LuxLib-v1.6.1

LuxLib LuxLib-v1.6.1

Diff since LuxLib-v1.6.0

Merged pull requests:

• fix: relax input types (#1226) (@avik-pal)

v1.6.0

Lux v1.6.0

Diff since v1.5.2

Merged pull requests:

• chore: bump Zygote version (#1182) (@avik-pal)
• CompatHelper: add new compat entry for Lux at version 1 for package RealNVP, (keep existing compat) (#1221) (@github-actions[bot])

Closed issues:

• Use TestExtras.jl for inference testing (#1098)

MLDataDevices-v1.6.9

MLDataDevices MLDataDevices-v1.6.9

Diff since MLDataDevices-v1.6.8

Merged pull requests:

• feat: emit batch_norm ops from stablehlo (#1142) (@avik-pal)
• chore: bump Zygote version (#1182) (@avik-pal)
• chore: bump CairoMakie to 0.13 (#1206) (@avik-pal)
• CompatHelper: bump compat for Turing to 0.36 for package BayesianNN, (keep existing compat) (#1207) (@github-actions[bot])
• docs: add GCN Cora example (#1210) (@avik-pal)
• CompatHelper: add new compat entry for GNNGraphs at version 1 for package GCN_Cora, (keep existing compat) (#1211) (@github-actions[bot])
• CompatHelper: add new compat entry for OneHotArrays at version 0.2 for package GCN_Cora, (keep existing compat) (#1212) (@github-actions[bot])
• docs: Normalizing Flow (RealNVP) example (#1215) (@avik-pal)
• CompatHelper: add new compat entry for Lux at version 1 for package RealNVP, (keep existing compat) (#1221) (@github-actions[bot])

Closed issues:

• Use TestExtras.jl for inference testing (#1098)
• Parallel is incompatible with Zygote nested gradient (#1199)

LuxTestUtils-v1.7.1

LuxTestUtils LuxTestUtils-v1.7.1

Diff since LuxTestUtils-v1.7.0

Merged pull requests:

