t5-small-finetuned-mydata

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

• Loss: 2.7077
• Rouge1: 41.6567
• Rouge2: 23.7942
• Rougel: 41.0101
• Rougelsum: 41.5048
• Gen Len: 7.6027

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: 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: 50
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	1.0	19	4.9039	20.0474	7.234	18.2098	17.9517	10.9589
No log	2.0	38	4.5878	23.0871	8.221	21.7521	21.6804	11.3425
No log	3.0	57	4.3925	23.4492	8.8479	22.0822	22.1146	12.0548
No log	4.0	76	4.2184	26.0031	9.4235	24.6843	24.6388	12.6438
No log	5.0	95	4.0619	26.7979	9.548	25.7363	25.7928	12.8219
No log	6.0	114	3.9334	26.9541	9.7913	25.9349	25.9444	12.726
No log	7.0	133	3.8185	28.0578	10.9266	26.9035	26.746	12.1507
No log	8.0	152	3.7113	28.296	10.9928	26.6577	26.446	12.0822
No log	9.0	171	3.6335	30.3027	11.4952	28.313	28.2952	11.7397
No log	10.0	190	3.5584	30.8405	11.0987	28.7148	28.8457	11.0822
No log	11.0	209	3.4895	30.2533	10.9185	28.3191	28.4837	11.0685
No log	12.0	228	3.4216	30.3158	11.3392	28.3347	28.5197	10.7534
No log	13.0	247	3.3705	30.8803	12.1903	29.3055	29.4952	10.4521
No log	14.0	266	3.3190	31.0433	12.2378	29.4309	29.6068	9.9315
No log	15.0	285	3.2699	31.8936	12.9061	30.1597	30.6298	9.6849
No log	16.0	304	3.2192	33.4292	13.8997	31.779	32.0884	9.1096
No log	17.0	323	3.1740	33.729	14.1086	32.0316	32.315	9.0411
No log	18.0	342	3.1394	36.7725	17.2736	35.2518	35.7599	8.7671
No log	19.0	361	3.1014	36.4014	17.4106	34.8341	35.3403	8.7397
No log	20.0	380	3.0691	36.6132	17.4341	35.0468	35.5194	8.5616
No log	21.0	399	3.0368	37.4634	18.3921	35.8956	36.3709	8.4658
No log	22.0	418	3.0071	37.1796	18.0799	35.6085	36.102	8.4247
No log	23.0	437	2.9806	37.6934	19.5239	36.4692	36.9152	8.2055
No log	24.0	456	2.9535	38.3271	20.1594	37.0697	37.6403	8.0959
No log	25.0	475	2.9325	38.5833	20.7699	37.3922	37.9437	8.1781
No log	26.0	494	2.9105	38.5591	21.1086	37.8183	38.2351	8.137
3.6364	27.0	513	2.8892	38.1741	20.492	37.4062	37.765	7.863
3.6364	28.0	532	2.8716	38.0978	20.3115	37.0709	37.3916	7.7808
3.6364	29.0	551	2.8541	38.7918	20.6816	37.4011	37.7503	7.8219
3.6364	30.0	570	2.8392	38.9202	20.7127	37.5863	37.8795	7.863
3.6364	31.0	589	2.8256	38.6036	21.0085	37.8739	38.1613	7.6164
3.6364	32.0	608	2.8122	39.0417	21.677	38.2494	38.6465	7.726
3.6364	33.0	627	2.7994	39.2329	21.7591	38.5074	38.8281	7.6986
3.6364	34.0	646	2.7862	40.9608	23.3487	39.9721	40.4826	7.6301
3.6364	35.0	665	2.7752	40.3292	23.0376	39.6256	40.123	7.6986
3.6364	36.0	684	2.7658	40.3589	22.9372	39.6409	40.1315	7.6438
3.6364	37.0	703	2.7562	40.6065	22.9372	39.8863	40.4343	7.6575
3.6364	38.0	722	2.7495	40.9141	22.9372	40.1929	40.7218	7.6575
3.6364	39.0	741	2.7425	40.5265	22.9372	39.7735	40.3237	7.6849
3.6364	40.0	760	2.7367	40.5265	22.9372	39.7735	40.3237	7.6849
3.6364	41.0	779	2.7308	40.5265	22.9372	39.7735	40.3237	7.6849
3.6364	42.0	798	2.7264	41.0514	22.9372	40.3332	40.8709	7.6986
3.6364	43.0	817	2.7233	41.0514	22.9372	40.3332	40.8709	7.6986
3.6364	44.0	836	2.7193	41.4655	23.3863	40.7719	41.274	7.7123
3.6364	45.0	855	2.7164	41.6567	23.7942	41.0101	41.5048	7.6027
3.6364	46.0	874	2.7135	41.6567	23.7942	41.0101	41.5048	7.6027
3.6364	47.0	893	2.7108	41.6567	23.7942	41.0101	41.5048	7.6027
3.6364	48.0	912	2.7092	41.6567	23.7942	41.0101	41.5048	7.6027
3.6364	49.0	931	2.7081	41.6567	23.7942	41.0101	41.5048	7.6027
3.6364	50.0	950	2.7077	41.6567	23.7942	41.0101	41.5048	7.6027

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu121
• Datasets 2.16.1
• Tokenizers 0.15.1