CoolFace
Modelpublic

sbartlett97/gqa-opus-mt-de-en

sourceHugging Faceapache-2.0updated 9mo agoView on Hugging Face
0likes17downloads
Model Card

<!-- 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. -->

IMNPORTANT: This is an experimental model that uses a custom modified architecture for the MarianMT models. (It has also seen insufficient data to learn properly so will perform badly)

gqa-opus-mt-de-en

This model is a fine-tuned version of sbartlett97/gqa-opus-mt-de-en on an unknown dataset. It achieves the following results on the evaluation set:

  • —Loss: 2.6134
  • —Bleu: 0.2252
  • —Gen Len: 13.0925

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
  • —trainbatchsize: 16
  • —evalbatchsize: 16
  • —seed: 42
  • —optimizer: Use OptimizerNames.ADAMWTORCHFUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
  • —lrschedulertype: linear
  • —lrschedulerwarmup_steps: 500
  • —num_epochs: 25
  • —mixedprecisiontraining: Native AMP

Training results

Training LossEpochStepValidation LossBleuGen Len
1.54031.0625001.96700.231812.8925
1.50172.01250001.95680.228512.8835
1.43143.01875001.96670.222712.863
1.40564.02500001.97810.230612.9815
1.34335.03125002.00430.231212.937
1.32286.03750002.02100.228312.9465
1.22247.04375002.04800.228412.9285
1.21038.05000002.08130.230512.901
1.16479.05625002.11050.228812.93
1.114810.06250002.13480.228912.989
1.101411.06875002.14480.228212.924
1.069012.07500002.17970.225812.9675
0.999413.08125002.21470.227212.996
0.974814.08750002.23850.225112.9535
0.950415.09375002.28630.22412.975
0.914716.010000002.31620.224112.9705
0.856517.010625002.35610.227213.012
0.820418.011250002.38460.227313.0055
0.778519.011875002.43340.221712.991
0.760320.012500002.46390.223713.0475
0.715321.013125002.50140.221313.051
0.676122.013750002.53000.221613.0385
0.635223.014375002.56240.221913.078
0.607924.015000002.59570.224513.087
0.572325.015625002.61340.225213.0925

Framework versions

  • —Transformers 5.0.0.dev0
  • —Pytorch 2.9.1+cu128
  • —Datasets 4.4.1
  • —Tokenizers 0.22.1