Fix ML predictions missing from report page

.env.example

@@ -13,6 +13,9 @@ HA_DB_BUCKET=homeassistant/autogen
 HA_DB_USER_NAME="YOUR DB USERNAME HERE"
 HA_DB_PASSWORD="YOUR DB PASSWORD HERE"
 
+# Container timezone (for log timestamps, etc.)
+TZ=Europe/Stockholm
+
 # Development mode
 FLASK_DEBUG=true
 

.gitignore

@@ -27,8 +27,11 @@ yarn-error.log*
 .env.local
 .env.*.local
 
-# Development options (contains sensitive settings)
+# Development options and personal config (contains sensitive settings)
 backend/dev-options.json
+ml_config.yaml
+myconfig.yaml
+ha.secret
 
 # Build output
 dist/
@@ -43,6 +46,9 @@ build/
 # TypeScript
 *.tsbuildinfo
 
+# ML model artifacts (generated by training)
+ml/trained_model.*
+
 # Miscellaneous
 *.log
 .bess-logs/

CLAUDE.md

@@ -11,85 +11,73 @@ BESS Battery Manager is a Home Assistant add-on for optimizing battery energy st
 ### Backend (Python)
 
 ```bash
+# Activate virtual environment (required for all Python commands)
+source .venv/bin/activate
 
 # Install dependencies
-
 pip install -r backend/requirements.txt
 
 # Run development server
-
 ./dev-run.sh
 
 # Run tests
-
 pytest
 pytest core/bess/tests/unit/
 pytest core/bess/tests/integration/
 pytest --cov=core.bess
 
 # Code quality
-
 black .
 ruff check --fix .
 mypy .
-```text
+```
 
 ### Frontend (React/TypeScript)
 
 ```bash
 cd frontend
 
 # Install dependencies
-
 npm install
 
 # Development server
-
 npm run dev
 
 # Build production
-
 npm run build
 
 # Generate API client from OpenAPI spec
-
 npm run generate-api
-```text
+```
 
 ### Docker Development
 
 ```bash
-
 # Start both backend and frontend
-
 docker-compose up -d
 
 # View logs
-
 docker-compose logs -f
-```text
+```
 
 ### Build Add-on
 
 ```bash
 chmod +x package-addon.sh
 ./package-addon.sh
-```text
+```
 
 ### Quality Checks
 
 ```bash
-
 # Run comprehensive quality checks
-
 ./scripts/quality-check.sh
 
 # Individual checks
-
 black .                    # Format Python code
 ruff check --fix .        # Fix Python linting issues
 cd frontend && npm run lint:fix  # Fix TypeScript issues
-```text
+```
 
 ## Architecture Overview
 
@@ -163,16 +151,17 @@ cd frontend && npm run lint:fix  # Fix TypeScript issues
 **Required Analysis Steps**:
 
 1. **Search for Existing Implementations**:
+
    ```bash
    # Search for similar functionality
    grep -r "dataclass\|serialization\|formatting" --include="*.py"
-   grep -r "HealthStatus\|SystemHealth" --include="*.ts" --include="*.tsx" 
+   grep -r "HealthStatus\|SystemHealth" --include="*.ts" --include="*.tsx"
    find . -name "*api*" -name "*model*" -name "*conversion*"
    ```
 
 1. **Examine Related Files**:
    - `backend/api_dataclasses.py` - existing API models
-   - `backend/api_conversion.py` - serialization utilities  
+   - `backend/api_conversion.py` - serialization utilities
    - `frontend/src/types.ts` - TypeScript interfaces
    - `core/bess/` - domain models and services
    - Any files matching the functionality you plan to add
@@ -200,18 +189,21 @@ cd frontend && npm run lint:fix  # Fix TypeScript issues
 ## Codebase Analysis for Sensor Formatting
 
 ### Existing Infrastructure Found:
-- ✅ API Dataclasses: `backend/api_dataclasses.py` 
+
+- ✅ API Dataclasses: `backend/api_dataclasses.py`
 - ✅ Serialization: `backend/api_conversion.py`
 - ✅ Health Types: `frontend/src/types.ts`
 - ✅ Health Endpoint: `/api/system-health` in `backend/api.py`
 
 ### What's Actually Missing:
+
 - ❌ Centralized sensor unit formatting (only frontend string matching exists)
 - ❌ Unit metadata in METHOD_SENSOR_MAP
 
 ### Minimal Required Changes:
+
 1. Add unit metadata to existing METHOD_SENSOR_MAP
-2. Create SensorFormattingService 
+2. Create SensorFormattingService
 3. Integrate with existing health check system
 ```
 
@@ -262,10 +254,11 @@ cd frontend && npm run lint:fix  # Fix TypeScript issues
 - **Proper list spacing**: Add blank line before lists, none between list items
 - **No trailing spaces**: Remove all trailing whitespace
 - **Single blank lines**: Never use multiple consecutive blank lines
-- **Consistent heading levels**: Don't skip heading levels (no h1 → h3)
+- **Consistent heading levels**: Don't skip heading levels (no h1 to h3)
 
-```markdown
+Good markdown example:
 
+```markdown
 # Good Example
 
 ## Header with proper spacing
@@ -279,23 +272,24 @@ This paragraph has proper spacing around it.
 - List item 3
 
 Another paragraph after the list.
+```
 
+Bad markdown example:
+
+```markdown
 ## Bad Example
 
 ###Missing blank lines
 
 - List immediately after header
 - No spacing
-
-
-Too many blank lines above.
-```text
+```
 
 #### Pre-Commit Quality Checklist
 
 Before creating or modifying any files, ALWAYS:
 
-1. **Check Problems Tab**: View → Problems in VS Code - must show zero errors/warnings for modified files
+1. **Check Problems Tab**: View > Problems in VS Code - must show zero errors/warnings for modified files
 2. **Run Code Formatters**:
    - Python: `black .` and `ruff check --fix .`
    - TypeScript: `npm run lint:fix` in frontend directory
@@ -310,7 +304,7 @@ Run the quality check script before committing:
 
