Update chip.rs

---------

Co-authored-by: dante <45801863+alexander-camuto@users.noreply.github.com>

.github/workflows/benchmarks.yml

@@ -8,6 +8,8 @@ on:
 jobs:
 
   bench_poseidon:
+    permissions:
+      contents: read
     runs-on: self-hosted
     steps:
       - uses: actions/checkout@v4
@@ -22,6 +24,8 @@ jobs:
         run: cargo bench --verbose --bench poseidon
 
   bench_einsum_accum_matmul:
+    permissions:
+      contents: read
     runs-on: self-hosted
     needs: [bench_poseidon]
     steps:
@@ -37,6 +41,8 @@ jobs:
         run: cargo bench --verbose --bench accum_einsum_matmul
 
   bench_accum_matmul_relu:
+    permissions:
+      contents: read
     runs-on: self-hosted
     needs: [bench_poseidon]
     steps:
@@ -52,6 +58,8 @@ jobs:
         run: cargo bench --verbose --bench accum_matmul_relu
 
   bench_accum_matmul_relu_overflow:
+    permissions:
+      contents: read
     runs-on: self-hosted
     needs: [bench_poseidon]
     steps:
@@ -67,6 +75,8 @@ jobs:
         run: cargo bench --verbose --bench accum_matmul_relu_overflow
 
   bench_relu:
+    permissions:
+      contents: read
     runs-on: self-hosted
     needs: [bench_poseidon]
     steps:
@@ -82,6 +92,8 @@ jobs:
         run: cargo bench --verbose --bench relu
 
   bench_accum_dot:
+    permissions:
+      contents: read
     runs-on: self-hosted
     needs: [bench_poseidon]
     steps:
@@ -97,6 +109,8 @@ jobs:
         run: cargo bench --verbose --bench accum_dot
 
   bench_accum_conv:
+    permissions:
+      contents: read
     runs-on: self-hosted
     needs: [bench_poseidon]
     steps:
@@ -112,6 +126,8 @@ jobs:
         run: cargo bench --verbose --bench accum_conv
 
   bench_accum_sumpool:
+    permissions:
+      contents: read
     runs-on: self-hosted
     needs: [bench_poseidon]
     steps:
@@ -127,6 +143,8 @@ jobs:
         run: cargo bench --verbose --bench accum_sumpool
 
   bench_pairwise_add:
+    permissions:
+      contents: read
     runs-on: self-hosted
     needs: [bench_poseidon]
     steps:
@@ -142,6 +160,8 @@ jobs:
         run: cargo bench --verbose --bench pairwise_add
 
   bench_accum_sum:
+    permissions:
+      contents: read
     runs-on: self-hosted
     needs: [bench_poseidon]
     steps:
@@ -157,6 +177,8 @@ jobs:
         run: cargo bench --verbose --bench accum_sum
 
   bench_pairwise_pow:
+    permissions:
+      contents: read
     runs-on: self-hosted
     needs: [bench_poseidon]
     steps:

.github/workflows/engine.yml

@@ -15,6 +15,9 @@ defaults:
     working-directory: .
 jobs:
   publish-wasm-bindings:
+    permissions:
+      contents: read
+      packages: write
     name: publish-wasm-bindings
     runs-on: ubuntu-latest
     if: startsWith(github.ref, 'refs/tags/')
@@ -186,6 +189,9 @@ jobs:
 
 
   in-browser-evm-ver-publish:
+    permissions:
+      contents: read
+      packages: write
     name: publish-in-browser-evm-verifier-package
     needs: [publish-wasm-bindings]
     runs-on: ubuntu-latest

.github/workflows/large-tests.yml

@@ -6,6 +6,8 @@ on:
         description: "Test scenario tags"
 jobs:
   large-tests:
+    permissions:
+      contents: read
     runs-on: kaiju
     steps:
       - uses: actions/checkout@v4

.github/workflows/pypi-gpu.yml

@@ -18,6 +18,9 @@ defaults:
 jobs:
 
   linux:
+    permissions:
+      contents: read
+      packages: write
     runs-on: GPU
     strategy:
       matrix:

.github/workflows/pypi.yml

@@ -16,6 +16,8 @@ defaults:
 
 jobs:
   macos:
+    permissions:
+      contents: read
     runs-on: macos-latest
     if: startsWith(github.ref, 'refs/tags/')
     strategy:
@@ -47,6 +49,13 @@ jobs:
           components: rustfmt, clippy
 
       - name: Build wheels
+        if: matrix.target == 'universal2-apple-darwin'
+        uses: PyO3/maturin-action@v1
+        with:
+          target: ${{ matrix.target }}
+          args: --release --out dist --features python-bindings
+      - name: Build wheels
+        if: matrix.target == 'x86_64'
         uses: PyO3/maturin-action@v1
         with:
           target: ${{ matrix.target }}
@@ -64,6 +73,8 @@ jobs:
           path: dist
 
   windows:
+    permissions:
+      contents: read
     runs-on: windows-latest
     if: startsWith(github.ref, 'refs/tags/')
     strategy:
@@ -111,6 +122,8 @@ jobs:
           path: dist
 
   linux:
+    permissions:
+      contents: read
     runs-on: ubuntu-latest
     if: startsWith(github.ref, 'refs/tags/')
     strategy:
@@ -226,6 +239,8 @@ jobs:
   #       path: dist
 
   musllinux:
+    permissions:
+      contents: read
     runs-on: ubuntu-latest
     if: startsWith(github.ref, 'refs/tags/')
     strategy:
@@ -291,6 +306,8 @@ jobs:
           path: dist
 
   musllinux-cross:
+    permissions:
+      contents: read
     runs-on: ubuntu-latest
     if: startsWith(github.ref, 'refs/tags/')
     strategy:
@@ -378,6 +395,8 @@ jobs:
           packages-dir: ./
 
   doc-publish:
+    permissions:
+      contents: read
     name: Trigger ReadTheDocs Build
     runs-on: ubuntu-latest
     needs: pypi-publish

.github/workflows/release.yml

@@ -10,6 +10,9 @@ on:
       - "*"
 jobs:
   create-release:
+    permissions:
+      contents: read
+      packages: write
     name: create-release
     runs-on: ubuntu-22.04
     if: startsWith(github.ref, 'refs/tags/')
@@ -33,6 +36,9 @@ jobs:
           tag_name: ${{ env.EZKL_VERSION }}
 
   build-release-gpu:
+    permissions:
+      contents: read
+      packages: write
     name: build-release-gpu
     needs: ["create-release"]
     runs-on: GPU
@@ -94,6 +100,10 @@ jobs:
           asset_content_type: application/octet-stream
 
   build-release:
+    permissions:
+      contents: read
+      packages: write
+      issues: write
     name: build-release
     needs: ["create-release"]
     runs-on: ${{ matrix.os }}
@@ -186,14 +196,18 @@ jobs:
           echo "target flag is: ${{ env.TARGET_FLAGS }}"
           echo "target dir is: ${{ env.TARGET_DIR }}"
 
