Lora_long_T5_sum_challenge

floflodebilbao/Lora_long_T5_sum_challenge

Lora_long_T5_sum_challenge is a machine learning model from floflodebilbao. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for peft. The card lists the license as apache-2.0.

should probably proofread and complete it, then remove this comment. --

Downloads · 30 days

48% of all-time downloads

All-time downloads

Public

Repo size

50.9 MB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors7.1 MB · 67%

At a glance

Library: peft
License: apache-2.0
Access: Public
Created: Jul 25, 2025
Updated: Jul 29, 2025
SHA: 8dade40d

Base models

google/long-t5-tglobal-base

Model card

From the Hugging Face model README

Lora_long_T5_sum_challenge

This model is a fine-tuned version of google/long-t5-tglobal-base on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 1.2142
Rouge1: 0.2852
Rouge2: 0.0966
Rougel: 0.2231
Rougelsum: 0.2243
Gen Len: 28.38
Bleu: 0.0405
Precisions: 0.0919
Brevity Penalty: 0.8771
Length Ratio: 0.8841
Translation Length: 1068.0
Reference Length: 1208.0
Precision: 0.8739
Recall: 0.8718
F1: 0.8728
Hashcode: roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1)

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: 0.002
train_batch_size: 1
eval_batch_size: 1
seed: 42
gradient_accumulation_steps: 16
total_train_batch_size: 16
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len	Bleu	Precisions	Brevity Penalty	Length Ratio	Translation Length	Reference Length	Precision	Recall	F1	Hashcode
22.2581	1.0	7	6.5353	0.084	0.0147	0.0714	0.0714	31.0	0.0047	0.0247	0.5558	0.63	761.0	1208.0	0.7817	0.8234	0.8014	roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1)
6.7792	2.0	14	5.1759	0.1642	0.0129	0.13	0.1296	30.46	0.0	0.044	0.755	0.7806	943.0	1208.0	0.8343	0.8356	0.8349	roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1)
4.5124	3.0	21	3.7445	0.2094	0.0517	0.1669	0.1666	28.9	0.021	0.0606	0.8336	0.846	1022.0	1208.0	0.8516	0.8529	0.8521	roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1)
3.5042	4.0	28	3.1497	0.2314	0.0579	0.1774	0.1772	29.1	0.0317	0.0716	0.8537	0.8634	1043.0	1208.0	0.855	0.8584	0.8567	roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1)
2.8574	5.0	35	2.0950	0.2342	0.0664	0.1895	0.1897	28.34	0.0325	0.0756	0.8584	0.8675	1048.0	1208.0	0.8581	0.8605	0.8593	roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1)
2.0046	6.0	42	1.4599	0.2643	0.0843	0.2074	0.2081	28.18	0.036	0.0853	0.8678	0.8758	1058.0	1208.0	0.8665	0.8652	0.8658	roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1)
1.4948	7.0	49	1.2786	0.2831	0.0921	0.2203	0.2208	28.3	0.0413	0.0893	0.8855	0.8916	1077.0	1208.0	0.8703	0.8681	0.8691	roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1)
1.2731	8.0	56	1.2338	0.2802	0.096	0.2204	0.2221	28.26	0.0406	0.0893	0.8753	0.8825	1066.0	1208.0	0.8729	0.8705	0.8717	roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1)
1.1977	9.0	63	1.2179	0.2834	0.0991	0.2233	0.2244	28.42	0.0409	0.0919	0.8725	0.88	1063.0	1208.0	0.8745	0.8722	0.8733	roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1)
1.1717	10.0	70	1.2142	0.2852	0.0966	0.2231	0.2243	28.38	0.0405	0.0919	0.8771	0.8841	1068.0	1208.0	0.8739	0.8718	0.8728	roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1)

Framework versions

PEFT 0.15.2
Transformers 4.53.1
Pytorch 2.7.0+cu126
Datasets 3.6.0
Tokenizers 0.21.1

Library: peft
License: apache-2.0
Created: Jul 25, 2025
Updated: Jul 29, 2025