- Database
- sqlite/postgres/duckDB database hosted on a server
- CI/CD
- Github actions to run tests, and to update db regularly?
- Serving/inference
- banana.dev for ML inference? Works with docker - see here
- Testing
- pytest
- coverage report in repo
- Repo structure
models.py
for ORM classes,pipelines.py
for data access and pipelines- OR: use
data_access/
dir?
- OR: use
data/
for example csv dataarchimedes/
for analysis codearchimedes/forecasting/
archimedes/plotting/
archimedes/preprocessing/
test/
for teststest_archimedes/
for unit teststest_integration/
for integration tests?
- Code and container images on GitLab
- Server has postgresql DB on port 5433. URL:
jdbc:postgresql://<address>:5433/cm_db
- STL decomposition
- Comparison with forecast
- Outlier detection
- Correlation with walking data? Read in data from weather API?
- Google Cloud run
- Cloud DBs:
graph TD;
A-->B;
A-->C;
B-->D;
C-->D;