-      - name: Build release binary (no asm)
-        if: matrix.build != 'linux-gnu'
+      - name: Build release binary (no asm or metal) 
+        if: matrix.build != 'linux-gnu' && matrix.build != 'macos-aarch64'
         run: ${{ env.CARGO }} build --release ${{ env.TARGET_FLAGS }} -Z sparse-registry
 
       - name: Build release binary (asm)
         if: matrix.build == 'linux-gnu'
         run: ${{ env.CARGO }} build --release ${{ env.TARGET_FLAGS }} -Z sparse-registry --features asm
 
+      - name: Build release binary (metal)
+        if: matrix.build == 'macos-aarch64'
+        run: ${{ env.CARGO }} build --release ${{ env.TARGET_FLAGS }} -Z sparse-registry --features macos-metal
+
       - name: Strip release binary
         if: matrix.build != 'windows-msvc' && matrix.build != 'linux-aarch64'
         run: strip "target/${{ matrix.target }}/release/ezkl"

.github/workflows/rust.yml

@@ -19,9 +19,27 @@ env:
   CARGO_TERM_COLOR: always
 
 jobs:
-  build:
-    runs-on: ubuntu-latest
 
+  fr-age-test:
+    permissions:
+      contents: read
+    runs-on: large-self-hosted
+    steps:
+      - uses: actions/checkout@v4
+        with:
+          persist-credentials: false
+      - uses: actions-rs/toolchain@v1
+        with:
+          toolchain: nightly-2024-07-18
+          override: true
+          components: rustfmt, clippy
+      - name: fr age Mock
+        run: cargo test --release --verbose tests::large_mock_::large_tests_6_expects -- --include-ignored
+
+  build:
+    permissions:
+      contents: read
+    runs-on: ubuntu-latest
     steps:
       - uses: actions/checkout@v4
         with:
@@ -35,6 +53,8 @@ jobs:
         run: cargo build --verbose
 
   docs:
+    permissions:
+      contents: read
     runs-on: ubuntu-latest
     steps:
       - uses: actions/checkout@v4
@@ -49,6 +69,8 @@ jobs:
         run: cargo doc --verbose
 
   library-tests:
+    permissions:
+      contents: read
     runs-on: ubuntu-latest-32-cores
     steps:
       - uses: actions/checkout@v4
@@ -109,6 +131,8 @@ jobs:
   #       run: cargo nextest run conv_relu_col_ultra_overflow --no-capture --features icicle -- --include-ignored
 
   ultra-overflow-tests_og-lookup:
+    permissions:
+      contents: read
     runs-on: non-gpu
     steps:
       - uses: actions/checkout@v4
@@ -144,6 +168,8 @@ jobs:
         run: cargo nextest run  --release conv_relu_col_ultra_overflow --no-capture --no-default-features --features ezkl -- --include-ignored
 
   ultra-overflow-tests:
+    permissions:
+      contents: read
     runs-on: non-gpu
     steps:
       - uses: actions/checkout@v4
@@ -179,6 +205,8 @@ jobs:
         run: cargo nextest run  --release conv_relu_col_ultra_overflow --no-capture -- --include-ignored
 
   model-serialization:
+    permissions:
+      contents: read
     runs-on: ubuntu-latest-16-cores
     steps:
       - uses: actions/checkout@v4
@@ -197,6 +225,8 @@ jobs:
         run: cargo nextest run native_tests::tests::model_serialization_different_binaries_  --test-threads 1
 
   wasm32-tests:
+    permissions:
+      contents: read
     runs-on: ubuntu-latest
     steps:
       - uses: actions/checkout@v4
@@ -224,6 +254,8 @@ jobs:
         run: wasm-pack test --chrome --headless -- -Z build-std="panic_abort,std" --features web
 
   mock-proving-tests:
+    permissions:
+      contents: read
     runs-on: non-gpu
     needs: [build, library-tests, docs, python-tests, python-integration-tests]
     steps:
@@ -289,6 +321,8 @@ jobs:
         run: cargo nextest run --release --verbose tests::mock_fixed_params_ --test-threads 32
 
   prove-and-verify-evm-tests:
+    permissions:
+      contents: read
     runs-on: non-gpu
     needs: [build, library-tests, docs, python-tests, python-integration-tests]
     steps:
@@ -371,7 +405,44 @@ jobs:
       - name: KZG prove and verify tests (EVM + hashed outputs)
         run: cargo nextest run --release --verbose tests_evm::kzg_evm_hashed_output_prove_and_verify --test-threads 1
 
+  # prove-and-verify-tests-metal:
+  #   permissions:
+  #     contents: read
+  #   runs-on: macos-13
+  #   # needs: [build, library-tests, docs]
+  #   steps:
+  #     - uses: actions/checkout@v4
+  #       with:
+  #         persist-credentials: false
+  #     - uses: actions-rs/toolchain@v1
+  #       with:
+  #         toolchain: nightly-2024-07-18
+  #         override: true
+  #         components: rustfmt, clippy
+  #     - uses: jetli/wasm-pack-action@v0.4.0
+  #       with:
+  #         # Pin to version 0.12.1
+  #         version: 'v0.12.1'
+  #     - name: Add rust-src
+  #       run: rustup component add rust-src --toolchain nightly-2024-07-18
+  #     - uses: actions/checkout@v3
+  #       with:
+  #         persist-credentials: false
+  #     - name: Use pnpm 8
+  #       uses: pnpm/action-setup@v2
+  #       with:
+  #         version: 8
+  #     - uses: baptiste0928/cargo-install@v1
+  #       with:
+  #         crate: cargo-nextest
+  #         locked: true
+  #     - name: KZG prove and verify tests (public outputs)
+  #       run: cargo nextest run --release --features macos-metal --verbose tests::kzg_prove_and_verify_::t --no-capture
+
+
   prove-and-verify-tests:
+    permissions:
+      contents: read
     runs-on: non-gpu
     needs: [build, library-tests, docs]
     steps:
@@ -489,6 +560,8 @@ jobs:
   #       run: cargo nextest run --release --verbose tests::kzg_prove_and_verify_hashed --features icicle  --test-threads 1
 
