kapuska/t5-small-finetuned-on-800-records-samsum

This model is a fine-tuned version of t5-small on an unknown dataset. It achieves the following results on the evaluation set:

• Train Loss: 0.7883
• Validation Loss: 2.3752
• Train Rouge1: 24.8093
• Train Rouge2: 8.8889
• Train Rougel: 22.6817
• Train Rougelsum: 22.6817
• Train Gen Len: 19.0
• Epoch: 99

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:

• optimizer: {'name': 'AdamWeightDecay', 'learning_rate': 2e-05, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False, 'weight_decay_rate': 0.01}
• training_precision: float32

Training results

Train Loss	Validation Loss	Train Rouge1	Train Rouge2	Train Rougel	Train Rougelsum	Train Gen Len	Epoch
1.9252	1.9205	19.5556	2.3256	15.1111	15.1111	19.0	0
1.9005	1.9227	17.5579	2.3810	15.2852	15.2852	19.0	1
1.8769	1.9228	17.5579	2.3810	15.2852	15.2852	19.0	2
1.8463	1.9192	17.5579	2.3810	15.2852	15.2852	19.0	3
1.8251	1.9132	17.4786	2.3256	13.0342	13.0342	19.0	4
1.8148	1.9147	15.5594	2.3810	13.2867	13.2867	19.0	5
1.7980	1.9142	15.5594	2.3810	13.2867	13.2867	19.0	6
1.7684	1.9158	15.6772	2.3810	13.4045	13.4045	19.0	7
1.7571	1.9161	17.5964	2.3256	13.1519	13.1519	19.0	8
1.7345	1.9221	19.6372	2.3256	15.1927	15.1927	19.0	9
1.7136	1.9141	19.6372	2.3256	15.1927	15.1927	19.0	10
1.6935	1.9249	19.6372	2.3256	15.1927	15.1927	19.0	11
1.6685	1.9226	19.6372	2.3256	15.1927	15.1927	19.0	12
1.6571	1.9258	19.6372	2.3256	15.1927	15.1927	19.0	13
1.6327	1.9308	19.6372	2.3256	15.1927	15.1927	19.0	14
1.6295	1.9271	19.6372	2.3256	15.1927	15.1927	19.0	15
1.6112	1.9314	19.5556	2.3256	15.1111	15.1111	19.0	16
1.6008	1.9357	19.6372	2.3256	15.1927	15.1927	19.0	17
1.5826	1.9277	19.3913	2.2727	15.0435	15.0435	19.0	18
1.5784	1.9342	21.3913	2.2727	17.0435	17.0435	19.0	19
1.5553	1.9364	19.3913	2.2727	15.0435	15.0435	19.0	20
1.5292	1.9461	19.3913	2.2727	15.0435	15.0435	19.0	21
1.5114	1.9505	19.3913	2.2727	15.0435	15.0435	19.0	22
1.5042	1.9540	17.5964	2.3256	13.1519	13.1519	19.0	23
1.4964	1.9494	19.0621	4.4444	16.9344	16.9344	19.0	24
1.4736	1.9569	24.7136	4.4444	20.6628	22.5859	19.0	25
1.4644	1.9618	24.7136	4.4444	20.6628	22.5859	19.0	26
1.4562	1.9693	18.9821	4.4444	16.8544	16.8544	19.0	27
1.4339	1.9597	22.7905	4.4444	18.7398	20.6628	19.0	28
1.4204	1.9702	18.9444	4.4444	16.8167	16.8167	19.0	29
1.4182	1.9715	18.9444	4.4444	16.8167	16.8167	19.0	30
1.4014	1.9768	18.9444	4.4444	16.8167	16.8167	19.0	31
1.3845	1.9847	20.9428	4.4444	18.8152	18.8152	19.0	32
1.3756	1.9790	18.9444	4.4444	16.8167	16.8167	19.0	33
1.3611	1.9936	18.9444	4.4444	16.8167	16.8167	19.0	34
1.3495	1.9900	18.9444	4.4444	16.8167	16.8167	19.0	35
1.3403	1.9998	20.9428	4.4444	18.8152	18.8152	19.0	36
1.3253	2.0060	18.9444	4.4444	16.8167	16.8167	19.0	37
1.3109	2.0088	18.9821	4.4444	16.8544	16.8544	19.0	38
1.3106	2.0121	20.8674	4.4444	18.7398	18.7398	19.0	39
1.2903	2.0142	18.9444	4.4444	16.8167	16.8167	19.0	40
1.2795	2.0239	20.8674	4.4444	18.7398	18.7398	19.0	41
1.2788	2.0322	18.9444	4.4444	16.8167	16.8167	19.0	42
1.2629	2.0284	18.9444	4.4444	16.8167	16.8167	19.0	43
