Web Machine Learning Working Group - Activity in 2026
No matches.
GitHub
webmachinelearning/webnn
-
2 issues opened.
-
Explore W3C Open Source Program as a way to support GPU-backed WPT runners webmachinelearning/webnn#972, by @anssiko.
-
Investigate how to integrate WebNN caching with Cross-Origin Storage webmachinelearning/webnn#971, by @anssiko.
-
-
14 issues received 15 new comments.
-
Proposal: Support build-time metadata to describe how a graph will be driven webmachinelearning/webnn#965 (2 comments by @anssiko, @wangw-1991)
-
Explore W3C Open Source Program as a way to support GPU-backed WPT runners webmachinelearning/webnn#972 (1 comment by @ianbjacobs)
-
Consider specify lower limit for conv2d/pool2d kernel sizes, dilations, strides webmachinelearning/webnn#928 (1 comment by @anssiko)
-
Expand the expand operator to support blockwise broadcasting webmachinelearning/webnn#903 (1 comment by @anssiko)
-
Make zeroPoint optional for quantizeLinear and dequantizeLinear operations webmachinelearning/webnn#943 (1 comment by @anssiko)
-
-
1 pull request received 23 new comments.
-
Add explainer: Support dynamic language models on NPUs via WebNN webmachinelearning/webnn#969 (23 comments by @matthargett, @mklimenko-nv, @wangw-1991)
-
TR publications
-
3 documents published.
-
Web Neural Network API published 3 times as a Candidate Recommendation Draft. History of Web Neural Network API
-
Past meetings
Participants
-
1 participant joined.
-
Dave Raggett (W3C) joined.
-
-
5 participants left.
-
Lingyan Zhao (Microsoft Corporation) left.
-
Chai Chaoweeraprasit (Microsoft Corporation) left.
-
Josi Rosenfeld (CleverThis, Inc.) left.
-
Adnaan Nazir (CleverThis, Inc.) left.
-
Aditya Chhabra (CleverThis, Inc.) left.
-
GitHub
webmachinelearning/webnn
-
9 issues opened.
-
Consider dropping negative indices support in gather*/scatter* operators webmachinelearning/webnn#970, by @shiyi9801.
-
Add `topK` operator (a sliced `sort`) webmachinelearning/webnn#968, by @fdwr.
-
Do subnormal results have to be preserved when the reference formula overflows in float32? webmachinelearning/webnn#967, by @matthargett.
-
Proposal: Support stateful model for WebNN webmachinelearning/webnn#966, by @wangw-1991.
-
Proposal: Support build-time metadata to describe how a graph will be driven webmachinelearning/webnn#965, by @wangw-1991.
-
What is the current testing/shipping status of WebNN in Chrome and Edge, and what should developers use today? webmachinelearning/webnn#964, by @webmaxru.
-
Expose engine-defined compute timing for WebNN (`MLComputeTiming`) webmachinelearning/webnn#963, by @shubhamg13.
-
Direct Video Plane Access for WebNN webmachinelearning/webnn#961, by @bbernhar.
-
Clarify finite/infinite classification at numerical tolerance boundaries webmachinelearning/webnn#960, by @matthargett.
-
-
3 issues closed.
-
conv2d: infer groups from input and filter shapes when not specified webmachinelearning/webnn#959.
-
Validate LSTM/GRU full-sequence output regardless of the `returnSequence` flag webmachinelearning/webnn#949.
-
Proposal: STC-1.0 Tensor Container for Zero-Copy WebGPU/Wasm Loading and Cryptographic Provenance webmachinelearning/webnn#952.
-
-
19 issues received 36 new comments.