   prove-and-verify-mock-aggr-tests:
+    permissions:
+      contents: read
     runs-on: self-hosted
     needs: [build, library-tests, docs, python-tests, python-integration-tests]
     steps:
@@ -528,6 +601,8 @@ jobs:
   #       run: cargo nextest run --verbose tests_aggr::kzg_aggr_prove_and_verify_  --features icicle --test-threads 1 -- --include-ignored
 
   prove-and-verify-aggr-tests:
+    permissions:
+      contents: read
     runs-on: large-self-hosted
     needs: [build, library-tests, docs, python-tests, python-integration-tests]
     steps:
@@ -547,6 +622,8 @@ jobs:
         run: cargo nextest run --release --verbose tests_aggr::kzg_aggr_prove_and_verify_  --test-threads 4 -- --include-ignored
 
   prove-and-verify-aggr-evm-tests:
+    permissions:
+      contents: read
     runs-on: large-self-hosted
     needs: [build, library-tests, docs, python-tests, python-integration-tests]
     steps:
@@ -570,6 +647,8 @@ jobs:
         run: cargo nextest run  --release --verbose tests_evm::kzg_evm_aggr_prove_and_verify_::t --test-threads 4 -- --include-ignored
 
   examples:
+    permissions:
+      contents: read
     runs-on: ubuntu-latest-32-cores
     needs: [build, library-tests, docs]
     steps:
@@ -589,6 +668,8 @@ jobs:
         run: cargo nextest run --release tests_examples
 
   python-tests:
+    permissions:
+      contents: read
     runs-on: non-gpu
     needs: [build, library-tests, docs]
     steps:
@@ -617,7 +698,9 @@ jobs:
         run: source .env/bin/activate; pip install pytest-asyncio; pytest -vv
 
   accuracy-measurement-tests:
-    runs-on: ubuntu-latest-32-cores
+    permissions:
+      contents: read
+    runs-on: non-gpu
     needs: [build, library-tests, docs, python-tests, python-integration-tests]
     steps:
       - uses: actions/checkout@v4
@@ -649,6 +732,8 @@ jobs:
         run: source .env/bin/activate; cargo nextest run --release --verbose tests::resources_accuracy_measurement_public_outputs_
 
   python-integration-tests:
+    permissions:
+      contents: read
     runs-on: large-self-hosted
     services:
       # Label used to access the service container
@@ -694,6 +779,8 @@ jobs:
         run: python -m venv .env --clear; source .env/bin/activate; pip install -r requirements.txt; python -m ensurepip --upgrade
       - name: Build python ezkl
         run: source .env/bin/activate; unset CONDA_PREFIX; maturin develop --features python-bindings --release
+      - name: Voice tutorial
+        run: source .env/bin/activate; cargo nextest run py_tests::tests::voice_
       - name: Neural bow
         run: source .env/bin/activate; cargo nextest run py_tests::tests::neural_bag_of_words_ --no-capture
       - name: Felt conversion
@@ -713,14 +800,14 @@ jobs:
       #     chmod 600 /home/ubuntu/.kaggle/kaggle.json
       - name: All notebooks
         run: source .env/bin/activate; cargo nextest run py_tests::tests::run_notebook_ --test-threads 1
-      - name: Voice tutorial
-        run: source .env/bin/activate; cargo nextest run py_tests::tests::voice_
       - name: NBEATS tutorial
         run: source .env/bin/activate; cargo nextest run py_tests::tests::nbeats_
       # - name: Reusable verifier tutorial
         # run: source .env/bin/activate; cargo nextest run py_tests::tests::reusable_
 
   ios-integration-tests:
+      permissions:
+        contents: read
       runs-on: macos-latest
       steps:
         - uses: actions/checkout@v4
@@ -739,6 +826,8 @@ jobs:
           run: CARGO_BUILD_TARGET=aarch64-apple-darwin RUSTUP_TOOLCHAIN=nightly-2024-07-18-aarch64-apple-darwin cargo test --test ios_integration_tests --features ios-bindings-test --no-default-features
 
   swift-package-tests:
+      permissions:
+        contents: read
       runs-on: macos-latest
       needs: [ios-integration-tests]
 

.github/workflows/static-analysis.yml

@@ -8,8 +8,9 @@ on:
 
 jobs:
   analyze:
+    permissions:
+      contents: read
     runs-on: ubuntu-latest
-
     steps:
       - uses: actions/checkout@v4
         with:

.github/workflows/swift-pm.yml

@@ -9,6 +9,9 @@ on:
 
 jobs:
   build-and-update:
+    permissions:
+      contents: read
+      packages: write
     runs-on: macos-latest
     env:
       EZKL_SWIFT_PACKAGE_REPO: github.com/zkonduit/ezkl-swift-package.git

Cargo.lock

@@ -1835,6 +1835,16 @@ dependencies = [
  "syn 2.0.90",
 ]
 
+[[package]]
+name = "env_filter"
+version = "0.1.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "186e05a59d4c50738528153b83b0b0194d3a29507dfec16eccd4b342903397d0"
+dependencies = [
+ "log",
+ "regex",
+]
+
 [[package]]
 name = "env_logger"
 version = "0.10.2"
@@ -1848,6 +1858,19 @@ dependencies = [
  "termcolor",
 ]
 
