mBART_skr-en_longerrun

Shamus/mBART_skr-en_longerrun

mBART_skr-en_longerrun is a machine learning model from Shamus. 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.

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

Downloads · 30 days

4% of all-time downloads

All-time downloads

316

Public

Repo size

9.8 GB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.bin2.4 GB · 99%

At a glance

Library: transformers
Model type: mbart
Access: Public
Created: Aug 27, 2022
Updated: Aug 27, 2022
SHA: 50556c8b

Model card

From the Hugging Face model README

mBART_skr-en_longerrun

This model was trained from scratch on the None dataset. It achieves the following results on the evaluation set:

Loss: 1.4577
Bleu: 30.8071
Gen Len: 34.548

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: 4
eval_batch_size: 4
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 16
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 4
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Gen Len
0.5444	0.72	500	1.3416	28.7505	34.228
0.8576	1.45	1000	1.3411	30.1776	34.328
0.6422	2.18	1500	1.3882	30.2815	34.164
0.532	2.9	2000	1.3716	30.8947	34.556
0.4473	3.63	2500	1.4577	30.8071	34.548

Framework versions

Transformers 4.21.2
Pytorch 1.12.1+cu113
Datasets 2.4.0
Tokenizers 0.12.1

Library: transformers
Type: mbart
Created: Aug 27, 2022
Updated: Aug 27, 2022