File size: 4,400 Bytes
03b5e99 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 |
---
license: other
library_name: peft
tags:
- generated_from_trainer
metrics:
- precision
- recall
- accuracy
base_model: microsoft/phi-2
model-index:
- name: Phi2-Seq-classification-LoRa
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# Phi2-Seq-classification-LoRa
This model is a fine-tuned version of [microsoft/phi-2](https://huggingface.co/microsoft/phi-2) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 2.5015
- Precision: 0.9783
- Recall: 0.8182
- F1-score: 0.8911
- Accuracy: 0.9236
## 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: 0.0002
- train_batch_size: 1
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 4
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant
- lr_scheduler_warmup_ratio: 0.03
- training_steps: 3000
### Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1-score | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:--------:|:--------:|
| 0.3914 | 0.7 | 100 | 0.5650 | 0.8571 | 0.6545 | 0.7423 | 0.8264 |
| 0.0065 | 1.39 | 200 | 0.8945 | 0.9375 | 0.8182 | 0.8738 | 0.9097 |
| 0.2111 | 2.09 | 300 | 0.6418 | 0.9388 | 0.8364 | 0.8846 | 0.9167 |
| 0.1823 | 2.78 | 400 | 0.5655 | 0.9773 | 0.7818 | 0.8687 | 0.9097 |
| 0.145 | 3.48 | 500 | 0.9232 | 0.9545 | 0.7636 | 0.8485 | 0.8958 |
| 0.1095 | 4.17 | 600 | 0.8011 | 0.9583 | 0.8364 | 0.8932 | 0.9236 |
| 0.0 | 4.87 | 700 | 0.9774 | 0.9796 | 0.8727 | 0.9231 | 0.9444 |
| 0.4204 | 5.57 | 800 | 0.6787 | 0.96 | 0.8727 | 0.9143 | 0.9375 |
| 0.0068 | 6.26 | 900 | 1.1916 | 0.9783 | 0.8182 | 0.8911 | 0.9236 |
| 0.0004 | 6.96 | 1000 | 0.9468 | 0.9783 | 0.8182 | 0.8911 | 0.9236 |
| 0.0 | 7.65 | 1100 | 1.2818 | 0.9787 | 0.8364 | 0.9020 | 0.9306 |
| 0.0 | 8.35 | 1200 | 1.2332 | 0.9787 | 0.8364 | 0.9020 | 0.9306 |
| 0.0 | 9.04 | 1300 | 1.3078 | 0.9583 | 0.8364 | 0.8932 | 0.9236 |
| 0.0 | 9.74 | 1400 | 1.9713 | 0.9773 | 0.7818 | 0.8687 | 0.9097 |
| 0.0001 | 10.43 | 1500 | 1.5254 | 0.9773 | 0.7818 | 0.8687 | 0.9097 |
| 0.0 | 11.13 | 1600 | 1.8045 | 0.9592 | 0.8545 | 0.9038 | 0.9306 |
| 0.0011 | 11.83 | 1700 | 2.7594 | 0.9574 | 0.8182 | 0.8824 | 0.9167 |
| 0.0 | 12.52 | 1800 | 1.9378 | 0.9592 | 0.8545 | 0.9038 | 0.9306 |
| 0.0 | 13.22 | 1900 | 1.8710 | 0.9592 | 0.8545 | 0.9038 | 0.9306 |
| 0.0 | 13.91 | 2000 | 1.8685 | 0.9592 | 0.8545 | 0.9038 | 0.9306 |
| 0.0 | 14.61 | 2100 | 1.8683 | 0.9592 | 0.8545 | 0.9038 | 0.9306 |
| 0.0384 | 15.3 | 2200 | 2.3028 | 0.9592 | 0.8545 | 0.9038 | 0.9306 |
| 0.001 | 16.0 | 2300 | 1.7787 | 0.9216 | 0.8545 | 0.8868 | 0.9167 |
| 0.0 | 16.7 | 2400 | 1.8615 | 0.9787 | 0.8364 | 0.9020 | 0.9306 |
| 0.8746 | 17.39 | 2500 | 2.6710 | 0.9756 | 0.7273 | 0.8333 | 0.8889 |
| 0.0 | 18.09 | 2600 | 1.0475 | 0.9412 | 0.8727 | 0.9057 | 0.9306 |
| 0.0 | 18.78 | 2700 | 2.3325 | 0.9778 | 0.8 | 0.8800 | 0.9167 |
| 0.0001 | 19.48 | 2800 | 1.3658 | 0.94 | 0.8545 | 0.8952 | 0.9236 |
| 0.0 | 20.17 | 2900 | 2.4668 | 0.9783 | 0.8182 | 0.8911 | 0.9236 |
| 0.0 | 20.87 | 3000 | 2.5015 | 0.9783 | 0.8182 | 0.8911 | 0.9236 |
### Framework versions
- PEFT 0.7.1
- Transformers 4.36.2
- Pytorch 2.1.0+cu121
- Datasets 2.16.1
- Tokenizers 0.15.0 |