CoolFace
Modelpublic

cartesinus/iva_mt_wslot-m2m100_418M-en-pl-plaintext

sourceHugging Facemitupdated 3y agoView on Hugging Face
1likes10downloads
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. -->

ivamtwslot-m2m100418M-en-pl-plaintext10e

This model is a fine-tuned version of facebook/m2m100_418M on the ivamtwslot dataset. It achieves the following results on the evaluation set:

  • —Loss: 0.0169
  • —Bleu: 41.3124
  • —Gen Len: 15.5197

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
  • —trainbatchsize: 4
  • —evalbatchsize: 4
  • —seed: 42
  • —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • —lrschedulertype: linear
  • —num_epochs: 10
  • —mixedprecisiontraining: Native AMP

Training results

Training LossEpochStepValidation LossBleuGen Len
0.01691.050910.016236.66315.6444
0.01242.0101820.015138.3615.6314
0.00863.0152730.015039.380815.5507
0.00694.0203640.015239.630715.5235
0.00495.0254550.015640.444115.5911
0.00386.0305460.015940.378115.47
0.00277.0356370.016340.133915.4722
0.00218.0407280.016641.442915.4906
0.00169.0458190.016841.102415.5249
0.001210.0509100.016941.312415.5197

Framework versions

  • —Transformers 4.28.1
  • —Pytorch 2.0.1+cu118
  • —Datasets 2.14.4
  • —Tokenizers 0.13.3