GraphCodeBERT vulnerability classifier (CPU-trained, 20K balanced subset)

README.md

@@ -0,0 +1,102 @@
+---
+library_name: transformers
+base_model: huggingface/CodeBERTa-small-v1
+tags:
+- security
+- vulnerability-detection
+- code-analysis
+- multi-label-classification
+- graphcodebert
+- owasp
+- cwe
+- generated_from_trainer
+model-index:
+- name: graphcodebert-vuln-classifier
+  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. -->
+
+# graphcodebert-vuln-classifier
+
+This model is a fine-tuned version of [huggingface/CodeBERTa-small-v1](https://huggingface.co/huggingface/CodeBERTa-small-v1) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.1884
+- Macro F1: 0.1157
+- Micro F1: 0.7043
+- Weighted F1: 0.8779
+- Macro Precision: 0.0871
+- Macro Recall: 0.2759
+- F1 Safe: 0.9464
+- F1 Cwe-20: 0.0312
+- F1 Cwe-22: 0.0
+- F1 Cwe-78: 0.0
+- F1 Cwe-79: 0.0
+- F1 Cwe-89: 0.6
+- F1 Cwe-94: 0.4348
+- F1 Cwe-119: 0.1290
+- F1 Cwe-125: 0.1333
+- F1 Cwe-190: 0.4
+- F1 Cwe-200: 0.0
+- F1 Cwe-264: 0.0
+- F1 Cwe-269: 0.0
+- F1 Cwe-276: 0.0
+- F1 Cwe-284: 0.0
+- F1 Cwe-287: 0.0
+- F1 Cwe-310: 0.0
+- F1 Cwe-327: 0.0
+- F1 Cwe-330: 0.0
+- F1 Cwe-352: 0.0
+- F1 Cwe-362: 0.0
+- F1 Cwe-399: 0.1818
+- F1 Cwe-401: 0.0
+- F1 Cwe-416: 0.0
+- F1 Cwe-434: 0.0
+- F1 Cwe-476: 0.2105
+- F1 Cwe-502: 0.2857
+- F1 Cwe-601: 0.0
+- F1 Cwe-787: 0.2326
+- F1 Cwe-798: 0.0
+- F1 Cwe-918: 0.0
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+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: 8
+- eval_batch_size: 16
+- seed: 42
+- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_steps: 50
+- num_epochs: 2
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Macro F1 | Micro F1 | Weighted F1 | Macro Precision | Macro Recall | F1 Safe | F1 Cwe-20 | F1 Cwe-22 | F1 Cwe-78 | F1 Cwe-79 | F1 Cwe-89 | F1 Cwe-94 | F1 Cwe-119 | F1 Cwe-125 | F1 Cwe-190 | F1 Cwe-200 | F1 Cwe-264 | F1 Cwe-269 | F1 Cwe-276 | F1 Cwe-284 | F1 Cwe-287 | F1 Cwe-310 | F1 Cwe-327 | F1 Cwe-330 | F1 Cwe-352 | F1 Cwe-362 | F1 Cwe-399 | F1 Cwe-401 | F1 Cwe-416 | F1 Cwe-434 | F1 Cwe-476 | F1 Cwe-502 | F1 Cwe-601 | F1 Cwe-787 | F1 Cwe-798 | F1 Cwe-918 |
+|:-------------:|:-----:|:----:|:---------------:|:--------:|:--------:|:-----------:|:---------------:|:------------:|:-------:|:---------:|:---------:|:---------:|:---------:|:---------:|:---------:|:----------:|:----------:|:----------:|:----------:|:----------:|:----------:|:----------:|:----------:|:----------:|:----------:|:----------:|:----------:|:----------:|:----------:|:----------:|:----------:|:----------:|:----------:|:----------:|:----------:|:----------:|:----------:|:----------:|:----------:|
+| 0.4733        | 1.0   | 250  | 0.2105          | 0.0955   | 0.7880   | 0.8794      | 0.0730          | 0.2053       | 0.9539  | 0.05      | 0.0       | 0.0       | 0.0       | 0.5455    | 0.3704    | 0.08       | 0.2        | 0.0        | 0.0        | 0.0        | 0.0        | 0.0        | 0.0        | 0.0        | 0.0        | 0.0        | 0.0        | 0.0        | 0.0        | 0.0        | 0.0        | 0.0        | 0.0        | 0.2857     | 0.1818     | 0.0        | 0.2941     | 0.0        | 0.0        |
+| 0.4130        | 2.0   | 500  | 0.1884          | 0.1157   | 0.7043   | 0.8779      | 0.0871          | 0.2759       | 0.9464  | 0.0312    | 0.0       | 0.0       | 0.0       | 0.6       | 0.4348    | 0.1290     | 0.1333     | 0.4        | 0.0        | 0.0        | 0.0        | 0.0        | 0.0        | 0.0        | 0.0        | 0.0        | 0.0        | 0.0        | 0.0        | 0.1818     | 0.0        | 0.0        | 0.0        | 0.2105     | 0.2857     | 0.0        | 0.2326     | 0.0        | 0.0        |
+
+
+### Framework versions
+
+- Transformers 5.6.1
+- Pytorch 2.11.0+cu130
+- Datasets 4.8.4
+- Tokenizers 0.22.2

