ChloeKa/lora-llama-2-7b-food-order-understanding

<Llama2 모델을 nsmc 데이터셋을 해결하는 모델이 되도록 미세튜닝 하기>

모델: Llama2
데이터셋: nsmc
https://huggingface.co/datasets/nsmc
Train 데이터: 3000
Test 데이터: 1000

[테스트 결과]

정확도: 86.10%

혼동행렬(Confusion Matrix)

	정답 Positive	정답 Negative
예측 Positive	395	26
예측 Negative	113	466

평가지표

정밀도(Precision)	0.938
재현율(Recall)	0.459
F1 Score	0.616

[성능 향상]

train 데이터 수를 2000에서 2500, 3000으로 늘려가며 성능을 약 11% 정도 높였으며, TrainingArguments의 max_steps 등의 파라미터를 조절해가며 성능을 높이고자 노력하였다.

---

lora-llama-2-7b-food-order-understanding

This model is a fine-tuned version of meta-llama/Llama-2-7b-chat-hf on an unknown dataset.

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: 0.0001
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 2
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.03
• training_steps: 300
• mixed_precision_training: Native AMP

Training results

Framework versions

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