File size: 12,212 Bytes
146a3a4 | 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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 | {
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# CYB002 Baseline Classifier — Inference Example\n",
"\n",
"End-to-end demo: load the trained XGBoost and PyTorch MLP models from the Hugging Face repo and predict the **MITRE ATT&CK kill-chain phase** of a new attack-event record.\n",
"\n",
"**Models predict one of 10 phases:** `dwell_idle`, `reconnaissance`, `initial_access`, `execution`, `persistence`, `privilege_escalation`, `lateral_movement`, `collection`, `exfiltration`, `impact`.\n",
"\n",
"**This is a baseline reference model**, not a production threat detector. See the model card for full metrics and limitations."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## 1. Install dependencies"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"%pip install --quiet xgboost torch safetensors pandas numpy huggingface_hub"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## 2. Download model artifacts from Hugging Face\n",
"\n",
"Five files are needed:\n",
"- `model_xgb.json` — XGBoost weights\n",
"- `model_mlp.safetensors` — PyTorch MLP weights\n",
"- `feature_engineering.py` — feature pipeline (must match the one used at training)\n",
"- `feature_meta.json` — feature column order + categorical levels\n",
"- `feature_scaler.json` — MLP input standardization (mean / std)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"from huggingface_hub import hf_hub_download\n",
"\n",
"REPO_ID = \"xpertsystems/cyb002-baseline-classifier\"\n",
"\n",
"files = {}\n",
"for name in [\"model_xgb.json\", \"model_mlp.safetensors\",\n",
" \"feature_engineering.py\", \"feature_meta.json\",\n",
" \"feature_scaler.json\"]:\n",
" files[name] = hf_hub_download(repo_id=REPO_ID, filename=name)\n",
" print(f\" downloaded: {name}\")"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"# Make feature_engineering.py importable\n",
"import sys, os\n",
"fe_dir = os.path.dirname(files[\"feature_engineering.py\"])\n",
"if fe_dir not in sys.path:\n",
" sys.path.insert(0, fe_dir)\n",
"\n",
"from feature_engineering import (\n",
" transform_single, load_meta, INT_TO_LABEL, build_segment_lookup\n",
")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## 3. Load models and metadata"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"import json\n",
"import numpy as np\n",
"import torch\n",
"import torch.nn as nn\n",
"import xgboost as xgb\n",
"from safetensors.torch import load_file\n",
"\n",
"meta = load_meta(files[\"feature_meta.json\"])\n",
"with open(files[\"feature_scaler.json\"]) as f:\n",
" scaler = json.load(f)\n",
"\n",
"N_FEATURES = len(meta[\"feature_names\"])\n",
"N_CLASSES = len(meta[\"int_to_label\"])\n",
"print(f\"feature count: {N_FEATURES}\")\n",
"print(f\"class count: {N_CLASSES}\")\n",
"print(f\"label classes: {list(meta['int_to_label'].values())}\")"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"# XGBoost\n",
"xgb_model = xgb.XGBClassifier()\n",
"xgb_model.load_model(files[\"model_xgb.json\"])\n",
"\n",
"# MLP architecture (must match training)\n",
"class PhaseMLP(nn.Module):\n",
" def __init__(self, n_features, n_classes=10, hidden1=128, hidden2=64, dropout=0.3):\n",
" super().__init__()\n",
" self.net = nn.Sequential(\n",
" nn.Linear(n_features, hidden1),\n",
" nn.BatchNorm1d(hidden1),\n",
" nn.ReLU(),\n",
" nn.Dropout(dropout),\n",
" nn.Linear(hidden1, hidden2),\n",
" nn.BatchNorm1d(hidden2),\n",
" nn.ReLU(),\n",
" nn.Dropout(dropout),\n",
" nn.Linear(hidden2, n_classes),\n",
" )\n",
" def forward(self, x):\n",
" return self.net(x)\n",
"\n",
"mlp_model = PhaseMLP(N_FEATURES, n_classes=N_CLASSES)\n",
"mlp_model.load_state_dict(load_file(files[\"model_mlp.safetensors\"]))\n",
"mlp_model.eval()\n",
"print(\"models loaded\")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## 4. Build segment-aggregate lookup from the dataset\n",
"\n",
"Per-segment topology aggregates (mean exposure, fraction with EDR, etc.) are computed at training time and must be available at inference time too. The helper `build_segment_lookup` pulls them from `network_topology.csv`."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"from huggingface_hub import snapshot_download\n",
"\n",
"ds_path = snapshot_download(repo_id=\"xpertsystems/cyb002-sample\", repo_type=\"dataset\")\n",
"\n",
"import os\n",
"segment_aggregates_lookup = build_segment_lookup(\n",
" os.path.join(ds_path, \"network_topology.csv\")\n",
")\n",
"print(f\"loaded {len(segment_aggregates_lookup)} segment aggregates\")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## 5. Prediction helper"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"MU = np.array(scaler[\"mean\"], dtype=np.float32)\n",
"SD = np.array(scaler[\"std\"], dtype=np.float32)\n",
"\n",
"def predict_phase(record: dict) -> dict:\n",
" \"\"\"Predict the kill-chain phase for one event record.\n",
"\n",
" `record` is a dict with event-level fields. Segment-level aggregates\n",
" are pulled automatically from `segment_aggregates_lookup` using the\n",
