Hugging Face's logo

haryoaw
/
scenario-NON-KD-PR-COPY-CDF-ALL-D2_data-cardiffnlp_tweet_sentiment_multilingual_

Transformers
PyTorch
Safetensors
deberta-v2
Generated from Trainer
Inference Endpoints
Model card Files Community
scenario-NON-KD-PR-COPY-CDF-ALL-D2_data-cardiffnlp_tweet_sentiment_multilingual_ / README.md
haryoaw's picture
haryoaw
Upload tokenizer
1e7ffe8 verified
preview code
|
raw
history blame
4.98 kB
metadata
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 on the tweet_sentiment_multilingual dataset. It achieves the following results on the evaluation set:

  • Loss: 5.1517
  • Accuracy: 0.5571
  • F1: 0.5567

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: 44
  • 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
1.0024 1.0870 500 0.9815 0.5556 0.5555
0.8411 2.1739 1000 0.9889 0.5772 0.5763
0.6502 3.2609 1500 1.1977 0.5633 0.5598
0.4723 4.3478 2000 1.6466 0.5617 0.5626
0.3276 5.4348 2500 1.7205 0.5498 0.5519
0.2221 6.5217 3000 2.0190 0.5590 0.5600
0.167 7.6087 3500 2.5446 0.5552 0.5562
0.1317 8.6957 4000 2.5112 0.5525 0.5539
0.1141 9.7826 4500 2.6152 0.5594 0.5545
0.1007 10.8696 5000 3.2079 0.5513 0.5416
0.0827 11.9565 5500 2.7099 0.5590 0.5590
0.0653 13.0435 6000 3.1595 0.5721 0.5678
0.0644 14.1304 6500 3.1304 0.5679 0.5667
0.054 15.2174 7000 3.0885 0.5590 0.5573
0.0504 16.3043 7500 3.5769 0.5583 0.5580
0.0394 17.3913 8000 3.5597 0.5606 0.5608
0.0419 18.4783 8500 3.8739 0.5525 0.5501
0.0406 19.5652 9000 3.5220 0.5667 0.5660
0.0355 20.6522 9500 4.0325 0.5691 0.5667
0.0281 21.7391 10000 3.7630 0.5602 0.5614
0.0266 22.8261 10500 4.0162 0.5617 0.5553
0.0283 23.9130 11000 3.9135 0.5525 0.5529
0.027 25.0 11500 4.0734 0.5563 0.5541
0.0205 26.0870 12000 4.2900 0.5583 0.5586
0.0198 27.1739 12500 4.2693 0.5579 0.5572
0.0155 28.2609 13000 4.7029 0.5563 0.5435
0.0187 29.3478 13500 4.4409 0.5640 0.5616
0.014 30.4348 14000 4.4588 0.5571 0.5568
0.0147 31.5217 14500 4.3420 0.5652 0.5640
0.0128 32.6087 15000 4.5721 0.5598 0.5575
0.0099 33.6957 15500 4.5574 0.5586 0.5599
0.0101 34.7826 16000 4.3777 0.5610 0.5613
0.0053 35.8696 16500 4.8103 0.5610 0.5617
0.0107 36.9565 17000 4.2925 0.5590 0.5589
0.0081 38.0435 17500 4.5884 0.5606 0.5591
0.0071 39.1304 18000 4.7187 0.5617 0.5621
0.0075 40.2174 18500 4.7305 0.5594 0.5591
0.0081 41.3043 19000 4.5589 0.5602 0.5607
0.0059 42.3913 19500 4.6516 0.5598 0.5589
0.0061 43.4783 20000 4.6553 0.5613 0.5605
0.0032 44.5652 20500 4.9672 0.5567 0.5569
0.0031 45.6522 21000 5.0283 0.5590 0.5595
0.0032 46.7391 21500 5.0801 0.5563 0.5552
0.0032 47.8261 22000 5.0934 0.5602 0.5604
0.0017 48.9130 22500 5.1305 0.5579 0.5581
0.0017 50.0 23000 5.1517 0.5571 0.5567

Framework versions

  • Transformers 4.44.2
  • Pytorch 2.1.1+cu121
  • Datasets 2.14.5
  • Tokenizers 0.19.1