File size: 3,574 Bytes
bc81432 23d31af bc81432 23d31af bc81432 f2c556c f5ea9eb f2c556c f5ea9eb f2c556c bc81432 23d31af bc81432 23d31af bc81432 23d31af bc81432 23d31af |
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 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 |
---
base_model: microsoft/codebert-base
tags:
- generated_from_trainer
model-index:
- name: CodeBertForCodeTrans
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. -->
# CodeBertForCodeTrans
This model is a fine-tuned version of [microsoft/codebert-base](https://huggingface.co/microsoft/codebert-base) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.0006
## Model description
More information needed
## Driectly uses
```Python
from transformers import AutoTokenizer, AutoModelForCausalLM
additional_special_tokens = {'additional_special_tokens':['<|begin_of_java_code|>','<|end_of_java_code|>'\
,'<|begin_of_c-sharp_code|>','<|end_of_c-sharp_code|>',\
'<|translate|>']}
basemodel = "ljcnju/CodeBertForCodeTrans"
tokenizer = AutoTokenizer.from_pretrained(basemodel)
tokenizer.pad_token = tokenizer.eos_token
config = AutoConfig.from_pretrained(basemodel)
config.is_decoder = True
model = AutoModelForCausalLM.from_pretrained(basemodel,config=config)
device = torch.device("cuda") if torch.cuda.is_available() else torch.device('cpu')
model.to(device)
code = "public void serialize(LittleEndianOutput out) {out.writeShort(field_1_vcenter);}\n"
prefix = additional_special_tokens['additional_special_tokens'][0]
input_str = prefix + code +additional_special_tokens['additional_special_tokens'][1] + additional_special_tokens['additional_special_tokens'][2]
input = tokenizer(input_str,return_tensors = "pt")
output = model.generate(**input, max_length = 256)
outputs_str = tokenizer.decode(output[0])
print(outputs_str)
```
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 12354.0
- num_epochs: 20
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:-----:|:---------------:|
| 5.7169 | 1.0 | 644 | 4.5075 |
| 3.0571 | 2.0 | 1288 | 2.1423 |
| 0.7391 | 3.0 | 1932 | 0.2866 |
| 0.1028 | 4.0 | 2576 | 0.0219 |
| 0.0158 | 5.0 | 3220 | 0.0047 |
| 0.0065 | 6.0 | 3864 | 0.0024 |
| 0.0036 | 7.0 | 4508 | 0.0020 |
| 0.0028 | 8.0 | 5152 | 0.0014 |
| 0.0018 | 9.0 | 5796 | 0.0010 |
| 0.0023 | 10.0 | 6440 | 0.0017 |
| 0.002 | 11.0 | 7084 | 0.0009 |
| 0.002 | 12.0 | 7728 | 0.0012 |
| 0.0015 | 13.0 | 8372 | 0.0020 |
| 0.0028 | 14.0 | 9016 | 0.0010 |
| 0.0015 | 15.0 | 9660 | 0.0007 |
| 0.0027 | 16.0 | 10304 | 0.0015 |
| 0.002 | 17.0 | 10948 | 0.0007 |
| 0.0011 | 18.0 | 11592 | 0.0009 |
| 0.0019 | 19.0 | 12236 | 0.0007 |
| 0.0003 | 20.0 | 12880 | 0.0006 |
### Framework versions
- Transformers 4.37.2
- Pytorch 2.1.2+cu121
- Datasets 2.15.0
- Tokenizers 0.15.0
|