update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,71 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
model-index:
|
| 6 |
+
- name: distilgpt2-magicprompt
|
| 7 |
+
results: []
|
| 8 |
+
---
|
| 9 |
+
|
| 10 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 11 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 12 |
+
|
| 13 |
+
# distilgpt2-magicprompt
|
| 14 |
+
|
| 15 |
+
This model is a fine-tuned version of [distilgpt2](https://huggingface.co/distilgpt2) on an unknown dataset.
|
| 16 |
+
It achieves the following results on the evaluation set:
|
| 17 |
+
- Loss: 1.3089
|
| 18 |
+
|
| 19 |
+
## Model description
|
| 20 |
+
|
| 21 |
+
More information needed
|
| 22 |
+
|
| 23 |
+
## Intended uses & limitations
|
| 24 |
+
|
| 25 |
+
More information needed
|
| 26 |
+
|
| 27 |
+
## Training and evaluation data
|
| 28 |
+
|
| 29 |
+
More information needed
|
| 30 |
+
|
| 31 |
+
## Training procedure
|
| 32 |
+
|
| 33 |
+
### Training hyperparameters
|
| 34 |
+
|
| 35 |
+
The following hyperparameters were used during training:
|
| 36 |
+
- learning_rate: 0.001
|
| 37 |
+
- train_batch_size: 16
|
| 38 |
+
- eval_batch_size: 2
|
| 39 |
+
- seed: 42
|
| 40 |
+
- distributed_type: multi-GPU
|
| 41 |
+
- num_devices: 2
|
| 42 |
+
- gradient_accumulation_steps: 8
|
| 43 |
+
- total_train_batch_size: 256
|
| 44 |
+
- total_eval_batch_size: 4
|
| 45 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 46 |
+
- lr_scheduler_type: cosine
|
| 47 |
+
- lr_scheduler_warmup_ratio: 0.05
|
| 48 |
+
- num_epochs: 10.0
|
| 49 |
+
|
| 50 |
+
### Training results
|
| 51 |
+
|
| 52 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 53 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 54 |
+
| 2.7061 | 0.99 | 33 | 2.5859 |
|
| 55 |
+
| 2.08 | 1.99 | 66 | 1.9965 |
|
| 56 |
+
| 1.7623 | 2.99 | 99 | 1.7248 |
|
| 57 |
+
| 1.5408 | 3.99 | 132 | 1.5449 |
|
| 58 |
+
| 1.4147 | 4.99 | 165 | 1.4437 |
|
| 59 |
+
| 1.3593 | 5.99 | 198 | 1.3768 |
|
| 60 |
+
| 1.2703 | 6.99 | 231 | 1.3362 |
|
| 61 |
+
| 1.2528 | 7.99 | 264 | 1.3175 |
|
| 62 |
+
| 1.1981 | 8.99 | 297 | 1.3091 |
|
| 63 |
+
| 1.2117 | 9.99 | 330 | 1.3089 |
|
| 64 |
+
|
| 65 |
+
|
| 66 |
+
### Framework versions
|
| 67 |
+
|
| 68 |
+
- Transformers 4.25.0.dev0
|
| 69 |
+
- Pytorch 1.13.0+cu117
|
| 70 |
+
- Datasets 2.6.1
|
| 71 |
+
- Tokenizers 0.13.1
|