+[[package]]
+name = "env_logger"
+version = "0.11.6"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "dcaee3d8e3cfc3fd92428d477bc97fc29ec8716d180c0d74c643bb26166660e0"
+dependencies = [
+ "anstream",
+ "anstyle",
+ "env_filter",
+ "humantime",
+ "log",
+]
+
 [[package]]
 name = "equivalent"
 version = "1.0.1"
@@ -1923,7 +1946,7 @@ dependencies = [
  "console_error_panic_hook",
  "criterion 0.5.1",
  "ecc",
- "env_logger",
+ "env_logger 0.10.2",
  "ethabi",
  "foundry-compilers",
  "gag",
@@ -1931,7 +1954,7 @@ dependencies = [
  "halo2_gadgets",
  "halo2_proofs",
  "halo2_solidity_verifier",
- "halo2curves 0.7.0",
+ "halo2curves 0.7.0 (git+https://github.com/privacy-scaling-explorations/halo2curves?rev=b753a832e92d5c86c5c997327a9cf9de86a18851)",
  "hex",
  "indicatif",
  "instant",
@@ -1939,7 +1962,6 @@ dependencies = [
  "lazy_static",
  "log",
  "maybe-rayon",
- "metal",
  "mimalloc",
  "mnist",
  "num",
@@ -2394,14 +2416,14 @@ dependencies = [
 [[package]]
 name = "halo2_proofs"
 version = "0.3.0"
-source = "git+https://github.com/zkonduit/halo2#ee4e1a09ebdb1f79f797685b78951c6034c430a6#ee4e1a09ebdb1f79f797685b78951c6034c430a6"
+source = "git+https://github.com/zkonduit/halo2#bf9d0057a82443be48c4779bbe14961c18fb5996#bf9d0057a82443be48c4779bbe14961c18fb5996"
 dependencies = [
  "bincode",
  "blake2b_simd",
- "env_logger",
+ "env_logger 0.10.2",
  "ff",
  "group",
- "halo2curves 0.7.0",
+ "halo2curves 0.7.0 (git+https://github.com/privacy-scaling-explorations/halo2curves?rev=b753a832e92d5c86c5c997327a9cf9de86a18851)",
  "icicle-bn254",
  "icicle-core",
  "icicle-cuda-runtime",
@@ -2409,6 +2431,7 @@ dependencies = [
  "lazy_static",
  "log",
  "maybe-rayon",
+ "mopro-msm",
  "rand_chacha",
  "rand_core 0.6.4",
  "rustc-hash 2.0.0",
@@ -2497,13 +2520,14 @@ dependencies = [
 [[package]]
 name = "halo2curves"
 version = "0.7.0"
-source = "git+https://github.com/privacy-scaling-explorations/halo2curves?rev=b753a832e92d5c86c5c997327a9cf9de86a18851#b753a832e92d5c86c5c997327a9cf9de86a18851"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "d380afeef3f1d4d3245b76895172018cfb087d9976a7cabcd5597775b2933e07"
 dependencies = [
  "blake2",
  "digest 0.10.7",
  "ff",
  "group",
- "halo2derive",
+ "halo2derive 0.1.0 (registry+https://github.com/rust-lang/crates.io-index)",
  "hex",
  "lazy_static",
  "num-bigint",
@@ -2523,6 +2547,49 @@ dependencies = [
  "unroll",
 ]
 
+[[package]]
+name = "halo2curves"
+version = "0.7.0"
+source = "git+https://github.com/privacy-scaling-explorations/halo2curves?rev=b753a832e92d5c86c5c997327a9cf9de86a18851#b753a832e92d5c86c5c997327a9cf9de86a18851"
+dependencies = [
+ "blake2",
+ "digest 0.10.7",
+ "ff",
+ "group",
+ "halo2derive 0.1.0 (git+https://github.com/privacy-scaling-explorations/halo2curves?rev=b753a832e92d5c86c5c997327a9cf9de86a18851)",
+ "hex",
+ "lazy_static",
+ "num-bigint",
+ "num-integer",
+ "num-traits",
+ "pairing",
+ "pasta_curves",
+ "paste",
+ "rand 0.8.5",
+ "rand_core 0.6.4",
+ "rayon",
+ "serde",
+ "serde_arrays",
+ "sha2",
+ "static_assertions",
+ "subtle",
+ "unroll",
+]
+
+[[package]]
+name = "halo2derive"
+version = "0.1.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "bdb99e7492b4f5ff469d238db464131b86c2eaac814a78715acba369f64d2c76"
+dependencies = [
+ "num-bigint",
+ "num-integer",
+ "num-traits",
+ "proc-macro2",
+ "quote",
+ "syn 1.0.109",
+]
+
 [[package]]
 name = "halo2derive"
 version = "0.1.0"
@@ -3283,7 +3350,8 @@ dependencies = [
 [[package]]
 name = "metal"
 version = "0.29.0"
-source = "git+https://github.com/gfx-rs/metal-rs#0e1918b34689c4b8cd13a43372f9898680547ee9"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "7ecfd3296f8c56b7c1f6fbac3c71cefa9d78ce009850c45000015f206dc7fa21"
 dependencies = [
  "bitflags 2.5.0",
  "block",
@@ -3354,6 +3422,28 @@ dependencies = [
  "byteorder",
 ]
 
+[[package]]
+name = "mopro-msm"
+version = "0.1.0"
+source = "git+https://github.com/zkonduit/metal-msm-gpu-acceleration.git#be5f647b1a6c1a6ea9024390744a2b4d87f5d002"
+dependencies = [
+ "bincode",
+ "env_logger 0.11.6",
+ "halo2curves 0.7.0 (registry+https://github.com/rust-lang/crates.io-index)",
+ "instant",
+ "itertools 0.13.0",
+ "lazy_static",
+ "log",
+ "metal",
+ "objc",
+ "once_cell",
+ "rand 0.8.5",
+ "rayon",
+ "serde",
+ "thiserror",
+ "walkdir",
+]
+
 [[package]]
 name = "native-tls"
 version = "0.2.11"

Cargo.toml

@@ -91,7 +91,6 @@ pyo3-async-runtimes = { git = "https://github.com/PyO3/pyo3-async-runtimes", ver
 pyo3-log = { version = "0.12.0", default-features = false, optional = true }
 tract-onnx = { git = "https://github.com/sonos/tract/", rev = "37132e0397d0a73e5bd3a8615d932dabe44f6736", default-features = false, optional = true }
 tabled = { version = "0.12.0", optional = true }
-metal = { git = "https://github.com/gfx-rs/metal-rs", optional = true }
 objc = { version = "0.2.4", optional = true }
 mimalloc = { version = "0.1", optional = true }
 pyo3-stub-gen = { version = "0.6.0", optional = true }
@@ -277,13 +276,14 @@ icicle = ["halo2_proofs/icicle_gpu"]
 empty-cmd = []
 no-banner = []
 no-update = []
-
+macos-metal = ["halo2_proofs/macos"]
+ios-metal = ["halo2_proofs/ios"]
 
 [patch.'https://github.com/zkonduit/halo2']
-halo2_proofs = { git = "https://github.com/zkonduit/halo2#ee4e1a09ebdb1f79f797685b78951c6034c430a6", package = "halo2_proofs" }
+halo2_proofs = { git = "https://github.com/zkonduit/halo2#bf9d0057a82443be48c4779bbe14961c18fb5996", package = "halo2_proofs" }
 
 [patch.'https://github.com/zkonduit/halo2#0654e92bdf725fd44d849bfef3643870a8c7d50b']
-halo2_proofs = { git = "https://github.com/zkonduit/halo2#ee4e1a09ebdb1f79f797685b78951c6034c430a6", package = "halo2_proofs" }
+halo2_proofs = { git = "https://github.com/zkonduit/halo2#bf9d0057a82443be48c4779bbe14961c18fb5996", package = "halo2_proofs" }
 