-
Proposal: Support build-time metadata to describe how a graph will be driven webmachinelearning/webnn#965 (6 comments by @fdwr, @matthargett, @reillyeon, @wangw-1991)
-
Proposal: Support stateful model for WebNN webmachinelearning/webnn#966 (5 comments by @fdwr, @matthargett, @sushraja-msft, @wangw-1991)
-
Expose engine-defined compute timing for WebNN (`MLComputeTiming`) webmachinelearning/webnn#963 (4 comments by @shubhamg13, @vmukhachev)
-
(de)quantization behaviors on CoreML webmachinelearning/webnn#822 (2 comments by @fdwr, @matthargett)
-
Add `topK` operator (a sliced `sort`) webmachinelearning/webnn#968 (2 comments by @fdwr, @matthargett)
-
Effective MLComputePolicy exposure webmachinelearning/webnn#934 (2 comments by @anssiko, @shubhamg13)
-
conv2d: infer groups from input and filter shapes when not specified webmachinelearning/webnn#959 (2 comments by @fdwr, @ibelem)
-
Proposal: STC-1.0 Tensor Container for Zero-Copy WebGPU/Wasm Loading and Cryptographic Provenance webmachinelearning/webnn#952 (2 comments by @anssiko, @sovereign-ecosystem)
-
Consider dropping negative indices support in gather*/scatter* operators webmachinelearning/webnn#970 (1 comment by @fdwr)
-
Expand the expand operator to support blockwise broadcasting webmachinelearning/webnn#903 (1 comment by @matthargett)
-
Clarify finite/infinite classification at numerical tolerance boundaries webmachinelearning/webnn#960 (1 comment by @fdwr)
-
Proposal: API to Separate Graph Building from Weight Loading to Reduce Peak Memory Usage webmachinelearning/webnn#901 (1 comment by @RafaelCintron)
-
Device selection criteria for usecase-driven scenarios webmachinelearning/webnn#902 (1 comment by @RattanPriya)
-
How to support scatterElements on TFLite backend when indices operand is not constant webmachinelearning/webnn#878 (1 comment by @vmukhachev)
-
What is the current testing/shipping status of WebNN in Chrome and Edge, and what should developers use today? webmachinelearning/webnn#964 (1 comment by @anssiko)
-
reduceLogSumExp reference decomposition contradicts conformance test; accuracy requirements missing webmachinelearning/webnn#947 (1 comment by @fdwr)
-
-
1 pull request opened.
-
Add explainer: Support dynamic language models on NPUs via WebNN webmachinelearning/webnn#969, by @wangw-1991.
-
-
5 pull requests merged.
-
Add `bfloat16` and `float8e4m3` to the spec webmachinelearning/webnn#956.
-
Add dynamic shape explainer webmachinelearning/webnn#945.
-
Add low-precision floating point data types explainer webmachinelearning/webnn#938.
-
Add stable decomposition clarification webmachinelearning/webnn#957.
-
Enforce validation when creating an MLOperandDescriptor webmachinelearning/webnn#955.
-
-
7 pull requests received 28 new comments.
-
Add dynamic shape explainer webmachinelearning/webnn#945 (12 comments by @anssiko, @huningxin, @miaobin, @reillyeon, @sushraja-msft)
-
Add `bfloat16` and `float8e4m3` to the spec webmachinelearning/webnn#956 (11 comments by @fdwr, @mklimenko-nv)
-
Add explainer: Support dynamic language models on NPUs via WebNN webmachinelearning/webnn#969 (1 comment by @wangw-1991)
-
Refactor device selection: Rename to computePolicy, remove accelerated, and add fallback webmachinelearning/webnn#923 (1 comment by @mingmingtasd)
-
Add low-precision floating point data types explainer webmachinelearning/webnn#938 (1 comment by @huningxin)
-
Enforce validation when creating an MLOperandDescriptor webmachinelearning/webnn#955 (1 comment by @shiyi9801)
-
TR publications
-
2 documents published.
-
Web Neural Network API published twice as a Candidate Recommendation Draft. History of Web Neural Network API
-
Past meetings
Participants
-
1 participant joined.
-
Rattan Priya Bhasin (Microsoft Corporation) joined.
-
-
3 participants left.
-
Kaushik Satpathy (Yahoo Holdings Inc.) left.
-
ChangSeok Oh (ByteDance) left.
-
Stephan Morris (See Me Please) left.
-
-
1 organization left.
-
Yahoo Holdings Inc. left.
-
GitHub
webmachinelearning/webnn
-
2 issues opened.
-
conv2d: infer groups from input and filter shapes when not specified webmachinelearning/webnn#959, by @ibelem.
-
MLTensor export to GPUBuffer with custom usages? webmachinelearning/webnn#954, by @mwyrzykowski.
-
-
3 issues closed.
-
A conformance suite with disallowed intra-op examples would be helpful for hardening. webmachinelearning/webnn#244.
-
Mandate `output_channels % groups == 0` validation for conv2d webmachinelearning/webnn#925.
