README.md

metadata

base_model: weny22/sum_model_t5_saved
tags:
  - generated_from_trainer
metrics:
  - rouge
model-index:
  - name: extract_long_text_unbalanced_smaller_5
    results: []

extract_long_text_unbalanced_smaller_5

This model is a fine-tuned version of weny22/sum_model_t5_saved on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.2242
• Rouge1: 0.2008
• Rouge2: 0.0688
• Rougel: 0.1593
• Rougelsum: 0.1594
• Gen Len: 18.9847

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: 64
• eval_batch_size: 64
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	1.0	72	2.3970	0.1842	0.0572	0.1461	0.1458	18.98
No log	2.0	144	2.2826	0.1923	0.0623	0.1516	0.1515	19.0
No log	3.0	216	2.2308	0.1945	0.0634	0.1529	0.1527	18.9953
No log	4.0	288	2.1962	0.1944	0.0636	0.1528	0.1527	18.9967
No log	5.0	360	2.1940	0.1948	0.0633	0.1529	0.1528	18.9953
No log	6.0	432	2.1734	0.1882	0.0628	0.1492	0.1491	18.99
3.0387	7.0	504	2.1584	0.1964	0.0663	0.156	0.1559	18.992
3.0387	8.0	576	2.1588	0.197	0.068	0.1563	0.1562	18.9847
3.0387	9.0	648	2.1852	0.1967	0.0669	0.156	0.1559	18.9793
3.0387	10.0	720	2.1859	0.201	0.0685	0.159	0.1587	18.982
3.0387	11.0	792	2.1760	0.1936	0.0643	0.1534	0.1531	18.9953
3.0387	12.0	864	2.2081	0.1978	0.0672	0.1566	0.1564	18.9753
3.0387	13.0	936	2.2030	0.1991	0.068	0.1584	0.158	18.9833
2.204	14.0	1008	2.2029	0.1981	0.0686	0.1578	0.1578	18.9867
2.204	15.0	1080	2.2076	0.2016	0.0694	0.1595	0.1592	18.9773
2.204	16.0	1152	2.2172	0.203	0.0716	0.1617	0.1617	18.9893
2.204	17.0	1224	2.2136	0.2018	0.0697	0.1604	0.1603	18.9827
2.204	18.0	1296	2.2147	0.2016	0.0695	0.1601	0.1599	18.988
2.204	19.0	1368	2.2224	0.2007	0.0687	0.1592	0.1592	18.9847
2.204	20.0	1440	2.2242	0.2008	0.0688	0.1593	0.1594	18.9847

Framework versions

• Transformers 4.39.1
• Pytorch 2.2.1+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2