CoolFace
Modelpublic

TestZee/t5-small-finetuned-pytorch-final

sourceHugging Faceapache-2.0updated 4y agoView on Hugging Face
0likes1.3kdownloads
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. -->

t5-small-finetuned-pytorch-final

This model is a fine-tuned version of t5-small on an unknown dataset. It achieves the following results on the evaluation set:

  • —Loss: 1.6589
  • —Rouge1: 25.8199
  • —Rouge2: 14.6736
  • —Rougel: 22.3682
  • —Rougelsum: 23.917
  • —Gen Len: 19.0

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: 8
  • —evalbatchsize: 8
  • —seed: 42
  • —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • —lrschedulertype: linear
  • —num_epochs: 35
  • —mixedprecisiontraining: Native AMP

Training results

Training LossEpochStepValidation LossRouge1Rouge2RougelRougelsumGen Len
2.09891.05031.815224.436513.311320.986222.635119.0
1.97042.010061.772024.884313.708421.455122.978719.0
1.92413.015091.746825.198814.05221.704123.298119.0
1.88564.020121.726825.50214.357321.944923.574119.0
1.86535.025151.715625.646114.387222.082423.650219.0
1.83676.030181.705725.651314.484222.231423.700919.0
1.81837.035211.699325.637714.48122.18523.685519.0
1.80688.040241.694525.527514.318422.0323.552419.0
1.79599.045271.688525.423214.244321.969123.471119.0
1.774110.050301.684025.516914.265422.051823.586419.0
1.766511.055331.681725.523714.375822.09423.589119.0
1.754112.060361.677925.257214.193921.81623.357719.0
1.747913.065391.676125.392214.417322.029923.516319.0
1.730814.070421.674225.363114.290622.022123.512819.0
1.726115.075451.671725.431814.349322.045423.527819.0
1.718116.080481.669125.404314.32522.025223.542319.0
1.704817.085511.669125.640614.542422.237723.732519.0
1.706418.090541.667125.498614.317722.062923.594319.0
1.700319.095571.668725.619614.454622.207923.718419.0
1.685820.0100601.666025.686414.587422.307123.815119.0
1.686121.0105631.664825.669814.528122.271723.79719.0
1.68422.0110661.663525.710414.539322.257323.82919.0
1.675123.0115691.661525.725414.592322.250923.843919.0
1.674124.0120721.662425.782114.66322.316423.880919.0
1.676525.0125751.662125.768914.579622.277923.876519.0
1.656226.0130781.661625.785614.622422.329823.921519.0
1.663627.0135811.661025.8314.656922.422923.940419.0
1.67128.0140841.659925.685714.512622.209323.78819.0
1.646729.0145871.660225.811114.678422.359923.913219.0
1.655630.0150901.659425.888714.724422.399823.973919.0
1.646331.0155931.659425.853414.696622.386723.943919.0
1.654832.0160961.660025.83914.67422.376323.943819.0
1.645833.0165991.659025.836414.658922.351223.924319.0
1.643134.0171021.659025.831414.673322.352623.928819.0
1.663735.0176051.658925.819914.673622.368223.91719.0

Framework versions

  • —Transformers 4.17.0
  • —Pytorch 1.12.1+cu113
  • —Datasets 2.6.1
  • —Tokenizers 0.13.2