-
`gather` should allow 0D output webmachinelearning/webnn#946.
-
-
13 issues received 31 new comments.
-
Composite operators / subgraphs webmachinelearning/webnn#907 (9 comments by @anssiko, @huningxin, @mwyrzykowski, @OlehZhyhinas, @snnn, @Yuhengwe1)
-
Accumulation precision is unspecified for matmul/gemm and reduction operations webmachinelearning/webnn#948 (3 comments by @anssiko, @fdwr, @matthargett)
-
MLTensor export to GPUBuffer with custom usages? webmachinelearning/webnn#954 (3 comments by @bbernhar, @fdwr, @mwyrzykowski)
-
Validate LSTM/GRU full-sequence output regardless of the `returnSequence` flag webmachinelearning/webnn#949 (3 comments by @anssiko, @huningxin, @shiyi9801)
-
Consider specify lower limit for conv2d/pool2d kernel sizes, dilations, strides webmachinelearning/webnn#928 (2 comments by @anssiko)
-
Discrepancy in actual and expected output for subgraph test : batchNormalization options.axis=0 + gelu webmachinelearning/webnn#950 (2 comments by @fdwr, @matthargett)
-
Allow 0 size dimensions (dimensions containing a 0 in the list of sizes, not a rank of 0 which is valid) webmachinelearning/webnn#391 (2 comments by @anssiko, @matthargett)
-
reduceLogSumExp reference decomposition contradicts conformance test; accuracy requirements missing webmachinelearning/webnn#947 (2 comments by @anssiko, @matthargett)
-
A conformance suite with disallowed intra-op examples would be helpful for hardening. webmachinelearning/webnn#244 (1 comment by @anssiko)
-
Mandate `output_channels % groups == 0` validation for conv2d webmachinelearning/webnn#925 (1 comment by @anssiko)
-
-
4 pull requests opened.
-
Add stable decomposition clarification webmachinelearning/webnn#957, by @mklimenko-nv.
-
Add `bfloat16` and `float8e4m3` to the spec webmachinelearning/webnn#956, by @mklimenko-nv.
-
Enforce validation when creating an MLOperandDescriptor webmachinelearning/webnn#955, by @shiyi9801.
-
Bugfix: Allow 0-D output for gather() webmachinelearning/webnn#953, by @huningxin.
-
-
2 pull requests merged.
-
Add initial WebNN-WebGPU Interop specification webmachinelearning/webnn#942.
-
Bugfix: Allow 0-D output for gather() webmachinelearning/webnn#953.
-
-
5 pull requests received 62 new comments.
-
Add dynamic shape explainer webmachinelearning/webnn#945 (45 comments by @anssiko, @fdwr, @matthargett, @miaobin, @reillyeon, @theHamsta)
-
Add initial WebNN-WebGPU Interop specification webmachinelearning/webnn#942 (10 comments by @anssiko, @bbernhar, @fdwr)
-
Enforce validation when creating an MLOperandDescriptor webmachinelearning/webnn#955 (4 comments by @anssiko, @huningxin, @shiyi9801)
-
Add low-precision floating point data types explainer webmachinelearning/webnn#938 (2 comments by @anssiko, @mklimenko-nv)
-
TR publications
-
1 document published.
-
Web Neural Network API published as a Candidate Recommendation Draft. History of Web Neural Network API
-
Past meetings
Participants
-
5 participants joined.
-
Rezvan Mahdavi Hezaveh (Google LLC) joined.
-
Thomas Steiner (Google LLC) joined.
-
Mikhail Klimenko (NVIDIA) joined.
-
Jay Mok (PayPal) joined.
-
Stephan Morris (See Me Please) joined.
-
-
1 organization joined.
-
See Me Please joined.
-
GitHub
webmachinelearning/webnn
-
6 issues opened.
-
Proposal: STC-1.0 Tensor Container for Zero-Copy WebGPU/Wasm Loading and Cryptographic Provenance webmachinelearning/webnn#952, by @sovereign-ecosystem.
-
Discrepancy in actual and expected output for subgraph test : batchNormalization options.axis=0 + gelu webmachinelearning/webnn#950, by @hthadicherla.
-
Validate LSTM/GRU full-sequence output regardless of the `returnSequence` flag webmachinelearning/webnn#949, by @shiyi9801.
