seq2seq-imlla-test – AI Model by mbole | AlphaNeural AI

You can deploy this model and start earning money today!

mbole/
seq2seq-imlla-test

like

0

generated_from_trainer

endpoints_compatible

Views

No views yet

seq2seq-imlla-test

This model is a fine-tuned version of on the iva_mt_wslot dataset. It achieves the following results on the evaluation set:

Loss: 6.0796
Bleu: 0.0023
Gen Len: 3.9927

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-05
train_batch_size: 8
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 32
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 500
num_epochs: 5
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Gen Len
7.3774	0.9992	636	7.0402	0.0	4.1578
6.528	2.0	1273	6.4745	0.0141	9.5716
6.0907	2.9992	1909	6.2223	0.0024	5.0011
5.903	4.0	2546	6.1164	0.0019	4.0774
5.7639	4.9961	3180	6.0796	0.0023	3.9927

Framework versions

Transformers 4.44.2
Pytorch 2.4.1+cu121
Datasets 3.1.0
Tokenizers 0.19.1