README.md

metadata

library_name: transformers
base_model: cardiffnlp/twitter-roberta-base-sentiment-latest
tags:
  - generated_from_trainer
metrics:
  - accuracy
model-index:
  - name: finetuning-sentiment-model-tweet-finalVersion
    results: []

finetuning-sentiment-model-tweet-finalVersion

This model is a fine-tuned version of cardiffnlp/twitter-roberta-base-sentiment-latest on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.8178
• Precision Negative: 0.8125
• Recall Negative: 0.7222
• F1 Negative: 0.7647
• Precision Neutral: 0.8140
• Recall Neutral: 0.875
• F1 Neutral: 0.8434
• Precision Positive: 0.8889
• Recall Positive: 0.8571
• F1 Positive: 0.8727
• Accuracy: 0.8372
• Confusion Matrix: [[26, 9, 1], [5, 70, 5], [1, 7, 48]]

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: 16
• seed: 42
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 6

Training results

Training Loss	Epoch	Step	Validation Loss	Precision Negative	Recall Negative	F1 Negative	Precision Neutral	Recall Neutral	F1 Neutral	Precision Positive	Recall Positive	F1 Positive	Accuracy	Confusion Matrix
0.496	1.0	22	0.7011	0.875	0.5833	0.7	0.7792	0.75	0.7643	0.7183	0.9107	0.8031	0.7674	[[21, 12, 3], [3, 60, 17], [0, 5, 51]]
0.3789	2.0	44	0.6227	0.725	0.8056	0.7632	0.7582	0.8625	0.8070	0.9756	0.7143	0.8247	0.8023	[[29, 7, 0], [10, 69, 1], [1, 15, 40]]
0.1735	3.0	66	0.6720	0.7879	0.7222	0.7536	0.8	0.85	0.8242	0.8704	0.8393	0.8545	0.8198	[[26, 9, 1], [6, 68, 6], [1, 8, 47]]
0.1261	4.0	88	0.7001	0.8387	0.7222	0.7761	0.8046	0.875	0.8383	0.8704	0.8393	0.8545	0.8314	[[26, 9, 1], [4, 70, 6], [1, 8, 47]]
0.0555	5.0	110	0.7969	0.8387	0.7222	0.7761	0.8140	0.875	0.8434	0.8727	0.8571	0.8649	0.8372	[[26, 9, 1], [4, 70, 6], [1, 7, 48]]
0.035	6.0	132	0.8178	0.8125	0.7222	0.7647	0.8140	0.875	0.8434	0.8889	0.8571	0.8727	0.8372	[[26, 9, 1], [5, 70, 5], [1, 7, 48]]

Framework versions

• Transformers 4.46.3
• Pytorch 2.5.1+cu121
• Datasets 3.1.0
• Tokenizers 0.20.3