README.md

metadata

base_model: d0rj/rut5-base-summ
tags:
  - generated_from_trainer
metrics:
  - rouge
model-index:
  - name: summary_resume_keywords
    results: []

summary_resume_keywords

This model is a fine-tuned version of d0rj/rut5-base-summ on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 2.9737
• Rouge1: 0.2285
• Rouge2: 0.1524
• Rougel: 0.2285
• Rougelsum: 0.2285
• Gen Len: 51.3333

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: 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: 25
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	1.0	90	2.7766	0.2485	0.1111	0.2485	0.2485	52.0
No log	2.0	180	2.7734	0.2556	0.1404	0.2389	0.2389	53.6667
No log	3.0	270	2.7763	0.2882	0.1368	0.2557	0.2557	51.6667
No log	4.0	360	2.7921	0.2722	0.1404	0.2389	0.2389	58.3333
No log	5.0	450	2.8146	0.2778	0.1622	0.2607	0.2607	57.3333
2.1351	6.0	540	2.8387	0.2778	0.1622	0.2607	0.2607	57.3333
2.1351	7.0	630	2.8569	0.2778	0.1622	0.2607	0.2607	57.3333
2.1351	8.0	720	2.8736	0.2538	0.1524	0.2538	0.2538	55.3333
2.1351	9.0	810	2.8883	0.2538	0.1524	0.2538	0.2538	55.3333
2.1351	10.0	900	2.9025	0.2315	0.1524	0.2315	0.2315	51.0
2.1351	11.0	990	2.9161	0.2315	0.1524	0.2315	0.2315	51.0
1.7131	12.0	1080	2.9269	0.2315	0.1524	0.2315	0.2315	51.0
1.7131	13.0	1170	2.9354	0.226	0.1524	0.226	0.226	54.0
1.7131	14.0	1260	2.9427	0.226	0.1524	0.226	0.226	54.0
1.7131	15.0	1350	2.9471	0.2272	0.1524	0.2272	0.2272	53.6667
1.7131	16.0	1440	2.9509	0.226	0.1524	0.226	0.226	54.0
1.5914	17.0	1530	2.9558	0.2272	0.1524	0.2272	0.2272	53.6667
1.5914	18.0	1620	2.9589	0.226	0.1524	0.226	0.226	54.0
1.5914	19.0	1710	2.9636	0.2285	0.1524	0.2285	0.2285	51.0
1.5914	20.0	1800	2.9660	0.2285	0.1524	0.2285	0.2285	51.0
1.5914	21.0	1890	2.9687	0.2285	0.1524	0.2285	0.2285	50.3333
1.5914	22.0	1980	2.9709	0.2285	0.1524	0.2285	0.2285	50.3333
1.5508	23.0	2070	2.9736	0.2285	0.1524	0.2285	0.2285	50.3333
1.5508	24.0	2160	2.9742	0.2285	0.1524	0.2285	0.2285	50.3333
1.5508	25.0	2250	2.9737	0.2285	0.1524	0.2285	0.2285	51.3333

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu121
• Datasets 2.17.0
• Tokenizers 0.15.2