poem_sentiment

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

• Loss: 0.4747
• 0: {'precision': 0.8571428571428571, 'recall': 0.9473684210526315, 'f1-score': 0.9, 'support': 19}
• 1: {'precision': 0.7222222222222222, 'recall': 0.7647058823529411, 'f1-score': 0.7428571428571428, 'support': 17}
• 2: {'precision': 0.9393939393939394, 'recall': 0.8985507246376812, 'f1-score': 0.9185185185185185, 'support': 69}
• Accuracy: 0.8857
• Macro avg: {'precision': 0.8395863395863395, 'recall': 0.8702083426810846, 'f1-score': 0.8537918871252205, 'support': 105}
• Weighted avg: {'precision': 0.8893492750635609, 'recall': 0.8857142857142857, 'f1-score': 0.8867271352985638, 'support': 105}

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	0	1	2	Accuracy	Macro avg	Weighted avg
1.0922	1.0	112	0.8825	{'precision': 0.0, 'recall': 0.0, 'f1-score': 0.0, 'support': 19}	{'precision': 0.0, 'recall': 0.0, 'f1-score': 0.0, 'support': 17}	{'precision': 0.6571428571428571, 'recall': 1.0, 'f1-score': 0.7931034482758621, 'support': 69}	0.6571	{'precision': 0.21904761904761905, 'recall': 0.3333333333333333, 'f1-score': 0.26436781609195403, 'support': 105}	{'precision': 0.43183673469387757, 'recall': 0.6571428571428571, 'f1-score': 0.5211822660098522, 'support': 105}
0.6877	2.0	224	0.4747	{'precision': 0.8571428571428571, 'recall': 0.9473684210526315, 'f1-score': 0.9, 'support': 19}	{'precision': 0.7222222222222222, 'recall': 0.7647058823529411, 'f1-score': 0.7428571428571428, 'support': 17}	{'precision': 0.9393939393939394, 'recall': 0.8985507246376812, 'f1-score': 0.9185185185185185, 'support': 69}	0.8857	{'precision': 0.8395863395863395, 'recall': 0.8702083426810846, 'f1-score': 0.8537918871252205, 'support': 105}	{'precision': 0.8893492750635609, 'recall': 0.8857142857142857, 'f1-score': 0.8867271352985638, 'support': 105}
0.5299	3.0	336	0.6595	{'precision': 0.8, 'recall': 0.8421052631578947, 'f1-score': 0.8205128205128205, 'support': 19}	{'precision': 1.0, 'recall': 0.4117647058823529, 'f1-score': 0.5833333333333334, 'support': 17}	{'precision': 0.8461538461538461, 'recall': 0.9565217391304348, 'f1-score': 0.8979591836734695, 'support': 69}	0.8476	{'precision': 0.882051282051282, 'recall': 0.7367972360568942, 'f1-score': 0.7672684458398744, 'support': 105}	{'precision': 0.8627106227106227, 'recall': 0.8476190476190476, 'f1-score': 0.8330056564750442, 'support': 105}
0.9027	4.0	448	0.5981	{'precision': 1.0, 'recall': 0.7368421052631579, 'f1-score': 0.8484848484848484, 'support': 19}	{'precision': 0.7333333333333333, 'recall': 0.6470588235294118, 'f1-score': 0.6875, 'support': 17}	{'precision': 0.868421052631579, 'recall': 0.9565217391304348, 'f1-score': 0.9103448275862069, 'support': 69}	0.8667	{'precision': 0.867251461988304, 'recall': 0.7801408893076681, 'f1-score': 0.8154432253570185, 'support': 105}	{'precision': 0.870359231411863, 'recall': 0.8666666666666667, 'f1-score': 0.863071478330099, 'support': 105}
0.4588	5.0	560	0.7815	{'precision': 0.7727272727272727, 'recall': 0.8947368421052632, 'f1-score': 0.8292682926829269, 'support': 19}	{'precision': 0.6470588235294118, 'recall': 0.6470588235294118, 'f1-score': 0.6470588235294118, 'support': 17}	{'precision': 0.8939393939393939, 'recall': 0.855072463768116, 'f1-score': 0.8740740740740741, 'support': 69}	0.8286	{'precision': 0.7712418300653595, 'recall': 0.7989560431342637, 'f1-score': 0.7834670634288043, 'support': 105}	{'precision': 0.832034632034632, 'recall': 0.8285714285714286, 'f1-score': 0.8292115111627308, 'support': 105}

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0