-
Accumulation precision is unspecified for matmul/gemm and reduction operations webmachinelearning/webnn#948, by @mtavenrath.
-
reduceLogSumExp reference decomposition contradicts conformance test; accuracy requirements missing webmachinelearning/webnn#947, by @mtavenrath.
-
`gather` should allow 0D output webmachinelearning/webnn#946, by @Honry.
-
-
7 issues received 8 new comments.
-
Discrepancy in actual and expected output for subgraph test : batchNormalization options.axis=0 + gelu webmachinelearning/webnn#950 (1 comment by @adrastogi)
-
reduceLogSumExp reference decomposition contradicts conformance test; accuracy requirements missing webmachinelearning/webnn#947 (1 comment by @fdwr)
-
Accumulation precision is unspecified for matmul/gemm and reduction operations webmachinelearning/webnn#948 (1 comment by @fdwr)
-
Validate LSTM/GRU full-sequence output regardless of the `returnSequence` flag webmachinelearning/webnn#949 (1 comment by @shiyi9801)
-
Make zeroPoint optional for quantizeLinear and dequantizeLinear operations webmachinelearning/webnn#943 (1 comment by @fdwr)
-
1 pull request opened.
-
Add dynamic shape explainer webmachinelearning/webnn#945, by @miaobin.
-
-
1 pull request merged.
-
Add roundEven(), isNaN(), isInfinite() to Operators webmachinelearning/webnn#936.
-
-
5 pull requests received 64 new comments.
-
Add dynamic shape explainer webmachinelearning/webnn#945 (37 comments by @anssiko, @fdwr, @huningxin, @miaobin, @w3cbot)
-
Add low-precision floating point data types explainer webmachinelearning/webnn#938 (17 comments by @anssiko, @fdwr, @huningxin, @mklimenko-nv)
-
Add initial WebNN-WebGPU Interop specification webmachinelearning/webnn#942 (5 comments by @huningxin, @mwyrzykowski)
-
Add roundEven(), isNaN(), isInfinite() to Operators webmachinelearning/webnn#936 (3 comments by @huningxin, @shubhamg13)
-
fix reduceLogSumExp, softplus, and tanh decompositions to not overflow so easily webmachinelearning/webnn#916 (2 comments by @fdwr, @vmukhachev)
-
Participants
-
2 Team contacts appointed.
-
Ruoxi Ran became a Team contact.
-
Dominique Hazaël-Massieux became a Team contact.
-
-
2 Team contacts stepped down.
-
Dominique Hazaël-Massieux stepped down as a Team contact.
-
Dominique Hazaël-Massieux stepped down as a Team contact.
-
-
4 participants joined.
-
Matt Kubej (Shopify) joined.
-
Lingyan Zhao (Microsoft Corporation) joined.
-
Amit Varia (Google LLC) joined.
-
Ruoxi Ran (W3C) joined.
-
-
1 participant left.
-
Dominique Hazaël-Massieux (W3C) left.
-
GitHub
webmachinelearning/webnn
-
3 issues opened.
-
Make zeroPoint optional for quantizeLinear and dequantizeLinear operations webmachinelearning/webnn#943, by @Yuhengwe1.
-
WebNN/WebGPU zero-copy device selection webmachinelearning/webnn#941, by @bbernhar.
-
Remove the constant tensor support webmachinelearning/webnn#940, by @mingmingtasd.
-
-
6 issues received 16 new comments.
-
Remove the constant tensor support webmachinelearning/webnn#940 (6 comments by @bbernhar, @fdwr, @reillyeon)
-
WebNN/WebGPU zero-copy device selection webmachinelearning/webnn#941 (2 comments by @bbernhar, @reillyeon)
-
Consider specify lower limit for conv2d/pool2d kernel sizes, dilations, strides webmachinelearning/webnn#928 (2 comments by @huningxin, @wangw-1991)
-
-
1 pull request opened.
-
Add initial WebNN-WebGPU Interop specification webmachinelearning/webnn#942, by @bbernhar.
-
-
1 pull request received 1 new comment.
-
Add initial WebNN-WebGPU Interop specification webmachinelearning/webnn#942 (1 comment by @bbernhar)
-
TR publications
-
1 document published.