config.json

@@ -0,0 +1,95 @@
+{
+  "add_cross_attention": false,
+  "architectures": [
+    "RobertaForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "dtype": "float32",
+  "eos_token_id": 2,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2",
+    "3": "LABEL_3",
+    "4": "LABEL_4",
+    "5": "LABEL_5",
+    "6": "LABEL_6",
+    "7": "LABEL_7",
+    "8": "LABEL_8",
+    "9": "LABEL_9",
+    "10": "LABEL_10",
+    "11": "LABEL_11",
+    "12": "LABEL_12",
+    "13": "LABEL_13",
+    "14": "LABEL_14",
+    "15": "LABEL_15",
+    "16": "LABEL_16",
+    "17": "LABEL_17",
+    "18": "LABEL_18",
+    "19": "LABEL_19",
+    "20": "LABEL_20",
+    "21": "LABEL_21",
+    "22": "LABEL_22",
+    "23": "LABEL_23",
+    "24": "LABEL_24",
+    "25": "LABEL_25",
+    "26": "LABEL_26",
+    "27": "LABEL_27",
+    "28": "LABEL_28",
+    "29": "LABEL_29",
+    "30": "LABEL_30"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "is_decoder": false,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_10": 10,
+    "LABEL_11": 11,
+    "LABEL_12": 12,
+    "LABEL_13": 13,
+    "LABEL_14": 14,
+    "LABEL_15": 15,
+    "LABEL_16": 16,
+    "LABEL_17": 17,
+    "LABEL_18": 18,
+    "LABEL_19": 19,
+    "LABEL_2": 2,
+    "LABEL_20": 20,
+    "LABEL_21": 21,
+    "LABEL_22": 22,
+    "LABEL_23": 23,
+    "LABEL_24": 24,
+    "LABEL_25": 25,
+    "LABEL_26": 26,
+    "LABEL_27": 27,
+    "LABEL_28": 28,
+    "LABEL_29": 29,
+    "LABEL_3": 3,
+    "LABEL_30": 30,
+    "LABEL_4": 4,
+    "LABEL_5": 5,
+    "LABEL_6": 6,
+    "LABEL_7": 7,
+    "LABEL_8": 8,
+    "LABEL_9": 9
+  },
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "roberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 6,
+  "pad_token_id": 1,
+  "problem_type": "multi_label_classification",
+  "tie_word_embeddings": true,
+  "transformers_version": "5.6.1",
+  "type_vocab_size": 1,
+  "use_cache": false,
+  "vocab_size": 52000
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7c2863b59f345fa82b44d21d25483aa4ddd11665213b069e302f9a815985add4
+size 333911404

tokenizer_config.json

@@ -0,0 +1,18 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "errors": "replace",
+  "is_local": false,
+  "local_files_only": false,
+  "mask_token": "<mask>",
+  "max_len": 512,
+  "model_max_length": 512,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "tokenizer_class": "RobertaTokenizer",
+  "trim_offsets": true,
+  "unk_token": "<unk>"
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:faf024c35995c1b5bd61eaa1ad4ddbbfca5175a19eb3f2e13c5cb7779bdbc19c
+size 5329