Conversation
…d on AUTO_REFRESH config
…illdown outer catch
… not assumed workload label
…instead of daemon uptime
… stale DB reads in dashboard
…utorefresh for proper 2s non-blocking refresh
…into without_llm
- Replace bare Popen with a watchdog thread that monitors cognios_as_daemon.py
- Auto-restarts daemon after 3s if it exits unexpectedly (fixes stale FocusOS pipeline)
- Add 2s startup grace period before dashboard launch so DB is ready
- Graceful shutdown on Ctrl+C via _stop_watchdog event (no zombie restarts)
…ual environment - Create requirements.txt with all pinned third-party dependencies across all modules - Update pyproject.toml with missing requests and umap-learn dependencies - Clean up merge conflict markers in .gitignore and ensure .venv, .conda, databases, and logs are ignored - Enhance main.py with auto-detection and seamless re-execution into project .venv - Add self-healing bootstrap in main.py to create .venv and install requirements if missing - Add cross-platform 1-click startup scripts (start.sh and start.bat) - Update README.md Quick Start with standardized environment setup instructions
…isualization script
…oise - Add _EXCLUDE_EXACT_NAMES and _EXCLUDE_SUBSTRINGS sets to silently drop Chrome/Chromium (Streamlit dashboard) and CogniOS daemon processes from process telemetry to prevent corrupting compiler_active signal and inflating CPU readings in FocusOS inference - Add UDP/TCP ratio calculation via psutil.net_connections with graceful fallback (0.10) on permission error - Pass udp_tcp_ratio through the metrics dict and into write_layer1() in the daemon loop
Doubled SLIDING_WIND_N to 30 to give FocusOS feature extraction more telemetry history per inference cycle, reducing variance in short-burst workload misclassifications.
Rewrote extract_features() to produce the full 22-feature statistical vector that matches the IdeaPad training dataset schema: - cpu_mean, cpu_max, cpu_variance - ram_mean, ram_growth_rate, swap_percent - network_mean, network_symmetry, net_variance, udp_tcp_ratio - disk_io_mean, process_count_mean, thread_count_mean - load_avg, ctx_switches_per_core, cpu_user_system_ratio - psi_cpu_some, psi_mem_some, psi_io_some - vscode_active, browser_active, compiler_active Added timestamp-aware time-delta computation, ddof=0 variance, and safe column presence checks throughout.
…ition Rewrote WorkloadPredictor and training pipeline in classifier.py: - Added full 22-feature schema (matches generate_dataset_ideapad.py) - Integrated StandardScaler alongside LabelEncoder for robust normalization - Added StratifiedKFold cross-validation with cross_val_score reporting - Candidate dataset path resolution (searches project root, ~/home, etc.) - Path now resolves to focusos/models/models_saved/ for local artifacts Updated cluster_trainer.py: - Minor config alignment with new MODELS_DIR path conventions
…l artifacts Removed: - focusos/models/focusos_training_data.csv (old pseudo-labeled dataset, 3001 rows) - focusos/models/generate_dataset.py (old synthetic dataset generator) - focusos/models_saved/kmeans_model.pkl (unsupervised clustering no longer used) - focusos/models_saved/pseudo_labeled_dataset.csv (replaced by IdeaPad dataset) Updated saved model artifacts to reflect the new 22-feature IdeaPad schema: - feature_columns.pkl: 15→22 feature names - label_encoder.pkl: updated class mappings - scaler.pkl: re-fit StandardScaler on IdeaPad data - xgboost_model.json: re-trained model checkpoint
…ne process loop, handle OS permissions
Extend the write_layer1() signature and INSERT statement to accept a udp_tcp_ratio column (default 0.10) in the layer1_sys table. This prepares the schema for the new FocusOS 22-feature network metric.
…n and changed the number of tokens
NikhByte
requested review from
Brijesh-Thakkar,
cherry-aggarwal and
coderTanisha22
as code owners
August 28, 2026 18:00
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
5 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Merging Frontend to the Staging