-
Web Neural Network API published as a Candidate Recommendation Draft. History of Web Neural Network API
-
Past meetings
Participants
-
4 participants joined.
-
Philippe Le Hegaret (W3C) joined.
-
Julien Bataille (Rakuten Group, Inc.) joined.
-
Yehonatan Daniv (Wix.com Ltd.) joined.
-
Severin Ferrand (Google LLC) joined.
-
-
2 participants left.
-
Simon Wijckmans (Client side development Inc) left.
-
Jonathan Schneerson (Temporal Series AI) left.
-
-
2 organizations joined.
-
Rakuten Group, Inc. joined.
-
Wix.com Ltd. joined.
-
-
3 organizations left.
-
Client side development Inc left.
-
Criteo left.
-
Temporal Series AI left.
-
GitHub
webmachinelearning/webnn
-
1 issue opened.
-
The base and exponent of pow() should be allowed to have different data types webmachinelearning/webnn#939, by @miaobin.
-
-
2 issues closed.
-
Add minimum data type set and rank range for input, constant, output and each operator into Spec webmachinelearning/webnn#896.
-
Consider specifying tensor element count limit webmachinelearning/webnn#924.
-
-
4 issues received 16 new comments.
-
RFE: Add support for more floating point low-precision ML data types (`bfloat16`, `fp8`, `nvfp4`) webmachinelearning/webnn#930 (6 comments by @anssiko, @fdwr, @mklimenko-nv, @snnn)
-
Effective MLComputePolicy exposure webmachinelearning/webnn#934 (4 comments by @anssiko, @huningxin, @mwyrzykowski)
-
Consider specify lower limit for conv2d/pool2d kernel sizes, dilations, strides webmachinelearning/webnn#928 (3 comments by @fdwr, @philloooo)
-
Discrepancies with maxPool2d padding behavior webmachinelearning/webnn#935 (3 comments by @anssiko, @fdwr, @huningxin)
-
-
2 pull requests opened.
-
Add low-precision floating point data types explainer webmachinelearning/webnn#938, by @mklimenko-nv.
-
Add roundEven(), isNaN(), isInfinite() to Operators webmachinelearning/webnn#936, by @shubhamg13.
-
-
1 pull request merged.
-
add minimum data types and rank range for operations webmachinelearning/webnn#910.
-
-
2 pull requests received 9 new comments.
-
add minimum data types and rank range for operations webmachinelearning/webnn#910 (7 comments by @BruceDai, @fdwr, @huningxin)
-
Add roundEven(), isNaN(), isInfinite() to Operators webmachinelearning/webnn#936 (2 comments by @shubhamg13, @w3cbot)
-
TR publications
-
1 document published.
-
Web Neural Network API published as a Candidate Recommendation Draft. History of Web Neural Network API
-
Past meetings
Participants
-
1 participant joined.
-
Geunhyung Kim (Sharing Economy Association of Korea (SEAK)) joined.
-
GitHub
webmachinelearning/webnn
-
2 issues opened.
-
Discrepancies with maxPool2d padding behavior webmachinelearning/webnn#935, by @philloooo.
-
Effective MLComputePolicy exposure webmachinelearning/webnn#934, by @anssiko.
-
-
2 issues closed.
-
Define an upper bound on number of concat inputs and split outputs webmachinelearning/webnn#931.
-
Get devices used for a graph after graph compilation webmachinelearning/webnn#836.
-
-
9 issues received 44 new comments.
-
Effective MLComputePolicy exposure webmachinelearning/webnn#934 (17 comments by @anssiko, @fdwr, @handellm, @mwyrzykowski, @zolkis)
-
RFE: Add support for more floating point low-precision ML data types (`bfloat16`, `fp8`, `nvfp4`) webmachinelearning/webnn#930 (10 comments by @anssiko, @fdwr, @huningxin, @mtavenrath, @reillyeon)
-
Support flexible input sizes webmachinelearning/webnn#883 (7 comments by @fdwr, @Honry, @huningxin, @miaobin)
-
Allow 0 size dimensions (dimensions containing a 0 in the list of sizes, not a rank of 0 which is valid) webmachinelearning/webnn#391 (4 comments by @anssiko, @bbernhar, @fdwr)
-
Consider specify lower limit for conv2d/pool2d kernel sizes, dilations, strides webmachinelearning/webnn#928 (2 comments by @anssiko, @dsharlet)
-
Define an upper bound on number of concat inputs and split outputs webmachinelearning/webnn#931 (1 comment by @anssiko)
-
Discrepancies with maxPool2d padding behavior webmachinelearning/webnn#935 (1 comment by @philloooo)
-
Get devices used for a graph after graph compilation webmachinelearning/webnn#836 (1 comment by @anssiko)
-
-
2 pull requests opened.
