CoolFace
Modelpublic

taiypeo/bart-large-wikilarge

sourceHugging Faceapache-2.0updated 7mo agoView on Hugging Face
0likes16downloads
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. -->

bart-large-wikilarge

This model is a fine-tuned version of facebook/bart-large on an unknown dataset. It achieves the following results on the evaluation set:

  • —Loss: 0.9051
  • —Sari: 37.0457
  • —Paper Sari: 37.1345
  • —Sari Add: 4.1443
  • —Sari Keep: 77.5177
  • —Sari Del: 29.475
  • —Paper Sari Add: 4.1443
  • —Paper Sari Keep: 77.7084
  • —Paper Sari Del: 29.5508
  • —Fkgl: 8.4619
  • —Bleu: 92.6286

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: 1e-05
  • —trainbatchsize: 16
  • —evalbatchsize: 16
  • —seed: 42
  • —gradientaccumulationsteps: 2
  • —totaltrainbatch_size: 32
  • —optimizer: Use OptimizerNames.ADAMWTORCHFUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
  • —lrschedulertype: linear
  • —lrschedulerwarmup_ratio: 0.06
  • —num_epochs: 20

Training results

Training LossEpochStepValidation LossSariPaper SariSari AddSari KeepSari DelPaper Sari AddPaper Sari KeepPaper Sari DelFkglBleu
2.08740.108010000.940131.929932.09780.545177.779917.46480.545678.108917.63888.86896.2302
1.54650.215920000.890631.448331.58490.719178.649614.97630.719578.979115.0569.012196.5849
1.48140.323930000.868131.213431.33830.915879.112913.61160.916379.444813.65399.143697.0903
1.44840.431840000.867731.413731.54240.967979.260714.01230.968379.597214.06199.13397.3217
1.42370.539850000.871232.653232.77871.359278.735517.86481.360179.041717.93438.994796.5928
1.40880.647760000.871133.850533.98911.679477.733522.13851.680478.012322.27458.781495.2315
1.38270.755770000.857733.455433.57881.827378.638919.89991.827878.931819.97678.917596.2523
1.37240.863780000.854633.982234.10921.902978.113621.931.903878.390222.03358.833995.6744
1.37650.971690000.856034.385834.51992.061877.588323.50742.062777.853423.64368.733694.8992
1.33941.0796100000.854333.946634.05912.293678.89320.65322.293979.176320.70728.874695.9382
1.32151.1875110000.858634.815434.94032.522677.711424.21222.523277.96724.33078.636994.8519
1.32011.2955120000.863535.064235.18212.647577.996624.54852.64878.248724.64978.692394.8076
1.3051.4034130000.853834.312534.41912.547778.946321.44352.54879.219921.48958.885595.7623
1.30071.5114140000.859435.292435.42132.697577.169526.01022.69877.409726.15638.515793.9044
1.29411.6193150000.867035.195535.30462.920178.083624.58272.920478.329724.66378.673994.4185
1.28431.7273160000.861834.782634.90152.521478.330823.49562.521678.595523.58758.774395.1965
1.28451.8353170000.866134.420934.53912.274678.147422.84072.27578.412222.93038.778195.3004
1.27521.9432180000.876335.441435.55192.931878.058925.33352.932178.301625.4228.697594.5009
1.22752.0512190000.869135.604235.70923.233678.074125.50483.233878.311125.58268.656894.5273
1.2122.1591200000.870935.772535.87863.167477.554526.59553.167777.778126.69018.577693.7323
1.21322.2671210000.866234.903835.01332.785278.577823.34862.785378.838923.41588.8195.2465
1.20522.3750220000.870735.453835.55473.122578.43824.8013.122678.680324.86138.703594.7578
1.21752.4830230000.874535.662535.76363.254678.271925.46123.254878.508425.52768.687994.7659
1.20142.5910240000.872535.672735.7793.020677.774326.22323.020978.003326.3138.549493.8561
1.1952.6989250000.872635.756435.86213.1277.987926.16133.120278.220226.2468.59693.9884
1.20452.8069260000.870235.57935.68773.060177.884425.79263.060378.119825.8838.603194.1975
1.18822.9148270000.865935.107735.21572.903978.390724.02862.904178.643724.09938.745894.7959
1.19463.0228280000.889536.224436.32773.461277.546927.66513.461377.761927.75998.552293.2346
1.13463.1307290000.877736.497836.59593.659377.823928.01033.659478.036628.09178.520193.3092
1.13253.2387300000.884135.862735.96533.327477.760126.50063.327577.984326.58418.552993.4478
1.13453.3466310000.877436.061936.1593.472677.818326.89473.472878.036126.96798.537193.6775
1.14563.4546320000.880636.15436.2583.321977.742827.39723.322177.962827.48928.541193.3262
1.12393.5626330000.876236.259436.35543.595377.916727.26613.595578.132927.33798.561393.631
1.12273.6705340000.880935.635.70393.160578.096925.54263.160678.333325.61798.68794.0505
1.14133.7785350000.879436.695536.79953.70877.188829.18973.708277.389629.30078.422792.7837
1.14233.8864360000.895636.131836.23473.357677.58127.45683.357877.797527.54888.571793.1018
1.13393.9944370000.882936.425836.52113.695877.468328.11323.69677.671728.19578.521392.9871
1.07474.1023380000.905137.045737.13454.144377.517729.4754.144377.708429.55088.461992.6286
1.07114.2103390000.893036.280536.37983.556578.005627.27943.556578.226527.35638.61193.7963
1.07854.3183400000.892936.61836.71623.692377.39228.76993.692477.593228.86318.457492.7116
1.08184.4262410000.889636.868236.96743.828577.404429.37163.828677.603429.47028.421292.9933
1.08114.5342420000.911936.735836.82583.88877.66928.65033.888177.867928.72148.48492.7557
1.08724.6421430000.892436.360636.463.556377.741227.78433.556477.954927.86888.51293.3203
1.09684.7501440000.892136.474236.56763.73577.765427.92213.735277.972527.9958.547693.3622
1.08354.8580450000.904036.722436.81124.05377.439728.67464.053177.633528.7478.520892.7691
1.08954.9660460000.890636.70636.79893.940277.605328.57243.940477.806328.65018.528293.1717
1.05335.0740470000.908536.474936.55743.883578.143427.39783.883778.348627.43988.625393.4822
1.0275.1819480000.905536.713236.80843.781477.184529.17373.781577.377329.26648.427592.3575

Framework versions

  • —Transformers 4.57.3
  • —Pytorch 2.9.1+cu128
  • —Datasets 3.6.0
  • —Tokenizers 0.22.1