• feat: update ConvMixer to support reactant (#1063) (@avik-pal)
• ci: use sources for docs (#1100) (@avik-pal)
• Add Reactant and TPU to autodiff.md (#1101) (@wsmoses)
• refactor: cleanup some old pre-1.0 hacks (#1102) (@avik-pal)
• feat: add bf16 function (#1104) (@avik-pal)
• docs: add CUDA.CURAND.default_rng() to docs (#1105) (@avik-pal)
• fix: use generic broadcasting for complex numbers (#1106) (@avik-pal)
• Update exporting_to_jax.md (#1107) (@wsmoses)
• CompatHelper: bump compat for LossFunctions in [weakdeps] to 1, (keep existing compat) (#1108) (@github-actions[bot])
• Add TrainState docstring with Optimisers API (#1110) (@abhro)
• Fix markdown list in docstring (#1111) (@abhro)
• chore: bump crate-ci/typos from 1.27.3 to 1.28.1 (#1113) (@dependabot[bot])
• fix: handle debug leafs with dispatch (#1115) (@avik-pal)
• test: allow the latest AMDGPU to be installed (#1116) (@avik-pal)
• test: add unsafe_free to skip list (#1117) (@avik-pal)
• fix: use the correct dispatches for device overloads (#1118) (@avik-pal)
• test: try fixing enzyme test (#1119) (@avik-pal)
• ci(github-actions): use julia-actions/cache (#1122) (@avik-pal)
• test: re-enable flux testing (#1123) (@avik-pal)
• chore: bump minimum Reactant version (#1125) (@avik-pal)
• fix: try fixing cuda install in tests (#1126) (@avik-pal)
• CompatHelper: bump compat for GPUArraysCore to 0.2, (keep existing compat) (#1127) (@github-actions[bot])
• docs: run partial dataset only on CI (#1128) (@avik-pal)
• chore: bump crate-ci/typos from 1.28.1 to 1.28.2 (#1130) (@dependabot[bot])
• fix: preserve object when device is same (#1133) (@avik-pal)
• fix: use functors for testing wrapped arrays (#1134) (@avik-pal)
• fix: remove old patches around reactant bug (#1135) (@avik-pal)
• CompatHelper: bump compat for Flux in [weakdeps] to 0.16, (keep existing compat) (#1136) (@github-actions[bot])
• chore: bump crate-ci/typos from 1.28.2 to 1.28.3 (#1137) (@dependabot[bot])
• fix: update to new reactant changes (#1140) (@avik-pal)
• feat: emit batch_norm ops from stablehlo (#1142) (@avik-pal)
• chore: bump crate-ci/typos from 1.28.3 to 1.28.4 (#1144) (@dependabot[bot])
• don't declare implicitly exported functions public (#1147) (@simeonschaub)
• use return_type instead of _return_type (#1148) (@simeonschaub)
• feat: more nested AD rules (#1151) (@avik-pal)
• fix: update default rng for reactant (#1152) (@avik-pal)
• Change update step to thousands in PINN 2D PDE (#1153) (@abhro)
• CompatHelper: add new compat entry for ProgressTables at version 0.1 for package CIFAR10, (keep existing compat) (#1154) (@github-actions[bot])
• device(NN) should only give warning once (#1156) (@vpuri3)
• feat: conditional VAE (#1157) (@avik-pal)
• fix: ConditionalVAE on CI (#1159) (@avik-pal)
• CompatHelper: add new compat entry for Optimisers at version 0.4 for package ConditionalVAE, (keep existing compat) (#1161) (@github-actions[bot])
• CompatHelper: add new compat entry for StableRNGs at version 1 for package ConditionalVAE, (keep existing compat) (#1162) (@github-actions[bot])
• CompatHelper: add new compat entry for Comonicon at version 1 for package ConditionalVAE, (keep existing compat) (#1163) (@github-actions[bot])
• [MLDataDevices] Bump deps (#1164) (@pxl-th)
• docs: migrate most examples to Reactant (#1180) (@avik-pal)
• chore: bump Zygote version (#1182) (@avik-pal)
• chore: bump crate-ci/typos from 1.28.4 to 1.29.4 (#1183) (@dependabot[bot])
• fix: pass in RNG to shuffle (#1188) (@avik-pal)
• feat: allow no grad option for reactant (#1190) (@avik-pal)
• CompatHelper: add new compat entry for Enzyme at version 0.13 for package PINN2DPDE, (keep existing compat) (#1191) (@github-actions[bot])
• CompatHelper: add new compat entry for Reactant at version 0.2 for package PINN2DPDE, (keep existing compat) (#1192) (@github-actions[bot])
• CompatHelper: add new compat entry for Reactant at version 0.2 for package SimpleChains, (keep existing compat) (#1193) (@github-actions[bot])
• CompatHelper: add new compat entry for Reactant at version 0.2 for package SimpleRNN, (keep existing compat) (#1194) (@github-actions[bot])
• CompatHelper: bump compat for oneAPI in [weakdeps] to 2 for package MLDataDevices, (keep existing compat) (#1195) (@github-actions[bot])
• CompatHelper: bump compat for oneAPI in [weakdeps] to 2 for package WeightInitializers, (keep existing compat) (#1196) (@github-actions[bot])
• chore: bump CairoMakie to 0.13 (#1206) (@avik-pal)
• CompatHelper: bump compat for Turing to 0.36 for package BayesianNN, (keep existing compat) (#1207) (@github-actions[bot])
• docs: add GCN Cora example (#1210) (@avik-pal)
• CompatHelper: add new compat entry for GNNGraphs at version 1 for package GCN_Cora, (keep existing compat) (#1211) (@github-actions[bot])
• CompatHelper: add new compat entry for OneHotArrays at version 0.2 for package GCN_Cora, (keep existing compat) (#1212) (@github-actions[bot])
• docs: Normalizing Flow (RealNVP) example (#1215) (@avik-pal)
• CompatHelper: add new compat entry for Lux at version 1 for package RealNVP, (keep existing compat) (#1221) (@github-actions[bot])

Closed issues:

