Fix self.oldmu initialization causing TypeError in KMeans.find_centers()

[Copilot]

KMeans.find_centers() called _has_converged() before self.oldmu was initialized to a valid value, causing TypeError: object of type 'NoneType' has no len() on first iteration.

Changes

• Initialize self.oldmu to K distinct arrays with infinity values before entering convergence loop
• Ensures convergence check can safely access len(self.oldmu) and returns False on first iteration
• Uses list comprehension to create independent array objects

# Before the convergence loop in find_centers()
self.oldmu = [np.array([float('inf')] * len(X[0])) for _ in range(K)]

iter = 0
while not self._has_converged() and iter < nitermax:
    self.oldmu = self.mu
    # ...

The infinity values guarantee initial centers differ from self.mu, allowing the algorithm to run while maintaining correct convergence detection.

---

💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.