generated_from_trainer

<!-- 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. -->

raw_disaster_tweets

This model is a fine-tuned version of cardiffnlp/twitter-roberta-base-sentiment on the None dataset. It achieves the following results on the evaluation set:

Training hyperparameters

The following hyperparameters were used during training:

Training results

Training Loss Epoch Step Validation Loss
0.7201 0.12 12 0.7164
0.7234 0.25 24 0.7148
0.7128 0.38 36 0.7121
0.7122 0.5 48 0.7086
0.7074 0.62 60 0.7042
0.699 0.75 72 0.6990
0.6989 0.88 84 0.6934
0.6904 1.0 96 0.6871
0.6884 1.12 108 0.6801
0.6793 1.25 120 0.6723
0.6746 1.38 132 0.6638
0.6635 1.5 144 0.6539
0.647 1.62 156 0.6426
0.6418 1.75 168 0.6293
0.6287 1.88 180 0.6140
0.5938 2.0 192 0.5959
0.5869 2.12 204 0.5768
0.5718 2.25 216 0.5586
0.5638 2.38 228 0.5413
0.5265 2.5 240 0.5259
0.5165 2.62 252 0.5111
0.4966 2.75 264 0.4987
0.5072 2.88 276 0.4886
0.456 3.0 288 0.4827
0.458 3.12 300 0.4704
0.434 3.25 312 0.4647
0.421 3.38 324 0.4604
0.4679 3.5 336 0.4546
0.4463 3.62 348 0.4498
0.4288 3.75 360 0.4500
0.4261 3.88 372 0.4406
0.4314 4.0 384 0.4420
0.3848 4.12 396 0.4349
0.4145 4.25 408 0.4354
0.4151 4.38 420 0.4277
0.4034 4.5 432 0.4273
0.3933 4.62 444 0.4295
0.4171 4.75 456 0.4216
0.3926 4.88 468 0.4179
0.3739 5.0 480 0.4184

Framework versions