sanchit-gandhi/xtreme_s_xlsr_2_bart_covost2_fr_en

This model was trained from scratch on the xtreme_s dataset. It achieves the following results on the evaluation set:

• Loss: 2.1356
• Bleu: 0.0000

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: 3e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 16
• total_train_batch_size: 128
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 3.0
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu
2.6458	0.31	500	4.3754	0.0
2.3505	0.62	1000	4.3071	0.0
2.2152	0.93	1500	3.9444	0.0
2.79	1.23	2000	3.2046	0.0000
2.569	1.54	2500	2.6812	0.0000
2.322	1.85	3000	2.4081	0.0000
2.3435	2.16	3500	2.2696	0.0000
2.2063	2.47	4000	2.2452	0.0000
2.1087	2.78	4500	2.1356	0.0000

Framework versions

• Transformers 4.19.0.dev0
• Pytorch 1.10.2+cu113
• Datasets 2.1.1.dev0
• Tokenizers 0.11.0