 [patch.crates-io]
 uniffi_testing = { git = "https://github.com/ElusAegis/uniffi-rs", branch = "feat/testing-feature-build-fix" }

examples/notebooks/mnist_gan_proof_splitting.ipynb

@@ -453,18 +453,18 @@
             "outputs": [],
             "source": [
                 "# now mock aggregate the proofs\n",
-                "proofs = []\n",
-                "for i in range(3):\n",
-                "    proof_path = os.path.join('proof_split_'+str(i)+'.json')\n",
-                "    proofs.append(proof_path)\n",
+                "# proofs = []\n",
+                "# for i in range(3):\n",
+                "#     proof_path = os.path.join('proof_split_'+str(i)+'.json')\n",
+                "#     proofs.append(proof_path)\n",
                 "\n",
-                "ezkl.mock_aggregate(proofs, logrows=23, split_proofs = True)"
+                "# ezkl.mock_aggregate(proofs, logrows=26, split_proofs = True)"
             ]
         }
     ],
     "metadata": {
         "kernelspec": {
-            "display_name": "ezkl",
+            "display_name": ".env",
             "language": "python",
             "name": "python3"
         },
@@ -478,7 +478,7 @@
             "name": "python",
             "nbconvert_exporter": "python",
             "pygments_lexer": "ipython3",
-            "version": "3.12.5"
+            "version": "3.12.7"
         },
         "orig_nbformat": 4
     },

src/circuit/modules/poseidon.rs

@@ -100,9 +100,6 @@ impl<S: Spec<Fp, WIDTH, RATE> + Sync, const WIDTH: usize, const RATE: usize, con
         let rc_a = (0..WIDTH).map(|_| meta.fixed_column()).collect::<Vec<_>>();
         let rc_b = (0..WIDTH).map(|_| meta.fixed_column()).collect::<Vec<_>>();
 
-        for input in hash_inputs.iter().take(WIDTH) {
-            meta.enable_equality(*input);
-        }
         meta.enable_constant(rc_b[0]);
 
         Self::configure_with_cols(
@@ -152,9 +149,6 @@ impl<S: Spec<Fp, WIDTH, RATE> + Sync, const WIDTH: usize, const RATE: usize, con
         let rc_a = (0..WIDTH).map(|_| meta.fixed_column()).collect::<Vec<_>>();
         let rc_b = (0..WIDTH).map(|_| meta.fixed_column()).collect::<Vec<_>>();
 
-        for input in hash_inputs.iter().take(WIDTH) {
-            meta.enable_equality(*input);
-        }
         meta.enable_constant(rc_b[0]);
 
         let instance = meta.instance_column();

src/circuit/ops/chip.rs

@@ -215,15 +215,16 @@ impl DynamicLookups {
 
 /// A struct representing the selectors for the dynamic lookup tables
 #[derive(Clone, Debug, Default)]
+
 pub struct Shuffles {
     /// [Selector]s generated when configuring the layer. We use a [BTreeMap] as we expect to configure many dynamic lookup ops.
     pub input_selectors: BTreeMap<(usize, (usize, usize)), Selector>,
     /// Selectors for the dynamic lookup tables
-    pub reference_selectors: Vec<Selector>,
+    pub output_selectors: Vec<Selector>,
     /// Inputs:
     pub inputs: Vec<VarTensor>,
     /// tables
-    pub references: Vec<VarTensor>,
+    pub outputs: Vec<VarTensor>,
 }
 
 impl Shuffles {
@@ -234,9 +235,13 @@ impl Shuffles {
 
         Self {
             input_selectors: BTreeMap::new(),
-            reference_selectors: vec![],
-            inputs: vec![dummy_var.clone(), dummy_var.clone()],
-            references: vec![single_col_dummy_var.clone(), single_col_dummy_var.clone()],
+            output_selectors: vec![],
+            inputs: vec![dummy_var.clone(), dummy_var.clone(), dummy_var.clone()],
+            outputs: vec![
+                single_col_dummy_var.clone(),
+                single_col_dummy_var.clone(),
+                single_col_dummy_var.clone(),
+            ],
         }
     }
 }
@@ -374,6 +379,12 @@ impl<F: PrimeField + TensorType + PartialOrd + std::hash::Hash> BaseConfig<F> {
         if inputs[0].num_cols() != output.num_cols() {
             log::warn!("input and output shapes do not match");
         }
+        if inputs[0].num_inner_cols() != inputs[1].num_inner_cols() {
+            log::warn!("input number of inner columns do not match");
+        }
+        if inputs[0].num_inner_cols() != output.num_inner_cols() {
+            log::warn!("input and output number of inner columns do not match");
+        }
 
         for i in 0..output.num_blocks() {
             for j in 0..output.num_inner_cols() {
@@ -581,9 +592,9 @@ impl<F: PrimeField + TensorType + PartialOrd + std::hash::Hash> BaseConfig<F> {
                         // this is 0 if the index is the same as the column index (starting from 1)
 
                         let col_expr = sel.clone()
-                            * table
+                            * (table
                                 .selector_constructor
-                                .get_expr_at_idx(col_idx, synthetic_sel);
+                                .get_expr_at_idx(col_idx, synthetic_sel));
 
                         let multiplier =
                             table.selector_constructor.get_selector_val_at_idx(col_idx);
@@ -615,6 +626,40 @@ impl<F: PrimeField + TensorType + PartialOrd + std::hash::Hash> BaseConfig<F> {
                         res
                     });
                 }
+
+                // add a degree-k custom constraint of the following form to the range check and
+                // static lookup configuration.
+                // 𝑚𝑢𝑙𝑡𝑖𝑠𝑒𝑙 · ∏ (𝑠𝑒𝑙 − 𝑖) = 0 where 𝑠𝑒𝑙 is the synthetic_sel, and the product is over the set of overflowed columns
+                // and 𝑚𝑢𝑙𝑡𝑖𝑠𝑒𝑙 is the selector value at the column index
+                cs.create_gate("range_check_on_sel", |cs| {
+                    let synthetic_sel = match len {
+                        1 => Expression::Constant(F::from(1)),
+                        _ => match index {
+                            VarTensor::Advice { inner: advices, .. } => {
+                                cs.query_advice(advices[x][y], Rotation(0))
+                            }
+                            _ => unreachable!(),
+                        },
+                    };
+
+                    let range_check_on_synthetic_sel = match len {
+                        1 => Expression::Constant(F::from(0)),
+                        _ => {
+                            let mut initial_expr = Expression::Constant(F::from(1));
+                            for i in 0..len {
+                                initial_expr = initial_expr
+                                    * (synthetic_sel.clone()
+                                        - Expression::Constant(F::from(i as u64)))
+                            }
+                            initial_expr
+                        }
+                    };
+
+                    let sel = cs.query_selector(multi_col_selector);
+
+                    Constraints::with_selector(sel, vec![range_check_on_synthetic_sel])
+                });
+
                 self.static_lookups
                     .selectors
                     .insert((nl.clone(), x, y), multi_col_selector);
@@ -740,8 +785,8 @@ impl<F: PrimeField + TensorType + PartialOrd + std::hash::Hash> BaseConfig<F> {
     pub fn configure_shuffles(
         &mut self,
         cs: &mut ConstraintSystem<F>,
-        inputs: &[VarTensor; 2],
-        references: &[VarTensor; 2],
+        inputs: &[VarTensor; 3],
+        outputs: &[VarTensor; 3],
     ) -> Result<(), CircuitError>
     where
         F: Field,
@@ -752,14 +797,14 @@ impl<F: PrimeField + TensorType + PartialOrd + std::hash::Hash> BaseConfig<F> {
             }
         }
 