1.2525	2.0423	18.9444	4.4444	16.8167	16.8167	19.0	44
1.2373	2.0424	27.0458	11.1111	22.9951	24.9182	19.0	45
1.2242	2.0454	18.9444	4.4444	16.8167	16.8167	19.0	46
1.2214	2.0541	18.9444	4.4444	16.8167	16.8167	19.0	47
1.2066	2.0567	27.0458	11.1111	22.9951	24.9182	19.0	48
1.1866	2.0632	26.9370	11.1111	24.8093	24.8093	19.0	49
1.1976	2.0684	27.0458	11.1111	22.9951	24.9182	19.0	50
1.1806	2.0725	27.0458	11.1111	22.9951	24.9182	19.0	51
1.1662	2.0803	27.0458	11.1111	22.9951	24.9182	19.0	52
1.1626	2.0840	23.1997	11.1111	21.0720	21.0720	19.0	53
1.1464	2.0855	23.1997	11.1111	21.0720	21.0720	19.0	54
1.1298	2.0956	18.9444	4.4444	16.8167	16.8167	19.0	55
1.1300	2.1050	23.1997	11.1111	21.0720	21.0720	19.0	56
1.1255	2.1025	18.9444	4.4444	16.8167	16.8167	19.0	57
1.1005	2.1188	18.9444	4.4444	16.8167	16.8167	19.0	58
1.1002	2.1261	23.1997	11.1111	21.0720	21.0720	19.0	59
1.0806	2.1318	22.6817	4.4444	20.5540	20.5540	19.0	60
1.0869	2.1425	23.1997	11.1111	21.0720	21.0720	19.0	61
1.0768	2.1492	18.9444	4.4444	16.8167	16.8167	19.0	62
1.0681	2.1473	18.9444	4.4444	16.8167	16.8167	19.0	63
1.0594	2.1440	18.9444	4.4444	16.8167	16.8167	19.0	64
1.0411	2.1461	22.6817	4.4444	20.5540	20.5540	19.0	65
1.0342	2.1727	22.6817	4.4444	20.5540	20.5540	19.0	66
1.0306	2.1677	22.6817	4.4444	20.5540	20.5540	19.0	67
1.0163	2.1753	22.6817	4.4444	20.5540	20.5540	19.0	68
1.0139	2.1767	22.6817	4.4444	20.5540	20.5540	19.0	69
1.0036	2.1929	18.9444	4.4444	16.8167	16.8167	19.0	70
1.0049	2.1902	23.1997	11.1111	21.0720	21.0720	19.0	71
0.9947	2.1936	18.9444	4.4444	16.8167	16.8167	19.0	72
0.9803	2.2084	18.9444	4.4444	16.8167	16.8167	19.0	73
0.9791	2.2106	19.3144	4.5455	17.1405	17.1405	19.0	74
0.9655	2.2172	20.8674	4.4444	18.7398	18.7398	19.0	75
0.9640	2.2215	22.6817	4.4444	20.5540	20.5540	19.0	76
0.9456	2.2341	26.9370	11.1111	24.8093	24.8093	19.0	77
0.9396	2.2414	23.0705	8.8889	20.9428	20.9428	19.0	78
0.9335	2.2455	18.9444	4.4444	16.8167	16.8167	19.0	79
0.9261	2.2560	23.1997	11.1111	21.0720	21.0720	19.0	80
0.9075	2.2642	23.1997	11.1111	21.0720	21.0720	19.0	81
0.9023	2.2763	22.9951	8.8889	20.8674	20.8674	19.0	82
0.9044	2.2782	21.0720	8.8889	18.9444	18.9444	19.0	83
0.8961	2.2812	24.8093	8.8889	22.6817	22.6817	19.0	84
0.8813	2.2794	24.8093	8.8889	22.6817	22.6817	19.0	85
0.8731	2.2886	21.0720	8.8889	18.9444	18.9444	19.0	86
0.8751	2.2930	24.8093	8.8889	22.6817	22.6817	19.0	87
0.8652	2.3024	25.2256	6.8182	23.0517	23.0517	19.0	88
0.8605	2.3131	24.8093	8.8889	22.6817	22.6817	19.0	89
0.8571	2.3070	22.9951	8.8889	20.8674	20.8674	19.0	90
0.8473	2.3123	25.1227	11.1111	22.9951	22.9951	19.0	91
0.8456	2.3272	25.1227	11.1111	22.9951	22.9951	19.0	92
0.8329	2.3427	26.9370	11.1111	24.8093	24.8093	19.0	93
0.8294	2.3419	25.1982	11.1111	23.0705	23.0705	19.0	94
0.8243	2.3507	25.1982	11.1111	23.0705	23.0705	19.0	95
0.8132	2.3600	24.8093	8.8889	22.6817	22.6817	19.0	96
0.8153	2.3501	24.8093	8.8889	22.6817	22.6817	19.0	97
0.8005	2.3579	20.8778	2.2727	18.7039	18.7039	19.0	98
0.7883	2.3752	24.8093	8.8889	22.6817	22.6817	19.0	99

Framework versions

• Transformers 4.20.1
• TensorFlow 2.8.2
• Datasets 2.3.2
• Tokenizers 0.12.1