--- base_model: microsoft/mdeberta-v3-base datasets: - tweet_sentiment_multilingual library_name: transformers license: mit metrics: - accuracy - f1 tags: - generated_from_trainer model-index: - name: scenario-NON-KD-PR-COPY-CDF-ALL-D2_data-cardiffnlp_tweet_sentiment_multilingual_ results: [] --- # scenario-NON-KD-PR-COPY-CDF-ALL-D2_data-cardiffnlp_tweet_sentiment_multilingual_ This model is a fine-tuned version of [microsoft/mdeberta-v3-base](https://huggingface.co/microsoft/mdeberta-v3-base) on the tweet_sentiment_multilingual dataset. It achieves the following results on the evaluation set: - Loss: 5.0035 - Accuracy: 0.5625 - F1: 0.5617 ## 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: 5e-05 - train_batch_size: 32 - eval_batch_size: 32 - seed: 55 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 50 ### Training results | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | |:-------------:|:-------:|:-----:|:---------------:|:--------:|:------:| | 1.021 | 1.0870 | 500 | 1.0027 | 0.5409 | 0.5367 | | 0.8432 | 2.1739 | 1000 | 1.0327 | 0.5814 | 0.5820 | | 0.6715 | 3.2609 | 1500 | 1.1554 | 0.5822 | 0.5778 | | 0.48 | 4.3478 | 2000 | 1.4182 | 0.5613 | 0.5573 | | 0.3384 | 5.4348 | 2500 | 1.8214 | 0.5567 | 0.5573 | | 0.2309 | 6.5217 | 3000 | 1.8385 | 0.5502 | 0.5445 | | 0.1737 | 7.6087 | 3500 | 2.0368 | 0.5444 | 0.5440 | | 0.1324 | 8.6957 | 4000 | 2.3667 | 0.5424 | 0.5414 | | 0.1132 | 9.7826 | 4500 | 2.0414 | 0.5509 | 0.5486 | | 0.1058 | 10.8696 | 5000 | 2.5673 | 0.5509 | 0.5491 | | 0.0833 | 11.9565 | 5500 | 2.7424 | 0.5513 | 0.5509 | | 0.0662 | 13.0435 | 6000 | 3.2582 | 0.5544 | 0.5529 | | 0.0664 | 14.1304 | 6500 | 3.5005 | 0.5556 | 0.5521 | | 0.0532 | 15.2174 | 7000 | 3.0692 | 0.5502 | 0.5509 | | 0.0494 | 16.3043 | 7500 | 3.1700 | 0.5478 | 0.5487 | | 0.0485 | 17.3913 | 8000 | 3.8948 | 0.5382 | 0.5377 | | 0.0359 | 18.4783 | 8500 | 3.5655 | 0.5583 | 0.5570 | | 0.0322 | 19.5652 | 9000 | 4.0121 | 0.5583 | 0.5547 | | 0.0294 | 20.6522 | 9500 | 3.5540 | 0.5579 | 0.5582 | | 0.026 | 21.7391 | 10000 | 4.0054 | 0.5525 | 0.5535 | | 0.0305 | 22.8261 | 10500 | 3.8289 | 0.5498 | 0.5453 | | 0.0232 | 23.9130 | 11000 | 4.4012 | 0.5556 | 0.5558 | | 0.0209 | 25.0 | 11500 | 4.0916 | 0.5559 | 0.5504 | | 0.0224 | 26.0870 | 12000 | 4.3087 | 0.5586 | 0.5583 | | 0.0192 | 27.1739 | 12500 | 4.0617 | 0.5467 | 0.5474 | | 0.0198 | 28.2609 | 13000 | 4.1456 | 0.5567 | 0.5555 | | 0.0148 | 29.3478 | 13500 | 4.5847 | 0.5505 | 0.5519 | | 0.016 | 30.4348 | 14000 | 4.3128 | 0.5494 | 0.5501 | | 0.0145 | 31.5217 | 14500 | 4.4021 | 0.5505 | 0.5500 | | 0.0146 | 32.6087 | 15000 | 4.3393 | 0.5509 | 0.5506 | | 0.0089 | 33.6957 | 15500 | 4.4852 | 0.5486 | 0.5499 | | 0.0089 | 34.7826 | 16000 | 4.8487 | 0.5475 | 0.5487 | | 0.0085 | 35.8696 | 16500 | 4.8052 | 0.5567 | 0.5573 | | 0.0077 | 36.9565 | 17000 | 4.6518 | 0.5502 | 0.5484 | | 0.0095 | 38.0435 | 17500 | 4.2742 | 0.5567 | 0.5554 | | 0.0054 | 39.1304 | 18000 | 4.7804 | 0.5548 | 0.5520 | | 0.0074 | 40.2174 | 18500 | 4.6940 | 0.5540 | 0.5516 | | 0.0053 | 41.3043 | 19000 | 4.6543 | 0.5590 | 0.5581 | | 0.003 | 42.3913 | 19500 | 5.0637 | 0.5563 | 0.5572 | | 0.0044 | 43.4783 | 20000 | 4.7918 | 0.5652 | 0.5657 | | 0.0053 | 44.5652 | 20500 | 4.7492 | 0.5625 | 0.5604 | | 0.0031 | 45.6522 | 21000 | 4.8642 | 0.5571 | 0.5567 | | 0.0026 | 46.7391 | 21500 | 4.9137 | 0.5617 | 0.5614 | | 0.0025 | 47.8261 | 22000 | 4.8985 | 0.5629 | 0.5626 | | 0.0007 | 48.9130 | 22500 | 4.9890 | 0.5633 | 0.5621 | | 0.0027 | 50.0 | 23000 | 5.0035 | 0.5625 | 0.5617 | ### Framework versions - Transformers 4.44.2 - Pytorch 2.1.1+cu121 - Datasets 2.14.5 - Tokenizers 0.19.1