-
validate concat inputs and split splits webmachinelearning/webnn#933, by @philloooo.
-
Validate output channels are a multiple of groups in conv2d webmachinelearning/webnn#932, by @lynne0326.
-
-
2 pull requests merged.
-
validate concat inputs and split splits webmachinelearning/webnn#933.
-
Validate output channels are a multiple of groups in conv2d webmachinelearning/webnn#932.
-
-
4 pull requests received 23 new comments.
-
add minimum data types and rank range for operations webmachinelearning/webnn#910 (13 comments by @BruceDai, @fdwr, @huningxin)
-
validate concat inputs and split splits webmachinelearning/webnn#933 (5 comments by @anssiko, @fdwr, @philloooo)
-
Validate output channels are a multiple of groups in conv2d webmachinelearning/webnn#932 (3 comments by @anssiko, @huningxin, @lynne0326)
-
Refactor device selection: Rename to computePolicy, remove accelerated, and add fallback webmachinelearning/webnn#923 (2 comments by @anssiko, @mingmingtasd)
-
Past meetings
-
1 meeting.
Participants
-
3 participants joined.
-
Sarah Drasner (Google LLC) joined.
-
Ali Spivak (Google LLC) joined.
-
Fidel Tian (Zoom) joined.
-
-
1 participant left.
-
Geunhyung Kim (Sharing Economy Association of Korea (SEAK)) left.
-
-
2 organizations joined.
-
Sharing Economy Association of Korea (SEAK) joined.
-
Zoom joined.
-
-
1 organization left.
-
Gooroomee Inc. left.
-
GitHub
webmachinelearning/webnn
-
4 issues opened.
-
Define an upper bound on number of concat inputs and split outputs webmachinelearning/webnn#931, by @philloooo.
-
RFE: Add support for more floating point low-precision ML data types (`bfloat16`, `fp8`, `nvfp4`) webmachinelearning/webnn#930, by @mtavenrath.
-
RFE: Add batching support (N-D tensors) to the gemm operator webmachinelearning/webnn#929, by @mtavenrath.
-
Consider specify lower limit for conv2d/pool2d kernel sizes, dilations, strides webmachinelearning/webnn#928, by @philloooo.
-
-
9 issues received 23 new comments.
-
Consider specify lower limit for conv2d/pool2d kernel sizes, dilations, strides webmachinelearning/webnn#928 (11 comments by @dsharlet, @fdwr, @huningxin, @philloooo)
-
Define an upper bound on number of concat inputs and split outputs webmachinelearning/webnn#931 (5 comments by @fdwr, @philloooo)
-
RFE: Add support for more floating point low-precision ML data types (`bfloat16`, `fp8`, `nvfp4`) webmachinelearning/webnn#930 (1 comment by @fdwr)
-
Get devices used for a graph after graph compilation webmachinelearning/webnn#836 (1 comment by @handellm)
-
Mandate `output_channels % groups == 0` validation for conv2d webmachinelearning/webnn#925 (1 comment by @anssiko)
-
Support `depth_multiplier > 1` for a depthwise conv2d op webmachinelearning/webnn#353 (1 comment by @vmukhachev)
-
-
2 pull requests received 10 new comments.
-
Refactor device selection: Rename to computePolicy, remove accelerated, and add fallback webmachinelearning/webnn#923 (7 comments by @anssiko, @handellm, @huningxin, @mingmingtasd, @zolkis)
-
add minimum data types and rank range for operations webmachinelearning/webnn#910 (3 comments by @BruceDai, @fdwr)
-
TR publications
-
3 documents published.
-
Web Neural Network API published 3 times as a Candidate Recommendation Draft. History of Web Neural Network API
-
Past meetings
Participants
-
5 participants joined.
-
Yoav Weiss (Shopify) joined.
-
Tarek Ziade (Hugging Face) joined.
-
Gregory Terzian (W3C Invited Experts) joined.