 ```bash
 ./scripts/quality-check.sh
-```text
+```
 
 This script automatically checks:
 
@@ -344,7 +338,7 @@ use snake_case. Added conversion to properly map keys before validation.
 Bad commit messages:
 
 ```text
-Fix issue 🤖 Generated with Claude Code
+Fix issue Generated with Claude Code
 Update settings (AI-assisted)
 Changes made by Claude
 ```
@@ -402,18 +396,14 @@ Changes made by Claude
 #### Code Examples
 
 ```python
-
 # Good: Use existing controller method
-
 soc_value = self.ha_controller.get_battery_soc()
 
 # Bad: Manual sensor key extraction
-
 sensor_info = self.ha_controller.METHOD_SENSOR_MAP["get_battery_soc"]
 soc_sensor_key = sensor_info.get("entity_id")
 
 # Good: Use centralized health check
-
 return perform_health_check(
     component_name="Battery Monitoring",
     description="Real-time battery state monitoring",
@@ -424,15 +414,14 @@ return perform_health_check(
 )
 
 # Bad: Custom health check logic with hardcoded thresholds
-
 working_count = sum(1 for method in methods if test_method(method))
 if working_count >= 3:
     return "OK"
 elif working_count >= 1:
     return "WARNING"
 else:
     return "ERROR"
-```text
+```
 
 #### API Conventions
 
@@ -545,10 +534,10 @@ assert slot_start_times == ['02:40', '05:20']  # Specific slot boundaries
 def test_export_arbitrage_enables_battery_discharge():
     strategic_intents[20] = 'EXPORT_ARBITRAGE'
     scheduler.apply_schedule(strategic_intents)
-    
+
     # Test BEHAVIOR: Battery should discharge during target hour
     assert scheduler.is_hour_configured_for_export(20)
-    
+
     # Test CONSTRAINTS: Hardware requirements must be met
     assert scheduler.has_no_overlapping_intervals()
     assert scheduler.intervals_are_chronologically_ordered()
@@ -591,3 +580,24 @@ def test_export_arbitrage_enables_battery_discharge():
 4. **Make tests algorithm-agnostic**: Could a different implementation pass?
 
 **Remember**: Good tests survive refactoring. Bad tests require updates when internal implementation changes.
+
+### Code Intelligence
+
+Prefer LSP over Grep/Read for code navigation - it's faster, precise, and avoids reading entire files:
+
+- `workspaceSymbol` to find where something is defined
+- `findReferences` to see all usages across the codebase
+- `goToDefinition` / `goToImplementation` to jump to source
+- `hover` for type info without reading the file
+
+Use Grep only when LSP isn't available or for text/pattern searches (comments, strings, config).
+
+After writing or editing code, check LSP diagnostics and fix errors before proceeding.
+
+## Python Environment
+
+**Always use the project virtual environment** (`.venv`) for all Python commands. Activate it before running any Python, pytest, black, ruff, or mypy commands:
+
+```bash
+source .venv/bin/activate
+```

DOCS.md

@@ -66,8 +66,19 @@ The add-on requires sensors for:
 
 - Battery: SOC, charge/discharge power, control switches
 - Solar: Production, consumption, grid import/export
-- Pricing: Nordpool spot prices (today and tomorrow)
-- Consumption: 48-hour average forecast
+- Pricing: Electricity spot prices via Nordpool or Octopus Energy (today and tomorrow)
+- Consumption: Depends on `consumption_strategy` setting (see below)
+
+### Consumption Forecasting
+
+Four strategies are available for consumption forecasting, configured via `consumption_strategy` in the `home` section:
+
+| Strategy | Sensor requirements |
+|---|---|
+| `sensor` (default) | `48h_avg_grid_import` sensor configured |
+| `fixed` | None (uses `home.consumption` value) |
+| `influxdb_profile` | `local_load_power` sensor + InfluxDB |
+| `ml_prediction` | Trained ML model + HA weather entity |
 
 See the [Installation Guide](https://github.com/johanzander/bess-manager/blob/main/INSTALLATION.md) for complete sensor configuration examples.
 

Dockerfile

@@ -22,24 +22,26 @@ LABEL \
     org.label-schema.vcs-url="https://github.com/johanzander/bess-manager"
 
 # Install requirements for add-on
-RUN apk add --no-cache \
+RUN apt-get update && apt-get install -y --no-install-recommends \
     python3 \
-    py3-pip \
+    python3-pip \
+    python3-venv \
     python3-dev \
     gcc \
-    musl-dev \
     bash \
     nodejs \
-    npm
+    npm \
+    && rm -rf /var/lib/apt/lists/*
 
 # Set working directory
 WORKDIR /app
 
 # Copy Python application files from backend directory
 COPY backend/app.py backend/api.py backend/api_conversion.py backend/api_dataclasses.py backend/log_config.py backend/requirements.txt ./
 
-# Copy core directory
+# Copy core directory and ML module
 COPY core/ /app/core/
+COPY ml/ /app/ml/
 
 # Build and copy frontend
 WORKDIR /tmp/frontend