bart-large-d-wikipedia-nested-wasserstein-simultaneous-d-wikipedia-filtered

taiypeo/bart-large-d-wikipedia-nested-wasserstein-simultaneous-d-wikipedia-filtered

bart-large-d-wikipedia-nested-wasserstein-simultaneous-d-wikipedia-filtered is a machine learning model from taiypeo. 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 transformers. The card lists the license as apache-2.0.

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

Downloads · 30 days

1% of all-time downloads

All-time downloads

179

Public

Parameters

406M

19.5 GB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors1.6 GB · 100%

At a glance

Library: transformers
License: apache-2.0
Model type: bart
Access: Public
Created: Mar 10, 2026
Updated: Mar 10, 2026
SHA: 5f1fe151

Base models

taiypeo/bart-large-d-wikipedia

Model card

From the Hugging Face model README

bart-large-d-wikipedia-nested-wasserstein-simultaneous-d-wikipedia-filtered

This model is a fine-tuned version of taiypeo/bart-large-d-wikipedia on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 7.8003
Sari: 48.07
Sari Add: 8.6156
Sari Keep: 46.8836
Sari Del: 88.7109
Fkgl: 6.773
Bleu: 18.057
D Sari: 0.4385
D Sari Keep: 0.3739
D Sari Del: 0.7925
D Sari Add: 0.149

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: 5e-06
train_batch_size: 4
eval_batch_size: 4
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 16
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.06
num_epochs: 5
label_smoothing_factor: 0.3

Training results

Training Loss	Epoch	Step	Validation Loss	Sari	Sari Add	Sari Keep	Sari Del	Fkgl	Bleu	D Sari	D Sari Keep	D Sari Del	D Sari Add
4.9128	0.0092	50	7.8191	48.0402	8.537	46.9126	88.6712	6.7916	18.1825	0.4382	0.3741	0.7924	0.1479
4.908	0.0183	100	7.8003	48.07	8.6156	46.8836	88.7109	6.773	18.057	0.4385	0.3739	0.7925	0.149
4.9255	0.0275	150	7.7912	48.0397	8.5857	46.8444	88.6888	6.8081	18.0571	0.4378	0.3737	0.7912	0.1485
4.9353	0.0366	200	7.7878	47.9469	8.4521	46.7635	88.625	6.8543	18.1372	0.4366	0.3719	0.7901	0.1477
4.9067	0.0458	250	7.7905	47.8839	8.4484	46.6134	88.59	6.8639	18.0847	0.4371	0.3725	0.7907	0.1481
4.9259	0.0549	300	7.7952	47.8745	8.4987	46.6361	88.4889	6.9028	18.4706	0.4369	0.3722	0.7895	0.1489
4.8911	0.0641	350	7.7962	47.8536	8.5256	46.617	88.4184	6.9143	18.6647	0.4365	0.3724	0.7886	0.1485
4.883	0.0732	400	7.7897	47.7662	8.4449	46.5661	88.2878	6.9882	18.9553	0.4363	0.3728	0.7877	0.1483
4.907	0.0824	450	7.7843	47.8667	8.5064	46.6532	88.4405	6.9241	18.6027	0.4367	0.3728	0.7886	0.1488
4.9278	0.0915	500	7.7898	47.8113	8.5584	46.5695	88.3061	6.98	18.9826	0.4369	0.3736	0.788	0.1491
4.8779	0.1007	550	7.7666	47.7729	8.5156	46.4751	88.3281	6.9531	18.7972	0.4357	0.3725	0.7864	0.1483
4.9066	0.1098	600	7.7690	47.7264	8.4605	46.5025	88.2162	7.0237	19.1396	0.435	0.3727	0.785	0.1474

Framework versions

Transformers 4.57.3
Pytorch 2.9.1+cu128
Datasets 3.6.0
Tokenizers 0.22.1

Library: transformers
Type: bart
License: apache-2.0
Created: Mar 10, 2026
Updated: Mar 10, 2026