Update README.md
Browse files
README.md
CHANGED
|
@@ -8,72 +8,6 @@ model-index:
|
|
| 8 |
results: []
|
| 9 |
---
|
| 10 |
|
| 11 |
-
|
| 12 |
-
|
| 13 |
-
|
| 14 |
-
# gpt2coder-8epochs
|
| 15 |
-
|
| 16 |
-
This model is a fine-tuned version of [Aravindan/gpt2out](https://huggingface.co/Aravindan/gpt2out) on an unknown dataset.
|
| 17 |
-
It achieves the following results on the evaluation set:
|
| 18 |
-
- Loss: 0.6964
|
| 19 |
-
|
| 20 |
-
## Model description
|
| 21 |
-
|
| 22 |
-
More information needed
|
| 23 |
-
|
| 24 |
-
## Intended uses & limitations
|
| 25 |
-
|
| 26 |
-
More information needed
|
| 27 |
-
|
| 28 |
-
## Training and evaluation data
|
| 29 |
-
|
| 30 |
-
More information needed
|
| 31 |
-
|
| 32 |
-
## Training procedure
|
| 33 |
-
|
| 34 |
-
### Training hyperparameters
|
| 35 |
-
|
| 36 |
-
The following hyperparameters were used during training:
|
| 37 |
-
- learning_rate: 5e-05
|
| 38 |
-
- train_batch_size: 8
|
| 39 |
-
- eval_batch_size: 8
|
| 40 |
-
- seed: 42
|
| 41 |
-
- gradient_accumulation_steps: 16
|
| 42 |
-
- total_train_batch_size: 128
|
| 43 |
-
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 44 |
-
- lr_scheduler_type: linear
|
| 45 |
-
- lr_scheduler_warmup_ratio: 0.1
|
| 46 |
-
- num_epochs: 20
|
| 47 |
-
|
| 48 |
-
### Training results
|
| 49 |
-
|
| 50 |
-
| Training Loss | Epoch | Step | Validation Loss |
|
| 51 |
-
|:-------------:|:-------:|:----:|:---------------:|
|
| 52 |
-
| No log | 0.9962 | 132 | 1.5428 |
|
| 53 |
-
| No log | 2.0 | 265 | 1.4204 |
|
| 54 |
-
| No log | 2.9962 | 397 | 1.2888 |
|
| 55 |
-
| 1.5725 | 4.0 | 530 | 1.1900 |
|
| 56 |
-
| 1.5725 | 4.9962 | 662 | 1.1045 |
|
| 57 |
-
| 1.5725 | 6.0 | 795 | 1.0314 |
|
| 58 |
-
| 1.5725 | 6.9962 | 927 | 0.9723 |
|
| 59 |
-
| 1.217 | 8.0 | 1060 | 0.9139 |
|
| 60 |
-
| 1.217 | 8.9962 | 1192 | 0.8689 |
|
| 61 |
-
| 1.217 | 10.0 | 1325 | 0.8274 |
|
| 62 |
-
| 1.217 | 10.9962 | 1457 | 0.7910 |
|
| 63 |
-
| 1.0164 | 12.0 | 1590 | 0.7555 |
|
| 64 |
-
| 1.0164 | 12.9962 | 1722 | 0.7266 |
|
| 65 |
-
| 1.0164 | 14.0 | 1855 | 0.7014 |
|
| 66 |
-
| 1.0164 | 14.9962 | 1987 | 0.6777 |
|
| 67 |
-
| 0.8885 | 16.0 | 2120 | 0.6597 |
|
| 68 |
-
| 0.8885 | 16.9962 | 2252 | 0.6440 |
|
| 69 |
-
| 0.8885 | 18.0 | 2385 | 0.6327 |
|
| 70 |
-
| 0.8106 | 18.9962 | 2517 | 0.6239 |
|
| 71 |
-
| 0.8106 | 19.9962 | 2640 | 0.6964 |
|
| 72 |
-
|
| 73 |
-
|
| 74 |
-
### Framework versions
|
| 75 |
-
|
| 76 |
-
- Transformers 4.41.1
|
| 77 |
-
- Pytorch 2.1.2
|
| 78 |
-
- Datasets 2.19.1
|
| 79 |
-
- Tokenizers 0.19.1
|
|
|
|
| 8 |
results: []
|
| 9 |
---
|
| 10 |
|
| 11 |
+
## GPT-OUT
|
| 12 |
+
* It's a coding foundation model trained on various python datasets to bring the model only for python.
|
| 13 |
+
* Pre-training is still on-going process, estimated to complete this process before end of the june month.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|