update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,82 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
metrics:
|
| 6 |
+
- bleu
|
| 7 |
+
model-index:
|
| 8 |
+
- name: h1
|
| 9 |
+
results: []
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 13 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 14 |
+
|
| 15 |
+
# h1
|
| 16 |
+
|
| 17 |
+
This model is a fine-tuned version of [distilgpt2](https://huggingface.co/distilgpt2) on an unknown dataset.
|
| 18 |
+
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 0.0890
|
| 20 |
+
- Exact Match: 0.1970
|
| 21 |
+
- Bleu: 0.9737
|
| 22 |
+
- Codebleu: 0.9172
|
| 23 |
+
- Ngram Match Score: 0.8984
|
| 24 |
+
- Weighted Ngram Match Score: 0.8985
|
| 25 |
+
- Syntax Match Score: 0.9293
|
| 26 |
+
- Dataflow Match Score: 0.9429
|
| 27 |
+
- Chrf: 97.5313
|
| 28 |
+
|
| 29 |
+
## Model description
|
| 30 |
+
|
| 31 |
+
More information needed
|
| 32 |
+
|
| 33 |
+
## Intended uses & limitations
|
| 34 |
+
|
| 35 |
+
More information needed
|
| 36 |
+
|
| 37 |
+
## Training and evaluation data
|
| 38 |
+
|
| 39 |
+
More information needed
|
| 40 |
+
|
| 41 |
+
## Training procedure
|
| 42 |
+
|
| 43 |
+
### Training hyperparameters
|
| 44 |
+
|
| 45 |
+
The following hyperparameters were used during training:
|
| 46 |
+
- learning_rate: 2e-05
|
| 47 |
+
- train_batch_size: 4
|
| 48 |
+
- eval_batch_size: 4
|
| 49 |
+
- seed: 17
|
| 50 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 51 |
+
- lr_scheduler_type: cosine
|
| 52 |
+
- num_epochs: 200
|
| 53 |
+
- mixed_precision_training: Native AMP
|
| 54 |
+
|
| 55 |
+
### Training results
|
| 56 |
+
|
| 57 |
+
| Training Loss | Epoch | Step | Validation Loss | Exact Match | Bleu | Codebleu | Ngram Match Score | Weighted Ngram Match Score | Syntax Match Score | Dataflow Match Score | Chrf |
|
| 58 |
+
|:-------------:|:------:|:-----:|:---------------:|:-----------:|:------:|:--------:|:-----------------:|:--------------------------:|:------------------:|:--------------------:|:-------:|
|
| 59 |
+
| 0.3871 | 11.94 | 1600 | 0.1043 | 0.0152 | 0.9499 | 0.8549 | 0.8089 | 0.8089 | 0.8653 | 0.9366 | 95.4674 |
|
| 60 |
+
| 0.0752 | 23.88 | 3200 | 0.0784 | 0.1212 | 0.9640 | 0.8874 | 0.8525 | 0.8526 | 0.8929 | 0.9516 | 96.7978 |
|
| 61 |
+
| 0.0448 | 35.82 | 4800 | 0.0717 | 0.1364 | 0.9693 | 0.9077 | 0.8782 | 0.8782 | 0.9069 | 0.9674 | 97.2100 |
|
| 62 |
+
| 0.0308 | 47.76 | 6400 | 0.0752 | 0.1364 | 0.9702 | 0.9061 | 0.8808 | 0.8810 | 0.9070 | 0.9554 | 97.1896 |
|
| 63 |
+
| 0.0223 | 59.7 | 8000 | 0.0762 | 0.1364 | 0.9724 | 0.9050 | 0.8877 | 0.8881 | 0.9093 | 0.9348 | 97.4616 |
|
| 64 |
+
| 0.0166 | 71.64 | 9600 | 0.0762 | 0.1667 | 0.9733 | 0.9140 | 0.8948 | 0.8951 | 0.9197 | 0.9461 | 97.4945 |
|
| 65 |
+
| 0.0128 | 83.58 | 11200 | 0.0793 | 0.1515 | 0.9728 | 0.9085 | 0.8911 | 0.8918 | 0.9189 | 0.9321 | 97.4152 |
|
| 66 |
+
| 0.0104 | 95.52 | 12800 | 0.0822 | 0.1667 | 0.9732 | 0.9165 | 0.8946 | 0.8950 | 0.9222 | 0.9541 | 97.4887 |
|
| 67 |
+
| 0.0084 | 107.46 | 14400 | 0.0832 | 0.1667 | 0.9737 | 0.9167 | 0.8970 | 0.8972 | 0.9254 | 0.9471 | 97.5326 |
|
| 68 |
+
| 0.007 | 119.4 | 16000 | 0.0837 | 0.1818 | 0.9743 | 0.9160 | 0.8983 | 0.8986 | 0.9238 | 0.9434 | 97.6638 |
|
| 69 |
+
| 0.0058 | 131.34 | 17600 | 0.0858 | 0.1818 | 0.9739 | 0.9200 | 0.8977 | 0.8977 | 0.9267 | 0.9579 | 97.5583 |
|
| 70 |
+
| 0.005 | 143.28 | 19200 | 0.0878 | 0.1818 | 0.9743 | 0.9180 | 0.8993 | 0.9001 | 0.9301 | 0.9426 | 97.5819 |
|
| 71 |
+
| 0.0044 | 155.22 | 20800 | 0.0877 | 0.1667 | 0.9736 | 0.9156 | 0.8957 | 0.8960 | 0.9278 | 0.9429 | 97.5109 |
|
| 72 |
+
| 0.0042 | 167.16 | 22400 | 0.0890 | 0.1970 | 0.9736 | 0.9171 | 0.8984 | 0.8984 | 0.9293 | 0.9424 | 97.5617 |
|
| 73 |
+
| 0.0038 | 179.1 | 24000 | 0.0891 | 0.2121 | 0.9738 | 0.9174 | 0.8991 | 0.8991 | 0.9285 | 0.9429 | 97.5452 |
|
| 74 |
+
| 0.0037 | 191.04 | 25600 | 0.0890 | 0.1970 | 0.9737 | 0.9172 | 0.8984 | 0.8985 | 0.9293 | 0.9429 | 97.5313 |
|
| 75 |
+
|
| 76 |
+
|
| 77 |
+
### Framework versions
|
| 78 |
+
|
| 79 |
+
- Transformers 4.24.0
|
| 80 |
+
- Pytorch 1.13.0
|
| 81 |
+
- Datasets 2.6.1
|
| 82 |
+
- Tokenizers 0.13.1
|