Skip to content

Releases: google-ai-edge/model-explorer

model-explorer-v0.1.10

26 Aug 18:25
Compare
Choose a tag to compare

What's Changed

  • Add support for HLO dialects like mhlo and other hlo family of dialects.

  • Add support for adding custom attributes to group/layer nodes.
    Use the new groupNodeAttributes field in graph_builder.Graph. Example:

    graph = gb.Graph(id='graph',
                     nodes=[top_node, node0, node1, node2],
                     groupNodeAttributes={
                         # Use empty name for graph-level attributes.
                         '': {'model attr': 'value'},
                         'layer1': {'layer attr': 'value'}})

    graph node attributes

  • Add a setting to show out-of-layer edges for an op node without needing to select it.
    Note that this will make the visualization more noisy but you might find it useful in some cases.

    show all edges

  • Allow regex filtering node data table rows by node.

    Screenshot 2024-08-26 at 11 18 55 AM

Full Changelog: model-explorer-v0.1.9...model-explorer-v0.1.10

adapter-v0.1.5

13 Aug 20:09
Compare
Choose a tag to compare
adapter-v0.1.5 Pre-release
Pre-release
Use StableHLO adapter implementation as default in MLIR

This should support dialects like `mhlo` and other `hlo` family of dialects.

PiperOrigin-RevId: 662596997

model-explorer-v0.1.9

12 Aug 18:33
Compare
Choose a tag to compare

What's Changed

  • Now the UI code is fully open-sourced. Thanks for your patience!

  • Add support to pin a single op node to the top of a layer:
    Set GraphNode.config.pinToGroupTop to true in your adapter to pin an op node to the top of the corresponding group.
    Screenshot 2024-08-12 at 11 29 06 AM

  • Fix ExportedProgram display when tensor is None by @justinchuby in #118

  • Minor update to the README file by @copybara-service in #122

Full Changelog: adapter-v0.1.4...model-explorer-v0.1.9

adapter-v0.1.4

31 Jul 19:39
Compare
Choose a tag to compare
adapter-v0.1.4 Pre-release
Pre-release
Rollback the autoformatting

PiperOrigin-RevId: 658075315

model-explorer-v0.1.8

22 Jul 21:27
Compare
Choose a tag to compare

What's Changed

  • Delete unused //tensorflow/python/integration_testing/... and all users by @copybara-service in #100
  • updating pyproject.toml to explicitly only allow supported torch vers… by @pkgoogle in #106
  • Allow users to pass a model name when adding node data through API.
    Example: config.add_node_data_from_path(model_data1, model_name='model1.tflite')
    Also set numpy version < 2 to suppress warning message.

Full Changelog: model-explorer-v0.1.7...model-explorer-v0.1.8

model-explorer-v0.1.7

29 Jun 17:42
Compare
Choose a tag to compare

What's Changed

  • Enable multiple ModuleOps in single MLIR file by @copybara-service in #94
  • [Issue #84] Implemented extensions for API Calls. by @vproHacks in #85
  • Implement stale management for Issues and PRs. by @shmishra99 in #89

New Contributors

Full Changelog: model-explorer-v0.1.6...model-explorer-v0.1.7

adapter-v0.1.3

29 Jun 00:35
01d9ae5
Compare
Choose a tag to compare
adapter-v0.1.3 Pre-release
Pre-release
Implement stale management for Issues and PRs. (#89)

model-explorer-v0.1.6

27 Jun 07:36
Compare
Choose a tag to compare

What's Changed

  • Fix the inputs_map error in pytorch adapter by @yijie-yang in #78
  • Adding auto-assignment scripts and configuration by @pkgoogle in #41
  • Fix incorrect number of constants shown in StableHLO MLIR by @copybara-service in #87
  • Fix a bug in dot backend by @wuruoyu in #88
  • Add Check dialect to the MLIR adapter. by @copybara-service in #91
  • Fix mlir adapter deps by @copybara-service in #92

New Contributors

Full Changelog: model-explorer-v0.1.5...model-explorer-v0.1.6

adapter-v0.1.2

26 Jun 23:45
Compare
Choose a tag to compare
adapter-v0.1.2 Pre-release
Pre-release
Fix mlir adapter deps

PiperOrigin-RevId: 647099419

model-explorer-v0.1.5

04 Jun 21:02
Compare
Choose a tag to compare

What's Changed

  • Updates the custom data overlay colab to use results from TFLite benchmarking. by @akshat14195 in #74
  • Add a command line flag to skip health check after server starts. by @jinjingforever in #76

New Contributors

Full Changelog: model-explorer-v0.1.4...model-explorer-v0.1.5