-
Jaewon Lee (Google LLC) joined.
-
ninja.zx zhangxiao (ByteDance) joined.
-
-
1 participant left.
-
Tarek Ziade (Hugging Face) left.
-
GitHub
webmachinelearning/webnn
-
4 issues opened.
-
Security considerations refresh webmachinelearning/webnn#927, by @anssiko.
-
Mandate `output_channels % groups == 0` validation for conv2d webmachinelearning/webnn#925, by @lynne0326.
-
Consider specifying tensor element count limit webmachinelearning/webnn#924, by @philloooo.
-
dequantizeLinear algorithm steps should validate zeroPoint's data type is the same as input's webmachinelearning/webnn#921, by @huningxin.
-
-
2 issues closed.
-
Wide review tracker webmachinelearning/webnn#239.
-
dequantizeLinear algorithm steps should validate zeroPoint's data type is the same as input's webmachinelearning/webnn#921.
-
-
7 issues received 14 new comments.
-
Consider specifying tensor element count limit webmachinelearning/webnn#924 (4 comments by @fdwr, @philloooo)
-
accelerated should be prior to powerPreference for device selection webmachinelearning/webnn#911 (2 comments by @anssiko, @mingmingtasd)
-
Get devices used for a graph after graph compilation webmachinelearning/webnn#836 (2 comments by @anssiko, @mwyrzykowski)
-
-
3 pull requests opened.
-
Specify element count limit webmachinelearning/webnn#926, by @philloooo.
-
Refactor device selection: Rename to computePolicy, remove accelerated, and add fallback webmachinelearning/webnn#923, by @mingmingtasd.
-
Bugfix: Add missing data type validations for QDQ ops webmachinelearning/webnn#922, by @huningxin.
-
-
3 pull requests merged.
-
Specify element count limit webmachinelearning/webnn#926.
-
Restrict scale's values of dequantizeLinear and quantizeLinear to be positive and nonzero webmachinelearning/webnn#906.
-
Bugfix: Add missing data type validations for QDQ ops webmachinelearning/webnn#922.
-
-
2 pull requests closed.
-
fix reduceLogSumExp, softplus, and tanh decompositions to not overflow so easily webmachinelearning/webnn#916.
-
define graph.devices webmachinelearning/webnn#854.
-
-
5 pull requests received 18 new comments.
-
Refactor device selection: Rename to computePolicy, remove accelerated, and add fallback webmachinelearning/webnn#923 (7 comments by @fdwr, @handellm, @mwyrzykowski, @zolkis)
-
Bugfix: Add missing data type validations for QDQ ops webmachinelearning/webnn#922 (5 comments by @fdwr, @huningxin)
-
Specify element count limit webmachinelearning/webnn#926 (3 comments by @anssiko, @fdwr, @huningxin)
-
fix reduceLogSumExp, softplus, and tanh decompositions to not overflow so easily webmachinelearning/webnn#916 (2 comments by @fdwr, @vmukhachev)
-
Restrict scale's values of dequantizeLinear and quantizeLinear to be positive and nonzero webmachinelearning/webnn#906 (1 comment by @fdwr)
-
Past meetings
Participants
-
1 participant joined.
-
Christopher Hebert (NVIDIA) joined.
-
-
2 participants left.
-
Wenhe Li (W3C Invited Experts) left.
-
Philip Laszkowicz (W3C Invited Experts) left.
-
GitHub
webmachinelearning/webnn
-
2 issues opened.
-
what `N-D tensor` is? webmachinelearning/webnn#920, by @Yaffle.
-
Creating a context should be done using finer-grained in-parallel steps webmachinelearning/webnn#919, by @gterzian.
-
-
1 issue closed.
-
what `N-D tensor` is? webmachinelearning/webnn#920.
-
-
3 issues received 14 new comments.
-
accelerated should be prior to powerPreference for device selection webmachinelearning/webnn#911 (7 comments by @anssiko, @handellm, @mwyrzykowski, @zolkis)
-
Support flexible input sizes webmachinelearning/webnn#883 (5 comments by @anssiko, @huningxin, @mtavenrath, @tarekziade)
-
Clarify `instanceNormalization`'s support for 1d, 2d, 3d use cases webmachinelearning/webnn#490 (2 comments by @fdwr, @Yaffle)
-
webmachinelearning/webmachinelearning-ethics
-
1 issue received 1 new comment.
