6 posts tagged with "Enhancement"

Weight each discrete parameter with its own coefficient, and see a preview of the rule as you build it.

---

What's included​

Linear replaces Sum​

• The Sum constraint for discrete numerical parameters is now called Linear
• Give each parameter a coefficient, for example 2 × x0 + x1 <= 100
• Coefficients start at 1, which gives a plain sum
• Use a negative coefficient to compare two parameters

Preview of your rule​

• A preview below the form shows the rule as you fill it in, for example 2 · x0 + x1 <= 100

Example: keep x1 at or below x0​

To make sure x1 ≤ x0, write it as x0 − x1 ≥ 0 and add a Linear constraint:

• Coefficients: x0 = 1, x1 = −1
• Operator: >=
• Right-Hand Side: 0

Other improvements​

• Negative numbers and decimals can be typed directly in Linear and Product constraints
• Removing a parameter from the Selected list now also removes its coefficient or condition

---

Notes​

• Your existing Sum constraints are now Linear constraints with every coefficient set to 1, so they work exactly as before
• Tolerance only applies to the = and != operators

Follow optimization progress more clearly with completed results ordered as observations and predictions grouped separately.

---

What's included​

A clearer optimization sequence​

• Completed target results are numbered in completion order, without gaps from trials that did not produce a measurement
• The original trial number and completion time remain available in the point details
• Predictions for Pending and Accepted trials appear after the completed observations
• A dashed divider separates measured results from predictions across selected targets

---

Notes​

• The observation number can differ from the original trial number when trials finish out of order or do not produce a target result
• Predictions remain model estimates and include error bars showing one posterior standard deviation above and below the mean

Catalyst Assist can now answer questions about the data in your study — counts, averages, distributions, and filtered breakdowns — by querying the dataset directly.

---

What's included​

• Ask data questions in natural language — "average overlay error by tool?", "how many wafers failed process control?", "distribution of risk score by lot" — and Assist queries your study dataset to answer
• Answers are computed from your actual data, returned as markdown tables, with a brief takeaway drawn from the results
• Date and duration math is supported — group by year, compare timestamps, measure elapsed time between events
• Platform and feature questions still work exactly as before — Assist picks the right mode for each question

---

Notes​

• Data answers come only from your dataset — Assist never invents, estimates, or extrapolates figures
• Data questions are available when Assist is opened from within a study
• Data queries run on the study's worker, so it must be provisioned and active; otherwise Assist still answers platform and feature questions but can't query the dataset
• AI responses may not always be accurate. If you receive an unexpected result, let us know with the thumbs up/down on the response, or share it via Support → Contact Support
• This is an incremental update during beta

The folder picker is now easier to navigate, supports deeper search drill-down, and lets you create folders in the right place.

---

What’s included​

Easier folder navigation​

• You can now drill into folders directly from search results instead of starting over in the main tree
• Search breadcrumbs stay clickable, so it’s easier to move back up when you’ve drilled in
• Search results only show folders you can actually use, so there’s less dead-end clicking

Folder creation in the right place​

• The picker follows the same folder structure used in Foundry: space -> project -> folders
• A space is like a shared drive, and its first-level listings are project resources
• Admins manage projects, so you can browse into them but cannot create new folders at the project level itself
• You can create folders once you’re inside the folders within a project, including after drilling in from search

---

Notes​

• Folder creation is only available once you’ve drilled into a valid folder location inside a project
• If a folder does not appear in search, you may not have permission to use it

Create PCA clusters from sigma bands and compare only clusters that contain points.

---

What's included​

Create clusters from sigma bands​

• Open the cluster menu in PCA -> Cluster Analysis to create ready-made Inside nσ and Outside nσ clusters without drawing a brush
• Sigma-based clusters automatically assign matching points as soon as the PCA view finishes loading
• The matching sigma contour adopts the cluster color so you can see the selected boundary at a glance

---

Compare populated clusters only​

• Cluster comparison menus now show only clusters that currently contain points
• If a selected cluster becomes empty or is removed, the comparison updates to the next valid populated cluster automatically

---

Notes​

• Sigma-based clusters are stored with your other PCA cluster selections for that study
• Outside nσ options are available for every displayed sigma band, while Inside nσ options are available for the inner bands
• Empty clusters are excluded from comparison choices until they contain points

We’re moving towards a sidebar-based navigation to improve scalability and deliver a more consistent navigation experience as the platform evolves.

---

What’s included​

Left-hand sidebar​

• Main navigation moved into a left-hand sidebar
• Collapsible sections for better organization

Simplified top bar​

• A simplified top bar, starting with breadcrumbs

Notes​

• This is an incremental update during beta
• Navigation patterns may continue to evolve based on feedback