" `target_segment_id` field.\n",
"\n",
" Returns a dict with both models' predictions and per-class probabilities.\n",
" \"\"\"\n",
" seg_id = record.get(\"target_segment_id\")\n",
" seg_agg = segment_aggregates_lookup.get(seg_id, {})\n",
" X = transform_single(record, meta, segment_aggregates=seg_agg)\n",
"\n",
" xgb_proba = xgb_model.predict_proba(X)[0]\n",
" xgb_label = INT_TO_LABEL[int(np.argmax(xgb_proba))]\n",
"\n",
" Xs = ((X - MU) / SD).astype(np.float32)\n",
" with torch.no_grad():\n",
" logits = mlp_model(torch.tensor(Xs))\n",
" mlp_proba = torch.softmax(logits, dim=1).numpy()[0]\n",
" mlp_label = INT_TO_LABEL[int(np.argmax(mlp_proba))]\n",
"\n",
" return {\n",
" \"xgboost\": {\n",
" \"label\": xgb_label,\n",
" \"probabilities\": {INT_TO_LABEL[i]: float(p) for i, p in enumerate(xgb_proba)},\n",
" },\n",
" \"mlp\": {\n",
" \"label\": mlp_label,\n",
" \"probabilities\": {INT_TO_LABEL[i]: float(p) for i, p in enumerate(mlp_proba)},\n",
" },\n",
" }"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## 6. Run on an example record\n",
"\n",
"This is a real `reconnaissance` event lifted from the sample dataset: opportunistic attacker scanning an email server early in a campaign (timestep 0). Both models should predict `reconnaissance`."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"# Real attack event from the sample dataset (true label: reconnaissance)\n",
"example_record = {\n",
" \"campaign_id\": \"CAMP-000030\",\n",
" \"attacker_id\": \"ATK-0003\",\n",
" \"timestep\": 0,\n",
" \"target_segment_id\": \"SEG-0008\",\n",
" \"target_asset_type\": \"email_server\",\n",
" \"source_ip_class\": \"vpn_tunnel\",\n",
" \"dest_port\": 22,\n",
" \"protocol\": \"icmp\",\n",
" \"bytes_transferred\": 15648.48,\n",
" \"connection_duration_s\": 3.913,\n",
" \"auth_failure_count\": 0,\n",
" \"process_injection_flag\": 0,\n",
" \"lateral_hop_count\": 0,\n",
" \"c2_beacon_interval_s\": 0.0,\n",
" \"detection_outcome\": \"edr_blocked\",\n",
" \"alert_severity\": \"critical\",\n",
" \"siem_rule_triggered\": 0,\n",
" \"edr_blocked_flag\": 1,\n",
" \"attacker_capability_tier\": \"opportunistic\",\n",
" \"defender_maturity_level\": \"baseline\",\n",
"}\n",
"\n",
"result = predict_phase(example_record)\n",
"\n",
"print(f\"XGBoost -> {result['xgboost']['label']}\")\n",
"for lbl, p in sorted(result['xgboost']['probabilities'].items(), key=lambda x: -x[1])[:5]:\n",
" print(f\" P({lbl:25s}) = {p:.4f}\")\n",
"\n",
"print(f\"\\nMLP -> {result['mlp']['label']}\")\n",
"for lbl, p in sorted(result['mlp']['probabilities'].items(), key=lambda x: -x[1])[:5]:\n",
" print(f\" P({lbl:25s}) = {p:.4f}\")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Note: when the two models disagree\n",
"\n",
"XGBoost and the MLP can disagree on out-of-distribution records — particularly hand-crafted inputs whose feature combinations don't sit on the training-data manifold. The MLP, with BatchNorm and a small training set, has narrower competence than the tree ensemble. Disagreement is a useful triage signal: in a SOC workflow, conflicting predictions are flows worth a human eyeball."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## 7. Batch prediction on the sample dataset"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"import pandas as pd\n",
"\n",
"events = pd.read_csv(os.path.join(ds_path, \"attack_events.csv\"))\n",
"\n",
"# Drop leakage columns the model was never trained on\n",
"events = events.drop(columns=[\"technique_id\", \"technique_name\", \"tactic_category\"],\n",
" errors=\"ignore\")\n",
"\n",
"# Score the first 200 events\n",
"sample = events.head(200).copy()\n",
"preds = [predict_phase(row.to_dict())[\"xgboost\"][\"label\"] for _, row in sample.iterrows()]\n",
"sample[\"xgb_pred\"] = preds\n",
"\n",
"ct = pd.crosstab(sample[\"kill_chain_phase\"], sample[\"xgb_pred\"],\n",
" rownames=[\"true\"], colnames=[\"pred\"])\n",
"print(\"Confusion on first 200 sample rows (XGBoost):\")\n",
"print(ct)\n",
"acc = (sample[\"kill_chain_phase\"] == sample[\"xgb_pred\"]).mean()\n",
"print(f\"\\nbatch accuracy on first 200 (in-distribution): {acc:.4f}\")\n",
"print(\"\\nNote: this includes training-set events. See validation_results.json\\n\"\n",
" \"for proper held-out test-set metrics from disjoint campaigns.\")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## 8. Next steps\n",
"\n",
"- See `validation_results.json` for held-out test-set metrics (15 disjoint campaigns, 726 events).\n",
"- See `ablation_results.json` for per-feature-group contribution. `timestep` is by far the most predictive feature, which is honest: kill-chain phases progress in time, so where you are in the campaign timeline carries most of the phase signal.\n",
"- The model card's **Limitations** section explains the gap between this baseline and production threat-detection systems.\n",
"- For the full 380k-row CYB002 dataset and commercial licensing, contact **pradeep@xpertsystems.ai**."
]
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3",
"language": "python",
"name": "python3"
},
"language_info": {
"name": "python",
"version": "3.10"
}
},
"nbformat": 4,
"nbformat_minor": 5
}
|