-
Snapshot Ethical thinking workshop and Ethical Risk Canvas for archival webmachinelearning/webmachinelearning-ethics#21 (1 comment by @anssiko)
-
TR publications
-
5 documents published.
-
Web Neural Network API published as a Candidate Recommendation Snapshot. History of Web Neural Network API
-
Web Neural Network API published 4 times as a Candidate Recommendation Draft. History of Web Neural Network API
-
Past meetings
Participants
-
2 participants joined.
-
Benjamin VanderSloot (Mozilla Foundation) joined.
-
Dominic Farolino (Google LLC) joined.
-
-
4 participants left.
-
Brad Triebwasser (Google LLC) left.
-
Suraj Bisht (Capital One Discover Network) left.
-
Peter Tanski (Capital One Discover Network) left.
-
Christos Bacharakis (eyeo GmbH) left.
-
-
1 organization left.
-
eyeo GmbH left.
-
GitHub
webmachinelearning/webnn
-
1 issue opened.
-
Floating point accuracy for sin and cos with range constraint webmachinelearning/webnn#914, by @philloooo.
-
-
2 issues closed.
-
Drop support of 8-bit integers input for CumulativeSum webmachinelearning/webnn#892.
-
The decomposition of `lstm` has issue for batch size 1 input webmachinelearning/webnn#889.
-
-
7 issues received 37 new comments.
-
accelerated should be prior to powerPreference for device selection webmachinelearning/webnn#911 (23 comments by @anssiko, @bbernhar, @fdwr, @handellm, @huningxin, @mingmingtasd, @mwyrzykowski, @zolkis)
-
Floating point accuracy for sin and cos with range constraint webmachinelearning/webnn#914 (6 comments by @BruceDai, @fdwr, @huningxin, @mwyrzykowski, @philloooo)
-
Device selection criteria for usecase-driven scenarios webmachinelearning/webnn#902 (1 comment by @anssiko)
-
The decomposition of `lstm` has issue for batch size 1 input webmachinelearning/webnn#889 (1 comment by @huningxin)
-
-
6 pull requests opened.
-
Add explicit start and end dates for the changelog webmachinelearning/webnn#918, by @anssiko.
-
Get back to auto-publishing CR Drafts webmachinelearning/webnn#917, by @dontcallmedom.
-
fix reduceLogSumExp, softplus, and tanh decompositions to not overflow so easily webmachinelearning/webnn#916, by @vmukhachev.
-
Add Candidate Recommendation Snapshot for staging webmachinelearning/webnn#915, by @anssiko.
-
Bugfix: avoid squeezing batch size 1 dimension in lstm and gru emulation code webmachinelearning/webnn#913, by @huningxin.
-
Drop support of 8-bit integers input for CumulativeSum webmachinelearning/webnn#912, by @huningxin.
-
-
5 pull requests merged.
-
Add explicit start and end dates for the changelog webmachinelearning/webnn#918.
-
Get back to auto-publishing CR Drafts webmachinelearning/webnn#917.
-
Add Candidate Recommendation Snapshot for staging webmachinelearning/webnn#915.
-
Drop support of 8-bit integers input for CumulativeSum webmachinelearning/webnn#912.
-
Bugfix: avoid squeezing batch size 1 dimension in lstm and gru emulation code webmachinelearning/webnn#913.
-
-
7 pull requests received 27 new comments.
-
add minimum data types and rank range for operations webmachinelearning/webnn#910 (11 comments by @BruceDai, @huningxin)
-
Restrict scale's values of dequantizeLinear and quantizeLinear to be positive and nonzero webmachinelearning/webnn#906 (7 comments by @BruceDai, @fdwr, @huningxin, @reillyeon)
-
Add Candidate Recommendation Snapshot for staging webmachinelearning/webnn#915 (5 comments by @anssiko, @dontcallmedom)
-
Add explicit start and end dates for the changelog webmachinelearning/webnn#918 (1 comment by @anssiko)
-
Clarify interpolation algorithms for resample2d webmachinelearning/webnn#816 (1 comment by @vmukhachev)
-
fix reduceLogSumExp, softplus, and tanh decompositions to not overflow so easily webmachinelearning/webnn#916 (1 comment by @w3cbot)
-