README.md

metadata

license: mit
base_model: arthurmluz/ptt5-xlsumm-30epochs
tags:
  - generated_from_trainer
metrics:
  - rouge
model-index:
  - name: ptt5-xlsumm-temario
    results: []

ptt5-xlsumm-temario

This model is a fine-tuned version of arthurmluz/ptt5-xlsumm-30epochs on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 2.4610
• Rouge1: 0.0891
• Rouge2: 0.0571
• Rougel: 0.0781
• Rougelsum: 0.0845
• Gen Len: 19.0

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: 2
• eval_batch_size: 2
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	1.0	88	2.7016	0.0862	0.0326	0.0681	0.0805	19.0
No log	2.0	176	2.6413	0.0879	0.0389	0.0701	0.0828	19.0
2.9296	3.0	264	2.5893	0.0881	0.0438	0.0707	0.0827	19.0
2.9296	4.0	352	2.5650	0.0923	0.0479	0.0748	0.0871	19.0
2.646	5.0	440	2.5429	0.0885	0.0469	0.0732	0.0834	19.0
2.646	6.0	528	2.5247	0.088	0.0503	0.0739	0.0831	19.0
2.5072	7.0	616	2.5108	0.0891	0.0534	0.0769	0.0851	19.0
2.5072	8.0	704	2.5039	0.0884	0.0547	0.0764	0.0848	19.0
2.5072	9.0	792	2.4948	0.0864	0.0536	0.0751	0.083	19.0
2.4128	10.0	880	2.4836	0.0869	0.0546	0.076	0.0839	19.0
2.4128	11.0	968	2.4813	0.0866	0.0543	0.0764	0.0832	19.0
2.356	12.0	1056	2.4768	0.0864	0.0533	0.076	0.0828	19.0
2.356	13.0	1144	2.4728	0.0872	0.0556	0.0775	0.0838	19.0
2.2815	14.0	1232	2.4666	0.0877	0.0557	0.0774	0.0841	19.0
2.2815	15.0	1320	2.4667	0.0866	0.0552	0.0764	0.0829	19.0
2.2106	16.0	1408	2.4680	0.0869	0.0553	0.0772	0.0824	19.0
2.2106	17.0	1496	2.4647	0.0867	0.0553	0.0771	0.0828	19.0
2.2106	18.0	1584	2.4597	0.0875	0.0561	0.0777	0.0837	19.0
2.1809	19.0	1672	2.4601	0.0873	0.0557	0.0773	0.0833	19.0
2.1809	20.0	1760	2.4596	0.0873	0.0561	0.0773	0.0835	19.0
2.1541	21.0	1848	2.4592	0.0875	0.0561	0.0777	0.0837	19.0
2.1541	22.0	1936	2.4620	0.0869	0.0551	0.0768	0.0828	19.0
2.1442	23.0	2024	2.4621	0.0869	0.0551	0.0768	0.0828	19.0
2.1442	24.0	2112	2.4619	0.0868	0.0553	0.0768	0.0828	19.0
2.1071	25.0	2200	2.4613	0.0868	0.0553	0.0768	0.0828	19.0
2.1071	26.0	2288	2.4618	0.0873	0.0557	0.0768	0.0828	19.0
2.1071	27.0	2376	2.4607	0.0892	0.0575	0.0782	0.0847	19.0
2.08	28.0	2464	2.4606	0.0874	0.056	0.0769	0.083	19.0
2.08	29.0	2552	2.4616	0.0891	0.0571	0.0781	0.0845	19.0
2.1013	30.0	2640	2.4610	0.0891	0.0571	0.0781	0.0845	19.0

Framework versions

• Transformers 4.34.0
• Pytorch 2.0.1+cu117
• Datasets 2.14.5
• Tokenizers 0.14.1