<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

Usage

Translates to Acholi, Lugbara, Luganda, Runyankole and Ateso

Make sure to add a target language and dataset tags before a source sentence.

Ex. >>lug_hq<< I want Posho ---> Njagala Posho

For biblical style translations attempt to use the ood tag

Ex. >>lug_ood<< And thus spoke the LORD to the masses on the mountain

We these other tags which you might want to try [ggl, bt, hq, ood]

Language tags [ach, lgg, lug, nyn, teo]

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 0.0001
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• gradient_accumulation_steps: 5000
• total_train_batch_size: 5000
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10
• label_smoothing_factor: 0.1

Framework versions

• Transformers 4.28.1
• Pytorch 2.0.0+cu118
• Tokenizers 0.13.3