Andreas Jörg commited on
Commit ·
2d7b712
1
Parent(s): f7dcac5
update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,77 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
model-index:
|
| 6 |
+
- name: kant-gpt2
|
| 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 |
+
# kant-gpt2
|
| 14 |
+
|
| 15 |
+
This model is a fine-tuned version of [dbmdz/german-gpt2](https://huggingface.co/dbmdz/german-gpt2) on an unknown dataset.
|
| 16 |
+
It achieves the following results on the evaluation set:
|
| 17 |
+
- Loss: 1.8022
|
| 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: 5e-05
|
| 37 |
+
- train_batch_size: 16
|
| 38 |
+
- eval_batch_size: 16
|
| 39 |
+
- seed: 42
|
| 40 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 41 |
+
- lr_scheduler_type: linear
|
| 42 |
+
- lr_scheduler_warmup_steps: 500
|
| 43 |
+
- num_epochs: 22
|
| 44 |
+
|
| 45 |
+
### Training results
|
| 46 |
+
|
| 47 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 48 |
+
|:-------------:|:-----:|:-----:|:---------------:|
|
| 49 |
+
| 3.3257 | 1.0 | 1825 | 3.2231 |
|
| 50 |
+
| 2.9885 | 2.0 | 3650 | 3.0069 |
|
| 51 |
+
| 2.7955 | 3.0 | 5475 | 2.8440 |
|
| 52 |
+
| 2.5748 | 4.0 | 7300 | 2.7059 |
|
| 53 |
+
| 2.3545 | 5.0 | 9125 | 2.5806 |
|
| 54 |
+
| 2.1759 | 6.0 | 10950 | 2.4618 |
|
| 55 |
+
| 1.9697 | 7.0 | 12775 | 2.3553 |
|
| 56 |
+
| 1.7778 | 8.0 | 14600 | 2.2517 |
|
| 57 |
+
| 1.6192 | 9.0 | 16425 | 2.1599 |
|
| 58 |
+
| 1.4675 | 10.0 | 18250 | 2.0895 |
|
| 59 |
+
| 1.3195 | 11.0 | 20075 | 2.0138 |
|
| 60 |
+
| 1.2012 | 12.0 | 21900 | 1.9602 |
|
| 61 |
+
| 1.0828 | 13.0 | 23725 | 1.9097 |
|
| 62 |
+
| 0.9926 | 14.0 | 25550 | 1.8720 |
|
| 63 |
+
| 0.9076 | 15.0 | 27375 | 1.8426 |
|
| 64 |
+
| 0.8336 | 16.0 | 29200 | 1.8214 |
|
| 65 |
+
| 0.7649 | 17.0 | 31025 | 1.8058 |
|
| 66 |
+
| 0.7208 | 18.0 | 32850 | 1.7980 |
|
| 67 |
+
| 0.6798 | 19.0 | 34675 | 1.7938 |
|
| 68 |
+
| 0.647 | 20.0 | 36500 | 1.7969 |
|
| 69 |
+
| 0.6226 | 21.0 | 38325 | 1.7975 |
|
| 70 |
+
| 0.601 | 22.0 | 40150 | 1.8022 |
|
| 71 |
+
|
| 72 |
+
|
| 73 |
+
### Framework versions
|
| 74 |
+
|
| 75 |
+
- Transformers 4.19.2
|
| 76 |
+
- Pytorch 1.11.0+cu113
|
| 77 |
+
- Tokenizers 0.12.1
|