	0.8172	0.7081	0.7588
0.8589	1.8829	12000	0.9013	0.7659	0.8481	0.6697	0.7484
0.865	1.9143	12200	1.0734	0.7612	0.7673	0.7760	0.7717
0.8068	1.9457	12400	0.9214	0.76	0.8381	0.6674	0.7431
0.6212	1.9771	12600	1.0116	0.7706	0.8539	0.6742	0.7535
0.7657	2.0085	12800	0.9830	0.7718	0.8605	0.6697	0.7532
0.6631	2.0399	13000	1.0075	0.7776	0.8005	0.7624	0.7810
0.3003	2.0712	13200	1.1456	0.7812	0.8333	0.7240	0.7748
0.5982	2.1026	13400	1.0728	0.7753	0.8438	0.6968	0.7633
0.4828	2.1340	13600	1.0474	0.7753	0.8177	0.7308	0.7718
0.5463	2.1654	13800	1.0521	0.7776	0.8252	0.7262	0.7726
0.5429	2.1968	14000	1.0990	0.7706	0.8365	0.6946	0.7590
0.7112	2.2282	14200	1.1072	0.7729	0.8507	0.6833	0.7578
0.4816	2.2595	14400	1.1528	0.7753	0.8277	0.7172	0.7685
0.7882	2.2909	14600	0.9670	0.7765	0.8214	0.7285	0.7722
0.5265	2.3223	14800	1.0724	0.7765	0.8298	0.7172	0.7694
0.6116	2.3537	15000	1.0316	0.7776	0.8203	0.7330	0.7742
0.575	2.3851	15200	1.1125	0.7741	0.8415	0.6968	0.7624
0.5599	2.4164	15400	1.0327	0.7765	0.8119	0.7421	0.7754
0.5821	2.4478	15600	1.0655	0.7776	0.8078	0.7511	0.7784
0.4777	2.4792	15800	1.1187	0.7835	0.8028	0.7738	0.7880
0.432	2.5106	16000	1.1973	0.7788	0.8256	0.7285	0.7740
0.4385	2.5420	16200	1.2155	0.7729	0.8029	0.7466	0.7737
0.6103	2.5734	16400	1.0527	0.78	0.8212	0.7376	0.7771
0.4618	2.6047	16600	1.1377	0.78	0.8164	0.7443	0.7787
0.471	2.6361	16800	1.1468	0.7788	0.8038	0.7602	0.7814
0.6206	2.6675	17000	1.1048	0.7765	0.8014	0.7579	0.7791
0.5869	2.6989	17200	1.1343	0.7776	0.7895	0.7805	0.7850
0.5647	2.7303	17400	1.0843	0.7859	0.8218	0.7511	0.7849
0.5527	2.7617	17600	1.0834	0.7847	0.8091	0.7670	0.7875
0.8013	2.7930	17800	0.9898	0.7894	0.8124	0.7738	0.7926
0.5232	2.8244	18000	1.0052	0.7859	0.8110	0.7670	0.7884
0.617	2.8558	18200	1.0083	0.7824	0.8157	0.7511	0.7821
0.5093	2.8872	18400	1.0510	0.7835	0.8241	0.7421	0.7810
0.5099	2.9186	18600	1.0758	0.78	0.8133	0.7489	0.7797
0.6239	2.9499	18800	1.0726	0.7812	0.8168	0.7466	0.7801
0.6592	2.9813	19000	1.0731	0.7788	0.8159	0.7421	0.7773

Framework versions

• Transformers 4.44.2
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1