-        for t in references.iter() {
+        for t in outputs.iter() {
             if !t.is_advice() || t.num_inner_cols() > 1 {
                 return Err(CircuitError::WrongDynamicColumnType(t.name().to_string()));
             }
         }
 
         // assert all tables have the same number of blocks
-        if references
+        if outputs
             .iter()
             .map(|t| t.num_blocks())
             .collect::<Vec<_>>()
@@ -767,23 +812,23 @@ impl<F: PrimeField + TensorType + PartialOrd + std::hash::Hash> BaseConfig<F> {
             .any(|w| w[0] != w[1])
         {
             return Err(CircuitError::WrongDynamicColumnType(
-                "references inner cols".to_string(),
+                "outputs inner cols".to_string(),
             ));
         }
 
         let one = Expression::Constant(F::ONE);
 
-        for q in 0..references[0].num_blocks() {
-            let s_reference = cs.complex_selector();
+        for q in 0..outputs[0].num_blocks() {
+            let s_output = cs.complex_selector();
 
             for x in 0..inputs[0].num_blocks() {
                 for y in 0..inputs[0].num_inner_cols() {
                     let s_input = cs.complex_selector();
 
-                    cs.lookup_any("lookup", |cs| {
+                    cs.lookup_any("shuffle", |cs| {
                         let s_inputq = cs.query_selector(s_input);
                         let mut expression = vec![];
-                        let s_referenceq = cs.query_selector(s_reference);
+                        let s_outputq = cs.query_selector(s_output);
                         let mut input_queries = vec![one.clone()];
 
                         for input in inputs {
@@ -795,9 +840,9 @@ impl<F: PrimeField + TensorType + PartialOrd + std::hash::Hash> BaseConfig<F> {
                             });
                         }
 
-                        let mut ref_queries = vec![one.clone()];
-                        for reference in references {
-                            ref_queries.push(match reference {
+                        let mut output_queries = vec![one.clone()];
+                        for output in outputs {
+                            output_queries.push(match output {
                                 VarTensor::Advice { inner: advices, .. } => {
                                     cs.query_advice(advices[q][0], Rotation(0))
                                 }
@@ -806,7 +851,7 @@ impl<F: PrimeField + TensorType + PartialOrd + std::hash::Hash> BaseConfig<F> {
                         }
 
                         let lhs = input_queries.into_iter().map(|c| c * s_inputq.clone());
-                        let rhs = ref_queries.into_iter().map(|c| c * s_referenceq.clone());
+                        let rhs = output_queries.into_iter().map(|c| c * s_outputq.clone());
                         expression.extend(lhs.zip(rhs));
 
                         expression
@@ -817,13 +862,13 @@ impl<F: PrimeField + TensorType + PartialOrd + std::hash::Hash> BaseConfig<F> {
                         .or_insert(s_input);
                 }
             }
-            self.shuffles.reference_selectors.push(s_reference);
+            self.shuffles.output_selectors.push(s_output);
         }
 
         // if we haven't previously initialized the input/output, do so now
-        if self.shuffles.references.is_empty() {
-            debug!("assigning shuffles reference");
-            self.shuffles.references = references.to_vec();
+        if self.shuffles.outputs.is_empty() {
+            debug!("assigning shuffles output");
+            self.shuffles.outputs = outputs.to_vec();
         }
         if self.shuffles.inputs.is_empty() {
             debug!("assigning shuffles input");
@@ -893,9 +938,9 @@ impl<F: PrimeField + TensorType + PartialOrd + std::hash::Hash> BaseConfig<F> {
                         let default_x = range_check.get_first_element(col_idx);
 
                         let col_expr = sel.clone()
-                            * range_check
+                            * (range_check
                                 .selector_constructor
-                                .get_expr_at_idx(col_idx, synthetic_sel);
+                                .get_expr_at_idx(col_idx, synthetic_sel));
 
                         let multiplier = range_check
                             .selector_constructor
@@ -918,6 +963,40 @@ impl<F: PrimeField + TensorType + PartialOrd + std::hash::Hash> BaseConfig<F> {
                         res
                     });
                 }
+
+                // add a degree-k custom constraint of the following form to the range check and
+                // static lookup configuration.
+                // 𝑚𝑢𝑙𝑡𝑖𝑠𝑒𝑙 · ∏ (𝑠𝑒𝑙 − 𝑖) = 0 where 𝑠𝑒𝑙 is the synthetic_sel, and the product is over the set of overflowed columns
+                // and 𝑚𝑢𝑙𝑡𝑖𝑠𝑒𝑙 is the selector value at the column index
+                cs.create_gate("range_check_on_sel", |cs| {
+                    let synthetic_sel = match len {
+                        1 => Expression::Constant(F::from(1)),
+                        _ => match index {
+                            VarTensor::Advice { inner: advices, .. } => {
+                                cs.query_advice(advices[x][y], Rotation(0))
+                            }
+                            _ => unreachable!(),
+                        },
+                    };
+
+                    let range_check_on_synthetic_sel = match len {
+                        1 => Expression::Constant(F::from(0)),
+                        _ => {
+                            let mut initial_expr = Expression::Constant(F::from(1));
+                            for i in 0..len {
+                                initial_expr = initial_expr
+                                    * (synthetic_sel.clone()
+                                        - Expression::Constant(F::from(i as u64)))
+                            }
+                            initial_expr
+                        }
+                    };
+
+                    let sel = cs.query_selector(multi_col_selector);
+
+                    Constraints::with_selector(sel, vec![range_check_on_synthetic_sel])
+                });
+
                 self.range_checks
                     .selectors
                     .insert((range, x, y), multi_col_selector);

