An open API service for providing issue and pull request metadata for open source projects.

GitHub / pymc-devs/pytensor issues and pull requests

Labelled with: memory_optimization

#1515 - Rethink the special-case for Elemwise/Blockwise InplaceOptimizer for "large graphs"

Issue - State: open - Opened by ricardoV94 5 months ago
Labels: graph rewriting, memory_optimization

#1494 - Do not skip validation between consecutive Elemwise inplace replacements

Pull Request - State: closed - Opened by ricardoV94 6 months ago - 3 comments
Labels: bug, graph rewriting, memory_optimization

#1494 - Do not skip validation between consecutive Elemwise inplace replacements

Pull Request - State: closed - Opened by ricardoV94 6 months ago - 3 comments
Labels: bug, graph rewriting, memory_optimization

#1457 - Blockwise inplace optimizer does not filter out duplicated/aliased inputs from candidates

Issue - State: closed - Opened by ricardoV94 6 months ago
Labels: graph rewriting, memory_optimization

#1377 - Does `destroy_map` need to specify which output destroys which input?

Issue - State: open - Opened by ricardoV94 8 months ago
Labels: needs info, graph objects, memory_optimization

#1346 - Speedup AdvancedSubtensor1 and AdvancedIncSubtensor1 in C backend

Pull Request - State: closed - Opened by ricardoV94 8 months ago - 4 comments
Labels: major, C-backend, performance, indexing, memory_optimization

#1346 - Speedup AdvancedSubtensor1 and AdvancedIncSubtensor1 in C backend

Pull Request - State: closed - Opened by ricardoV94 8 months ago - 4 comments
Labels: major, C-backend, performance, indexing, memory_optimization

#1343 - Propagate static output shapes in Split and avoid copy in C-impl

Pull Request - State: closed - Opened by ricardoV94 8 months ago - 1 comment
Labels: C-backend, performance, memory_optimization

#1340 - Reuse output of Join in C backend

Pull Request - State: closed - Opened by ricardoV94 8 months ago - 1 comment
Labels: C-backend, performance, memory_optimization

#1334 - Specialize away Split

Pull Request - State: closed - Opened by ricardoV94 8 months ago - 1 comment
Labels: maintenance, performance, memory_optimization

#1334 - Specialize away Split

Pull Request - State: closed - Opened by ricardoV94 8 months ago - 1 comment
Labels: maintenance, performance, memory_optimization

#1333 - Inplace join

Pull Request - State: closed - Opened by ricardoV94 8 months ago - 1 comment
Labels: memory_optimization

#1322 - Inplace Composite and ScalarLoop Ops with multiple outputs

Pull Request - State: closed - Opened by ricardoV94 9 months ago - 2 comments
Labels: bug, enhancement, numba, performance, memory_optimization

#1322 - Inplace Composite and ScalarLoop Ops with multiple outputs

Pull Request - State: closed - Opened by ricardoV94 9 months ago - 2 comments
Labels: bug, enhancement, numba, performance, memory_optimization

#1283 - Inplace on sit-sot / mit-sot when nsteps is symbolic

Issue - State: open - Opened by ricardoV94 9 months ago
Labels: numba, scan, memory_optimization

#1282 - Allow inner-function inplace of mit-mot in Scan backend

Issue - State: open - Opened by ricardoV94 9 months ago
Labels: numba, performance, memory_optimization

#1281 - Speedup Scan in different backends

Pull Request - State: closed - Opened by ricardoV94 9 months ago - 2 comments
Labels: performance, scan, memory_optimization

#1281 - Speedup `Scan` in different backends

Pull Request - State: closed - Opened by ricardoV94 9 months ago - 2 comments
Labels: performance, scan, memory_optimization

#1247 - Fix inner graph inplace rewrites in Numba / PyTorch backends

Pull Request - State: closed - Opened by ricardoV94 10 months ago - 5 comments
Labels: bug, numba, scan, OpFromGraph, torch, memory_optimization

#1247 - Fix inner graph inplace rewrites in Numba / PyTorch backends

Pull Request - State: closed - Opened by ricardoV94 10 months ago - 5 comments
Labels: bug, numba, scan, OpFromGraph, torch, memory_optimization