Upload detecode model v1
Browse files- README.md +32 -1
- config.json +58 -0
- model.safetensors +3 -0
- tokenizer.json +0 -0
- tokenizer_config.json +17 -0
- training_args.bin +3 -0
README.md
CHANGED
|
@@ -1,3 +1,34 @@
|
|
| 1 |
-
---
|
| 2 |
license: mit
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
license: mit
|
| 3 |
+
library_name: transformers
|
| 4 |
+
tags:
|
| 5 |
+
- codebert
|
| 6 |
+
- vulnerability-detection
|
| 7 |
+
- php
|
| 8 |
+
- javascript
|
| 9 |
+
- detecode
|
| 10 |
---
|
| 11 |
+
|
| 12 |
+
# DeteCode Model v1
|
| 13 |
+
|
| 14 |
+
Fine-tuned CodeBERT checkpoint for the optional AI engine in [DeteCode](https://github.com/rizkidsaputra/detecode).
|
| 15 |
+
|
| 16 |
+
This model is experimental and should be used as an assistive semantic layer, not as the only source of truth. DeteCode's local rules and taint analysis are recommended for stable CLI scanning.
|
| 17 |
+
|
| 18 |
+
## Usage
|
| 19 |
+
|
| 20 |
+
Download or clone this repository into:
|
| 21 |
+
|
| 22 |
+
```text
|
| 23 |
+
models/codebert-webvuln
|
| 24 |
+
```
|
| 25 |
+
|
| 26 |
+
Then run:
|
| 27 |
+
|
| 28 |
+
```powershell
|
| 29 |
+
python -m detecode scan .\tests\samples --engine hybrid --model-path .\models\codebert-webvuln --format table
|
| 30 |
+
```
|
| 31 |
+
|
| 32 |
+
## Training Data
|
| 33 |
+
|
| 34 |
+
The model was fine-tuned using the DeteCode training script with PHP/JavaScript vulnerability data, primarily CrossVul (`hitoshura25/crossvul`).
|
config.json
ADDED
|
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_cross_attention": false,
|
| 3 |
+
"architectures": [
|
| 4 |
+
"RobertaForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"bos_token_id": 0,
|
| 8 |
+
"classifier_dropout": null,
|
| 9 |
+
"dtype": "float32",
|
| 10 |
+
"eos_token_id": 2,
|
| 11 |
+
"hidden_act": "gelu",
|
| 12 |
+
"hidden_dropout_prob": 0.1,
|
| 13 |
+
"hidden_size": 768,
|
| 14 |
+
"id2label": {
|
| 15 |
+
"0": "safe",
|
| 16 |
+
"1": "CWE-200",
|
| 17 |
+
"2": "CWE-22",
|
| 18 |
+
"3": "CWE-352",
|
| 19 |
+
"4": "CWE-434",
|
| 20 |
+
"5": "CWE-502",
|
| 21 |
+
"6": "CWE-78",
|
| 22 |
+
"7": "CWE-79",
|
| 23 |
+
"8": "CWE-89",
|
| 24 |
+
"9": "CWE-918",
|
| 25 |
+
"10": "CWE-94",
|
| 26 |
+
"11": "CWE-98"
|
| 27 |
+
},
|
| 28 |
+
"initializer_range": 0.02,
|
| 29 |
+
"intermediate_size": 3072,
|
| 30 |
+
"is_decoder": false,
|
| 31 |
+
"label2id": {
|
| 32 |
+
"CWE-200": 1,
|
| 33 |
+
"CWE-22": 2,
|
| 34 |
+
"CWE-352": 3,
|
| 35 |
+
"CWE-434": 4,
|
| 36 |
+
"CWE-502": 5,
|
| 37 |
+
"CWE-78": 6,
|
| 38 |
+
"CWE-79": 7,
|
| 39 |
+
"CWE-89": 8,
|
| 40 |
+
"CWE-918": 9,
|
| 41 |
+
"CWE-94": 10,
|
| 42 |
+
"CWE-98": 11,
|
| 43 |
+
"safe": 0
|
| 44 |
+
},
|
| 45 |
+
"layer_norm_eps": 1e-05,
|
| 46 |
+
"max_position_embeddings": 514,
|
| 47 |
+
"model_type": "roberta",
|
| 48 |
+
"num_attention_heads": 12,
|
| 49 |
+
"num_hidden_layers": 12,
|
| 50 |
+
"output_past": true,
|
| 51 |
+
"pad_token_id": 1,
|
| 52 |
+
"problem_type": "single_label_classification",
|
| 53 |
+
"tie_word_embeddings": true,
|
| 54 |
+
"transformers_version": "5.9.0",
|
| 55 |
+
"type_vocab_size": 1,
|
| 56 |
+
"use_cache": false,
|
| 57 |
+
"vocab_size": 50265
|
| 58 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:064f2a5b2599baf25b1e0da0f61bb2697bb817a86eefcf4cdc6da73e1e50e2ee
|
| 3 |
+
size 498643584
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,17 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"backend": "tokenizers",
|
| 4 |
+
"bos_token": "<s>",
|
| 5 |
+
"cls_token": "<s>",
|
| 6 |
+
"eos_token": "</s>",
|
| 7 |
+
"errors": "replace",
|
| 8 |
+
"is_local": false,
|
| 9 |
+
"local_files_only": false,
|
| 10 |
+
"mask_token": "<mask>",
|
| 11 |
+
"model_max_length": 512,
|
| 12 |
+
"pad_token": "<pad>",
|
| 13 |
+
"sep_token": "</s>",
|
| 14 |
+
"tokenizer_class": "RobertaTokenizer",
|
| 15 |
+
"trim_offsets": true,
|
| 16 |
+
"unk_token": "<unk>"
|
| 17 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d29f62b2668fb205e8ad29350c893addf69b6eeb3733f1bccbf2cc42687e7b34
|
| 3 |
+
size 5265
|