src/circuit/ops/errors.rs

@@ -100,4 +100,7 @@ pub enum CircuitError {
     #[error("invalid input type {0}")]
     /// Invalid input type
     InvalidInputType(String),
+    #[error("an element is missing from the shuffled version of the tensor")]
+    /// An element is missing from the shuffled version of the tensor
+    MissingShuffleElement,
 }

src/circuit/ops/hybrid.rs

@@ -1,7 +1,7 @@
 use super::*;
 use crate::{
     circuit::{layouts, utils, Tolerance},
-    fieldutils::integer_rep_to_felt,
+    fieldutils::{integer_rep_to_felt, IntegerRep},
     graph::multiplier_to_scale,
     tensor::{self, Tensor, TensorType, ValTensor},
 };
@@ -250,8 +250,8 @@ impl<F: PrimeField + TensorType + PartialOrd + std::hash::Hash> Op<F> for Hybrid
                 config,
                 region,
                 values[..].try_into()?,
-                integer_rep_to_felt(input_scale.0 as i128),
-                integer_rep_to_felt(output_scale.0 as i128),
+                integer_rep_to_felt(input_scale.0 as IntegerRep),
+                integer_rep_to_felt(output_scale.0 as IntegerRep),
             )?,
             HybridOp::Div { denom, .. } => {
                 if denom.0.fract() == 0.0 {
@@ -259,7 +259,7 @@ impl<F: PrimeField + TensorType + PartialOrd + std::hash::Hash> Op<F> for Hybrid
                         config,
                         region,
                         values[..].try_into()?,
-                        integer_rep_to_felt(denom.0 as i128),
+                        integer_rep_to_felt(denom.0 as IntegerRep),
                     )?
                 } else {
                     layouts::nonlinearity(

src/circuit/ops/layouts.rs

@@ -43,7 +43,7 @@ const ASCII_ALPHABET: &str = "abcdefghijklmnopqrstuvwxyz";
 /// use ezkl::circuit::BaseConfig;
 /// use ezkl::tensor::ValTensor;
 /// let dummy_config = BaseConfig::dummy(12, 2);
-/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(128,2));
+/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(65536, 4));
 /// let x = ValTensor::from_integer_rep_tensor(Tensor::<IntegerRep>::new(
 ///    Some(&[1, 2, 3, 2, 3, 4, 3, 4, 5]),
 /// &[3, 3],
@@ -160,9 +160,26 @@ pub(crate) fn div<F: PrimeField + TensorType + PartialOrd + std::hash::Hash>(
         .into()
     };
     claimed_output.reshape(input_dims)?;
-    region.assign(&config.custom_gates.output, &claimed_output)?;
+    let claimed_output = region.assign(&config.custom_gates.output, &claimed_output)?;
     region.increment(claimed_output.len());
 
+    // here we decompose and extract the sign of the input
+    let sign = sign(config, region, &[claimed_output.clone()])?;
+    // check if x is too large only if the decomp would support overflow in the previous op
+    if (IntegerRep::MAX).abs() < ((region.base() as i128).pow(region.legs() as u32)) - 1 {
+        let abs_value = pairwise(
+            config,
+            region,
+            &[claimed_output.clone(), sign],
+            BaseOp::Mult,
+        )?;
+        let max_val = create_constant_tensor(integer_rep_to_felt(IntegerRep::MAX), 1);
+        let less_than_max = less(config, region, &[abs_value.clone(), max_val])?;
+        // assert the result is 1
+        let comparison_unit = create_constant_tensor(F::ONE, less_than_max.len());
+        enforce_equality(config, region, &[abs_value, comparison_unit])?;
+    }
+
     let product = pairwise(
         config,
         region,
@@ -235,6 +252,30 @@ pub(crate) fn recip<F: PrimeField + TensorType + PartialOrd + std::hash::Hash>(
         &[equal_zero_mask.clone(), equal_inverse_mask],
     )?;
 
+    let masked_output = pairwise(
+        config,
+        region,
+        &[claimed_output.clone(), not_equal_zero_mask.clone()],
+        BaseOp::Mult,
+    )?;
+
+    // here we decompose and extract the sign of the input
+    let sign = sign(config, region, &[masked_output.clone()])?;
+    // check if x is too large only if the decomp would support overflow in the previous op
+    if (IntegerRep::MAX).abs() < ((region.base() as i128).pow(region.legs() as u32)) - 1 {
+        let abs_value = pairwise(
+            config,
+            region,
+            &[claimed_output.clone(), sign],
+            BaseOp::Mult,
+        )?;
+        let max_val = create_constant_tensor(integer_rep_to_felt(IntegerRep::MAX), 1);
+        let less_than_max = less(config, region, &[abs_value.clone(), max_val])?;
+        // assert the result is 1
+        let comparison_unit = create_constant_tensor(F::ONE, less_than_max.len());
+        enforce_equality(config, region, &[abs_value, comparison_unit])?;
+    }
+
     let err_func = |config: &BaseConfig<F>,
                     region: &mut RegionCtx<F>,
                     x: &ValTensor<F>|
@@ -266,7 +307,7 @@ pub(crate) fn recip<F: PrimeField + TensorType + PartialOrd + std::hash::Hash>(
 /// use ezkl::circuit::BaseConfig;
 /// use ezkl::tensor::ValTensor;
 /// let dummy_config = BaseConfig::dummy(12, 2);
-/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(128,2));
+/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(65536, 4));
 /// let x = ValTensor::from_integer_rep_tensor(Tensor::<IntegerRep>::new(
 ///     Some(&[1, 2, 3, 2, 3, 4, 3, 4, 9]),
 ///    &[3, 3],
@@ -338,7 +379,7 @@ pub fn sqrt<F: PrimeField + TensorType + PartialOrd + std::hash::Hash>(
 /// use ezkl::circuit::BaseConfig;
 /// use ezkl::tensor::ValTensor;
 /// let dummy_config = BaseConfig::dummy(12, 2);
-/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(128,2));
+/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(65536, 4));
 /// let x = ValTensor::from_integer_rep_tensor(Tensor::<IntegerRep>::new(
 ///    Some(&[1, 2, 3, 2, 3, 4, 3, 4, 5]),
 /// &[3, 3],
@@ -377,7 +418,7 @@ pub fn rsqrt<F: PrimeField + TensorType + PartialOrd + std::hash::Hash>(
 /// use ezkl::circuit::layouts::dot;
 ///
 /// let dummy_config = BaseConfig::dummy(12, 2);
-/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(128,2));
+/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(65536, 4));
 ///
 /// let x = ValTensor::from_integer_rep_tensor(Tensor::<IntegerRep>::new(
 ///     Some(&[5, 2, 3, 0, 4, -1, 3, 1, 6]),
@@ -404,22 +445,6 @@ pub fn dot<F: PrimeField + TensorType + PartialOrd + std::hash::Hash>(
 
     let mut values = values.clone();
 
