57cf6c3828
- TrafficAutoEncoder class: symmetric AE (n→64→32→16→32→64→n) with BatchNorm+ReLU - Trained alongside EIF on human_baseline, saved/loaded with model versioning - Score = per-sample MSE reconstruction error, combined with EIF via AE_WEIGHT (α=0.30) - AE latent space (16-dim) used for HDBSCAN clustering instead of raw features - Configurable: AE_WEIGHT, AE_EPOCHS, AE_LATENT_DIM, AE_LEARNING_RATE - Graceful fallback: if torch unavailable or AE fails, EIF-only scoring continues - ClickHouse: ae_recon_error column added to ml_all_scores - Tests: 5 new tests (AE train/score, encode latent, state dict save/load, weight combination) Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
ClickHouse Migrations — ja4-platform
Migration Order
Apply these files in numeric order against the ClickHouse server:
clickhouse-client --multiquery < 00_database.sql
clickhouse-client --multiquery < 01_raw_tables.sql
clickhouse-client --multiquery < 02_dictionaries.sql
clickhouse-client --multiquery < 03_anubis_tables.sql
clickhouse-client --multiquery < 04_mv_http_logs.sql
clickhouse-client --multiquery < 05_aggregation_tables.sql
clickhouse-client --multiquery < 06_ml_tables.sql
clickhouse-client --multiquery < 07_ai_features_view.sql
clickhouse-client --multiquery < 08_users.sql
clickhouse-client --multiquery < 09_audit_table.sql
File Descriptions
| File | Contents |
|---|---|
00_database.sql |
CREATE DATABASE |
01_raw_tables.sql |
http_logs_raw ingest table |
02_dictionaries.sql |
ASN geo dict, bot IP/JA4/network reference tables |
03_anubis_tables.sql |
Anubis crawler rule tables and dictionaries (UA, IP, ASN, country) |
04_mv_http_logs.sql |
Canonical http_logs target table + mv_http_logs materialized view with full Anubis enrichment |
05_aggregation_tables.sql |
agg_host_ip_ja4_1h, agg_header_fingerprint_1h + their MVs |
06_ml_tables.sql |
ml_detected_anomalies, ml_all_scores |
07_ai_features_view.sql |
view_ai_features_1h with Anubis enrichment |
08_users.sql |
ClickHouse users and grants |
09_audit_table.sql |
audit_logs table for SOC dashboard audit trail |
Prerequisites
Place CSV data files in /var/lib/clickhouse/user_files/:
iplocate-ip-to-asn.csv— IP-to-ASN mapping (from IPLocate)bot_ip.csv— Known bot IP prefixesbot_ja4.csv— Known bot JA4 fingerprintsasn_reputation.csv— ASN reputation labels
Notes
04_mv_http_logs.sqlis the canonical version of the MV, superseding the base version inservices/correlator/sql/init.sql. It includes full Anubis enrichment.- All migrations are idempotent (use
IF NOT EXISTS/IF EXISTS). - Anubis dictionary passwords in
03_anubis_tables.sqlmust be changed before production use.