Design backup and recovery for classification and prediction models and prove it by restoring.
Building Machine Learning Models for Classification and Prediction
Build a working method for classification and prediction models that stands up to scrutiny and survives daily pressure.
📍 Tunis🗓️ 5 training days📚 4 modules🎓 Accredited certificate
5intensive training days
4scientific modules
8training sessions
32detailed points
Course Overview
Availability targets for classification and prediction models are commercial commitments, whatever the engineering team calls them. Security and delivery speed are traded against each other in prediction models whether or not anyone says so. Participants test their assumptions about the technology and systems delivery discipline against scenarios designed to break weak ones. Participants develop a defensible line of reasoning for the choices they make about classification and prediction models. Field experience suggests that the barrier to better this strand of technology and systems delivery is rarely technical. Plans for prediction models often fail at the handover point between functions. The content is relevant to those who own classification and prediction models and to those who are held accountable for its results. The programme takes participants through prediction models end to end, from framing the problem to closing it out. Participants finish with a short, specific brief on the wider technology and systems delivery agenda ready to put in front of a decision maker.
Expected Learning Outcomes
Control technical debt in prediction models deliberately rather than by neglect.
Test classification and prediction models under realistic load before it meets real load.
Measure delivery and reliability of prediction models with indicators teams trust.
Set acceptance criteria for classification and prediction models before work begins rather than after.
Apply a repeatable review cycle to prediction models and act on what it produces.
Communicate the purpose of classification and prediction models to those who have to comply with it.
Who Should Attend
Quality and test engineers verifying classification and prediction models.
Project managers delivering changes to prediction models.
IT managers and service owners responsible for classification and prediction models.
Staff seconded into improvement work on prediction models.
IT governance and audit staff reviewing classification and prediction models.
Business partners who must understand prediction models well enough to challenge it.
Course Modules
Classification and prediction models: change control and rollback
2 sessions · 8 pointsSession 1The cost of classification and prediction models and how to present it
- Establish who is informed, consulted and accountable in classification and prediction models.
- Measure current load on prediction models and project it forward twelve months.
- Set out the decisions in classification and prediction models that require sign-off and by whom.
- Assess the exit route from any cloud or vendor dependency in prediction models.
Session 2Dependencies and supply chain risk in prediction models
- Set out how exceptions to classification and prediction models are requested and approved.
- Build the internal briefing that explains prediction models to those affected.
- Identify the single points of failure in classification and prediction models.
- Confirm data retention and deletion rules applied within prediction models.
Prediction models: incident response and severity
2 sessions · 8 pointsSession 1Comparing prediction models with recognised practice
- Establish the boundary of classification and prediction models and record what sits outside it.
- Confirm every release of prediction models can be rolled back within a defined time.
- Decide what will be stopped to create capacity for classification and prediction models.
- Agree the smallest change to prediction models that would be visibly useful.
Session 2What has to be agreed before work on prediction models starts
- Define acceptance criteria for classification and prediction models in advance.
- Identify where judgement in prediction models is legitimate and where it is not.
- Set delivery and reliability indicators for classification and prediction models the team trusts.
- Test the procedure for prediction models against a realistic scenario.
Prediction models: architecture and designing for failure
2 sessions · 8 pointsSession 1Monitoring prediction models from the user's point of view
- Review logging on classification and prediction models for coverage and retention.
- Record the technical debt in prediction models and schedule repayment.
- Test the failover for classification and prediction models rather than assuming it works.
- Write down the assumptions underpinning the approach to prediction models.
Session 2Sizing capacity for classification and prediction models on measured growth
- Confirm the support model and escalation path for classification and prediction models.
- Inventory third-party dependencies inside prediction models and their update status.
- State the availability and recovery objectives for classification and prediction models as numbers.
- Name a single owner for each element of prediction models.
Prediction models: security, identity and least privilege
2 sessions · 8 pointsSession 1Setting availability and recovery targets for prediction models honestly
- Confirm that those complying with classification and prediction models understand why it exists.
- Close out actions on prediction models rather than leaving them open indefinitely.
- Identify single points of dependency in classification and prediction models and reduce them.
- Define incident severity levels for prediction models and the response each triggers.
Session 2Proving the backup of prediction models by restoring it
- Restore a backup of classification and prediction models in a test environment and time it.
- Configure alerting on prediction models that reflects what users experience.
- Review whether classification and prediction models is aligned with the objectives of the technical platform.
- Apply change control to prediction models including emergency changes.
Choose the package that suits you
Silver Package
At least 3 people
USD1,250
- Workshop or Program Participation
- Airport Transfers
- Customized Badge
- Expert Mentorship (Private Sessions)
- Supervision & Secretarial Services
- Accredited Certificate of Participation
- Complete Training Kit
- Coffee Break
- Closing Ceremony
Gold Package
At least 3 people
USD1,850
- 5-night stay in a 5-star hotel
- Workshop or Program Participation
- Airport Transfers
- Customized Badge
- Expert Mentorship (Private Sessions)
- Supervision & Secretarial Services
- Accredited Certificate of Participation
- Complete Training Kit
- Coffee Break
- Closing Ceremony
Complete your registration
We will contact you within one business day to confirm.