File size: 4,725 Bytes
73e7c3e |
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 |
---
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- bleu
model-index:
- name: h1
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. -->
# h1
This model is a fine-tuned version of [distilgpt2](https://huggingface.co/distilgpt2) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.0890
- Exact Match: 0.1970
- Bleu: 0.9737
- Codebleu: 0.9172
- Ngram Match Score: 0.8984
- Weighted Ngram Match Score: 0.8985
- Syntax Match Score: 0.9293
- Dataflow Match Score: 0.9429
- Chrf: 97.5313
## 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: 2e-05
- train_batch_size: 4
- eval_batch_size: 4
- seed: 17
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- num_epochs: 200
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Exact Match | Bleu | Codebleu | Ngram Match Score | Weighted Ngram Match Score | Syntax Match Score | Dataflow Match Score | Chrf |
|:-------------:|:------:|:-----:|:---------------:|:-----------:|:------:|:--------:|:-----------------:|:--------------------------:|:------------------:|:--------------------:|:-------:|
| 0.3871 | 11.94 | 1600 | 0.1043 | 0.0152 | 0.9499 | 0.8549 | 0.8089 | 0.8089 | 0.8653 | 0.9366 | 95.4674 |
| 0.0752 | 23.88 | 3200 | 0.0784 | 0.1212 | 0.9640 | 0.8874 | 0.8525 | 0.8526 | 0.8929 | 0.9516 | 96.7978 |
| 0.0448 | 35.82 | 4800 | 0.0717 | 0.1364 | 0.9693 | 0.9077 | 0.8782 | 0.8782 | 0.9069 | 0.9674 | 97.2100 |
| 0.0308 | 47.76 | 6400 | 0.0752 | 0.1364 | 0.9702 | 0.9061 | 0.8808 | 0.8810 | 0.9070 | 0.9554 | 97.1896 |
| 0.0223 | 59.7 | 8000 | 0.0762 | 0.1364 | 0.9724 | 0.9050 | 0.8877 | 0.8881 | 0.9093 | 0.9348 | 97.4616 |
| 0.0166 | 71.64 | 9600 | 0.0762 | 0.1667 | 0.9733 | 0.9140 | 0.8948 | 0.8951 | 0.9197 | 0.9461 | 97.4945 |
| 0.0128 | 83.58 | 11200 | 0.0793 | 0.1515 | 0.9728 | 0.9085 | 0.8911 | 0.8918 | 0.9189 | 0.9321 | 97.4152 |
| 0.0104 | 95.52 | 12800 | 0.0822 | 0.1667 | 0.9732 | 0.9165 | 0.8946 | 0.8950 | 0.9222 | 0.9541 | 97.4887 |
| 0.0084 | 107.46 | 14400 | 0.0832 | 0.1667 | 0.9737 | 0.9167 | 0.8970 | 0.8972 | 0.9254 | 0.9471 | 97.5326 |
| 0.007 | 119.4 | 16000 | 0.0837 | 0.1818 | 0.9743 | 0.9160 | 0.8983 | 0.8986 | 0.9238 | 0.9434 | 97.6638 |
| 0.0058 | 131.34 | 17600 | 0.0858 | 0.1818 | 0.9739 | 0.9200 | 0.8977 | 0.8977 | 0.9267 | 0.9579 | 97.5583 |
| 0.005 | 143.28 | 19200 | 0.0878 | 0.1818 | 0.9743 | 0.9180 | 0.8993 | 0.9001 | 0.9301 | 0.9426 | 97.5819 |
| 0.0044 | 155.22 | 20800 | 0.0877 | 0.1667 | 0.9736 | 0.9156 | 0.8957 | 0.8960 | 0.9278 | 0.9429 | 97.5109 |
| 0.0042 | 167.16 | 22400 | 0.0890 | 0.1970 | 0.9736 | 0.9171 | 0.8984 | 0.8984 | 0.9293 | 0.9424 | 97.5617 |
| 0.0038 | 179.1 | 24000 | 0.0891 | 0.2121 | 0.9738 | 0.9174 | 0.8991 | 0.8991 | 0.9285 | 0.9429 | 97.5452 |
| 0.0037 | 191.04 | 25600 | 0.0890 | 0.1970 | 0.9737 | 0.9172 | 0.8984 | 0.8985 | 0.9293 | 0.9429 | 97.5313 |
### Framework versions
- Transformers 4.24.0
- Pytorch 1.13.0
- Datasets 2.6.1
- Tokenizers 0.13.1
|