CoolFace
Modelpublic

Etelis/TweetEval_XLNET_5E

sourceHugging Facemitupdated 4y agoView on Hugging Face
0likes23downloads
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. -->

TweetEvalXLNET5E

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

  • Loss: 0.4591
  • Accuracy: 0.9333

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: 3e-05
  • trainbatchsize: 16
  • evalbatchsize: 8
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lrschedulertype: linear
  • num_epochs: 5

Training results

Training LossEpochStepValidation LossAccuracy
0.55750.04500.26750.9
0.41770.081000.21930.9067
0.29110.121500.24820.9
0.35030.162000.24240.9
0.34120.22500.19130.9267
0.27470.243000.17830.92
0.29990.283500.24950.9133
0.31410.324000.24600.9
0.29350.374500.20340.92
0.26190.415000.26000.9067
0.24540.455500.21780.92
0.28090.496000.22540.9133
0.2880.536500.18490.92
0.27690.577000.18960.9267
0.30790.617500.21530.9133
0.25980.658000.32790.9067
0.31490.698500.19850.92
0.28720.739000.18010.9333
0.25540.779500.20230.9267
0.26450.8110000.22080.9067
0.25090.8510500.20120.9333
0.24040.8911000.19950.9067
0.23610.9311500.18080.9133
0.22980.9712000.22260.9333
0.1931.0112500.25350.9267
0.16031.0613000.21630.9467
0.19161.113500.24790.92
0.19631.1414000.19640.94
0.16671.1814500.31390.9133
0.16681.2215000.22040.9267
0.16771.2615500.24680.9333
0.16011.316000.23940.94
0.17141.3416500.23260.94
0.1971.3817000.18610.94
0.17771.4217500.25180.94
0.19251.4618000.18060.94
0.20681.518500.13190.9467
0.17161.5419000.11990.9667
0.14421.5819500.16940.96
0.19291.6220000.19900.9467
0.16541.6620500.29720.9333
0.17591.721000.15840.9467
0.17881.7521500.22660.94
0.17961.7922000.27460.9333
0.1721.8322500.23130.9333
0.16371.8723000.29180.9267
0.23591.9123500.21210.9267
0.17781.9524000.20220.9333
0.15811.9924500.29360.9067
0.13122.0325000.25310.9333
0.11782.0725500.25250.9267
0.09242.1126000.27150.9333
0.07742.1526500.21230.9533
0.0912.1927000.21280.9467
0.09482.2327500.21870.9533
0.11212.2728000.24380.9467
0.12592.3128500.21970.9467
0.07472.3529000.27270.9333
0.1142.3929500.31970.9333
0.0862.4430000.36430.9333
0.13262.4830500.27910.94
0.10172.5231000.26610.9333
0.07192.5631500.27970.94
0.14242.632000.18190.96
0.1062.6432500.27700.94
0.09962.6833000.22130.94
0.08352.7233500.28940.9333
0.08082.7634000.34240.9333
0.14062.834500.21660.94
0.03452.8435000.31460.9333
0.12472.8835500.28240.9467
0.0762.9236000.26500.9467
0.1342.9636500.27580.9267
0.05213.037000.26930.9467
0.03663.0437500.34280.9333
0.06823.0838000.27790.9533
0.06243.1238500.25630.9467
0.04023.1739000.30860.94
0.0523.2139500.33240.94
0.05793.2540000.31650.9467
0.04113.2940500.35070.9467
0.05073.3341000.31080.9533
0.03263.3741500.36450.94
0.0853.4142000.33900.94
0.0223.4542500.33670.94
0.06893.4943000.34330.94
0.04583.5343500.33590.9533
0.03843.5744000.36420.9467
0.04153.6144500.34290.9467
0.03623.6545000.37270.9467
0.03513.6945500.32930.9467
0.063.7346000.47170.92
0.03443.7746500.36680.94
0.05183.8147000.34610.94
0.0463.8647500.40200.9267
0.07353.948000.26600.9467
0.04533.9448500.33640.9333
0.0393.9849000.43980.92
0.04974.0249500.34760.94
0.01834.0650000.38710.94
0.05584.150500.40660.9267
0.03584.1451000.39260.92
0.05074.1851500.33120.9467
0.01114.2252000.39760.9267
0.03634.2652500.47530.92
0.02834.353000.42340.9267
0.00974.3453500.45470.9333
0.00184.3854000.46870.9267
0.03444.4254500.42740.9333
0.0214.4655000.44480.9333
0.00924.555500.46720.9333
0.03544.5556000.46660.9333
0.0294.5956500.46140.9333
0.01824.6357000.48400.9333
0.0434.6757500.43270.9333
0.02594.7158000.46390.9333
0.02244.7558500.46070.9333
0.03024.7959000.46060.9333
0.02244.8359500.46540.9333
0.04314.8760000.46810.9333
0.02844.9160500.46220.9333
0.03264.9561000.46020.9333
0.0184.9961500.45910.9333

Framework versions

  • Transformers 4.24.0
  • Pytorch 1.13.0
  • Datasets 2.7.1
  • Tokenizers 0.13.2