• Immutable Arrays (#8)
• Downstream Compat Updates (#880)
• Zygote + ForwardDiff support for complex differentiation (#977)
• Add CUDA.CURAND.default_rng() to the table (#1003)
• Enzyme 0.13 fails with batched matrix multiply (#1024)
• getkeypath and layer_map not fully working with model with Parallel layers (#1068)
• Re-enable Flux compatibility testing (#1070)
• [AMDGPU CI] Circular dependencies disabling precompilation (#1095)
• Use TestExtras.jl for inference testing (#1098)
• LuxTestUtils.Constant clashes with DifferentiationInterface.Constant (#1103)
• Error in trying to use Optimization.jl for LSTM training based on Lux.jl (#1114)
• Documentation Build Stalls (#1120)
• CUDA Test CI is broken (#1121)
• [MLDataDevices] devices don't preserve identity (#1129)
• Random Numbers & Reactant (#1131)
• Problem with Lux & SymbolicsLuxExt (#1132)
• How to implement a detach operation similar to Pytorch? (#1138)
• Unexpected handling of LR Schedulers in TrainState (#1143)
• Directly construct Optimiser state on Reactant buffers (#1145)
• (AbstractDevice)(x) should respect Adapt.adapt_structure (#1149)
• CUDA 2nd order AD with MaxPool and logsoftmax (#1150)
• Recurrent cells cannot be chained with other layers (#1155)
• No Grad option for TrainState single_train_step(!) (#1181)
• sparse_init doesn't use provided rng fully (#1185)
• Incorrect IR generated for some neural networks (#1186)
• WeightInitializers.DeviceAgnostic doesn't respect Reactant (#1187)
• Export utilities in partial.jl (#1189)
• Parallel is incompatible with Zygote nested gradient (#1199)

LuxLib-v1.6.0

LuxLib LuxLib-v1.6.0

Diff since LuxLib-v1.5.0

Merged pull requests:

• chore: bump Zygote version (#1182) (@avik-pal)
• docs: add GCN Cora example (#1210) (@avik-pal)
• CompatHelper: add new compat entry for GNNGraphs at version 1 for package GCN_Cora, (keep existing compat) (#1211) (@github-actions[bot])
• CompatHelper: add new compat entry for OneHotArrays at version 0.2 for package GCN_Cora, (keep existing compat) (#1212) (@github-actions[bot])
• docs: Normalizing Flow (RealNVP) example (#1215) (@avik-pal)
• CompatHelper: add new compat entry for Lux at version 1 for package RealNVP, (keep existing compat) (#1221) (@github-actions[bot])

Closed issues:

• Use TestExtras.jl for inference testing (#1098)

v1.5.2

Lux v1.5.2

Diff since v1.5.1

Merged pull requests:

• feat: emit batch_norm ops from stablehlo (#1142) (@avik-pal)
• chore: bump CairoMakie to 0.13 (#1206) (@avik-pal)
• CompatHelper: bump compat for Turing to 0.36 for package BayesianNN, (keep existing compat) (#1207) (@github-actions[bot])
• docs: add GCN Cora example (#1210) (@avik-pal)
• CompatHelper: add new compat entry for GNNGraphs at version 1 for package GCN_Cora, (keep existing compat) (#1211) (@github-actions[bot])
• CompatHelper: add new compat entry for OneHotArrays at version 0.2 for package GCN_Cora, (keep existing compat) (#1212) (@github-actions[bot])
• docs: Normalizing Flow (RealNVP) example (#1215) (@avik-pal)

Closed issues:

• Parallel is incompatible with Zygote nested gradient (#1199)

LuxLib-v1.5.0

LuxLib LuxLib-v1.5.0

Diff since LuxLib-v1.4.1

Merged pull requests:

• CompatHelper: bump compat for GPUArraysCore to 0.2, (keep existing compat) (#1127) (@github-actions[bot])
• feat: emit batch_norm ops from stablehlo (#1142) (@avik-pal)
• feat: allow no grad option for reactant (#1190) (@avik-pal)
• CompatHelper: add new compat entry for Enzyme at version 0.13 for package PINN2DPDE, (keep existing compat) (#1191) (@github-actions[bot])
• CompatHelper: add new compat entry for Reactant at version 0.2 for package PINN2DPDE, (keep existing compat) (#1192) (@github-actions[bot])
• CompatHelper: add new compat entry for Reactant at version 0.2 for package SimpleChains, (keep existing compat) (#1193) (@github-actions[bot])
• CompatHelper: add new compat entry for Reactant at version 0.2 for package SimpleRNN, (keep existing compat) (#1194) (@github-actions[bot])
• CompatHelper: bump compat for oneAPI in [weakdeps] to 2 for package MLDataDevices, (keep existing compat) (#1195) (@github-actions[bot])
• CompatHelper: bump compat for oneAPI in [weakdeps] to 2 for package WeightInitializers, (keep existing compat) (#1196) (@github-actions[bot])
• chore: bump CairoMakie to 0.13 (#1206) (@avik-pal)
• CompatHelper: bump compat for Turing to 0.36 for package BayesianNN, (keep existing compat) (#1207) (@github-actions[bot])

Closed issues:

• No Grad option for TrainState single_train_step(!) (#1181)
• Parallel is incompatible with Zygote nested gradient (#1199)

v1.5.1

Lux v1.5.1

Diff since v1.5.0

Merged pull requests:

• CompatHelper: bump compat for GPUArraysCore to 0.2, (keep existing compat) (#1127) (@github-actions[bot])
• CompatHelper: add new compat entry for Enzyme at version 0.13 for package PINN2DPDE, (keep existing compat) (#1191) (@github-actions[bot])
• CompatHelper: add new compat entry for Reactant at version 0.2 for package PINN2DPDE, (keep existing compat) (#1192) (@github-actions[bot])
• CompatHelper: add new compat entry for Reactant at version 0.2 for package SimpleChains, (keep existing compat) (#1193) (@github-actions[bot])
• CompatHelper: add new compat entry for Reactant at version 0.2 for package SimpleRNN, (keep existing compat) (#1194) (@github-actions[bot])
• CompatHelper: bump compat for oneAPI in [weakdeps] to 2 for package MLDataDevices, (keep existing compat) (#1195) (@github-actions[bot])
• CompatHelper: bump compat for oneAPI in [weakdeps] to 2 for package WeightInitializers, (keep existing compat) (#1196) (@github-actions[bot])

WeightInitializers-v1.1.1

WeightInitializers WeightInitializers-v1.1.1

Diff since WeightInitializers-v1.1.0

Merged pull requests:

• CompatHelper: bump compat for GPUArraysCore to 0.2, (keep existing compat) (#1127) (@github-actions[bot])
• feat: allow no grad option for reactant (#1190) (@avik-pal)
• CompatHelper: add new compat entry for Enzyme at version 0.13 for package PINN2DPDE, (keep existing compat) (#1191) (@github-actions[bot])
• CompatHelper: add new compat entry for Reactant at version 0.2 for package PINN2DPDE, (keep existing compat) (#1192) (@github-actions[bot])
• CompatHelper: add new compat entry for Reactant at version 0.2 for package SimpleChains, (keep existing compat) (#1193) (@github-actions[bot])
• CompatHelper: add new compat entry for Reactant at version 0.2 for package SimpleRNN, (keep existing compat) (#1194) (@github-actions[bot])
• CompatHelper: bump compat for oneAPI in [weakdeps] to 2 for package MLDataDevices, (keep existing compat) (#1195) (@github-actions[bot])
• CompatHelper: bump compat for oneAPI in [weakdeps] to 2 for package WeightInitializers, (keep existing compat) (#1196) (@github-actions[bot])

Closed issues:

• No Grad option for TrainState single_train_step(!) (#1181)

MLDataDevices-v1.6.8

MLDataDevices MLDataDevices-v1.6.8

Diff since MLDataDevices-v1.6.7

Merged pull requests:

• CompatHelper: bump compat for GPUArraysCore to 0.2, (keep existing compat) (#1127) (@github-actions[bot])
• docs: migrate most examples to Reactant (#1180) (@avik-pal)
• chore: bump crate-ci/typos from 1.28.4 to 1.29.4 (#1183) (@dependabot[bot])
• fix: pass in RNG to shuffle (#1188) (@avik-pal)
• feat: allow no grad option for reactant (#1190) (@avik-pal)
• CompatHelper: add new compat entry for Enzyme at version 0.13 for package PINN2DPDE, (keep existing compat) (#1191) (@github-actions[bot])
• CompatHelper: add new compat entry for Reactant at version 0.2 for package PINN2DPDE, (keep existing compat) (#1192) (@github-actions[bot])
• CompatHelper: add new compat entry for Reactant at version 0.2 for package SimpleChains, (keep existing compat) (#1193) (@github-actions[bot])
• CompatHelper: add new compat entry for Reactant at version 0.2 for package SimpleRNN, (keep existing compat) (#1194) (@github-actions[bot])
• CompatHelper: bump compat for oneAPI in [weakdeps] to 2 for package MLDataDevices, (keep existing compat) (#1195) (@github-actions[bot])
• CompatHelper: bump compat for oneAPI in [weakdeps] to 2 for package WeightInitializers, (keep existing compat) (#1196) (@github-actions[bot])

Closed issues:

• No Grad option for TrainState single_train_step(!) (#1181)
• sparse_init doesn't use provided rng fully (#1185)
• Incorrect IR generated for some neural networks (#1186)
• WeightInitializers.DeviceAgnostic doesn't respect Reactant (#1187)
• Export utilities in partial.jl (#1189)