-    // this section has been optimized to death, don't mess with it
-    let mut removal_indices = values[0].get_const_zero_indices();
-    let second_zero_indices = values[1].get_const_zero_indices();
-    removal_indices.extend(second_zero_indices);
-    removal_indices.par_sort_unstable();
-    removal_indices.dedup();
-
-    // if empty return a const
-    if removal_indices.len() == values[0].len() {
-        return Ok(create_zero_tensor(1));
-    }
-
-    // is already sorted
-    values[0].remove_indices(&mut removal_indices, true)?;
-    values[1].remove_indices(&mut removal_indices, true)?;
-
     let mut inputs = vec![];
     let block_width = config.custom_gates.output.num_inner_cols();
 
@@ -492,7 +517,7 @@ pub fn dot<F: PrimeField + TensorType + PartialOrd + std::hash::Hash>(
 /// use ezkl::tensor::ValTensor;
 ///
 /// let dummy_config = BaseConfig::dummy(12, 2);
-/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(128,2));
+/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(65536, 4));
 ///
 /// // matmul case
 /// let x = ValTensor::from_integer_rep_tensor(Tensor::<IntegerRep>::new(
@@ -906,7 +931,6 @@ fn _sort_ascending<F: PrimeField + TensorType + PartialOrd + std::hash::Hash>(
     let window_b = assigned_sort.get_slice(&[1..assigned_sort.len()])?;
 
     let is_greater = greater_equal(config, region, &[window_b.clone(), window_a.clone()])?;
-
     let unit = create_unit_tensor(is_greater.len());
 
     enforce_equality(config, region, &[unit, is_greater])?;
@@ -945,7 +969,7 @@ fn _select_topk<F: PrimeField + TensorType + PartialOrd + std::hash::Hash>(
 /// use ezkl::tensor::ValTensor;
 ///
 /// let dummy_config = BaseConfig::dummy(12, 2);
-/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(128,2));
+/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(65536, 4));
 ///
 /// let x = ValTensor::from_integer_rep_tensor(Tensor::<IntegerRep>::new(
 ///     Some(&[2, 15, 2, 1, 1, 0]),
@@ -1168,30 +1192,38 @@ pub(crate) fn dynamic_lookup<F: PrimeField + TensorType + PartialOrd + std::hash
 }
 
 /// Shuffle arg
+/// 1. the input is a set of pairs (index_input, value_input) -- looked up against a (dynamic) set of values (index_output, value_output).
+/// 2. index_input is copy constrained to values in a fixed column and is thus a fixed set of incrementing values over the input
+/// 3. value_input is just the input that we are ascertaining is shuffled
+/// 4. index_output is (typically) a prover generated witness committed to in an advice column
+/// 5. value_output is (typically) a prover generated witness committed to in an advice column
+/// 6. Given the above, and given the fixed index_input , we go through every (index_input, value_input) pair and ascertain that it is contained in the input.
+/// Given the fixed incrementing index index_input, we avoid multiplicity in the output by leveraging this surrogate index: if index_output isn't matched to the exact value where for `index_input=index_output` -> `value_input=value_output`, then the lookup fails
 pub(crate) fn shuffles<F: PrimeField + TensorType + PartialOrd + std::hash::Hash>(
     config: &BaseConfig<F>,
     region: &mut RegionCtx<F>,
+    output: &[ValTensor<F>; 1],
     input: &[ValTensor<F>; 1],
-    reference: &[ValTensor<F>; 1],
 ) -> Result<ValTensor<F>, CircuitError> {
     let shuffle_index = region.shuffle_index();
-    let (input, reference) = (input[0].clone(), reference[0].clone());
+    let (output, input) = (output[0].clone(), input[0].clone());
 
     // assert input and reference are same length
-    if input.len() != reference.len() {
+    if output.len() != input.len() {
         return Err(CircuitError::MismatchedShuffleLength(
+            output.len(),
             input.len(),
-            reference.len(),
         ));
     }
 
-    let (reference, flush_len_ref) =
-        region.assign_shuffle(&config.shuffles.references[0], &reference)?;
-    let reference_len = reference.len();
+    let (output, flush_len_ref) = region.assign_shuffle(&config.shuffles.outputs[0], &output)?;
+    let output_len = output.len();
+    let input = region.assign(&config.shuffles.inputs[0], &input)?;
 
     // now create a vartensor of constants for the shuffle index
-    let index = create_constant_tensor(F::from(shuffle_index as u64), reference_len);
-    let (index, flush_len_index) = region.assign_shuffle(&config.shuffles.references[1], &index)?;
+    let index = create_constant_tensor(F::from(shuffle_index as u64), output_len);
+    let (index, flush_len_index) = region.assign_shuffle(&config.shuffles.outputs[1], &index)?;
+    region.assign(&config.shuffles.inputs[1], &index)?;
 
     if flush_len_index != flush_len_ref {
         return Err(CircuitError::MismatchedShuffleLength(
@@ -1200,18 +1232,71 @@ pub(crate) fn shuffles<F: PrimeField + TensorType + PartialOrd + std::hash::Hash
         ));
     }
 
-    let input = region.assign(&config.shuffles.inputs[0], &input)?;
-    region.assign(&config.shuffles.inputs[1], &index)?;
+    // now found the position of each element of the reference to the input
+
+    let is_known = !output.any_unknowns()? && !input.any_unknowns()?;
+
+    let claimed_index_output = if is_known {
+        let input = input.int_evals()?;
+        let output = output.int_evals()?;
+
+        // Keep track of which positions we've used for each value
+        let mut used_positions: HashMap<usize, bool> = HashMap::new();
+
+        let index_output = output
+            .iter()
+            .map(|x| {
+                // Find all positions of the current element
+                let positions: Vec<usize> = input
+                    .iter()
+                    .enumerate()
+                    .filter(|(_, y)| *y == x)
+                    .map(|(i, _)| i)
+                    .collect();
+
+                // Find the first unused position for this element
+                let pos = positions
+                    .iter()
+                    .find(|&&i| !used_positions.get(&i).unwrap_or(&false))
+                    .unwrap_or(&0);
+
+                // Mark this position as used
+                used_positions.insert(*pos, true);
+
+                Ok::<_, CircuitError>(Value::known(F::from(*pos as u64)))
+            })
+            .collect::<Result<Vec<_>, _>>()?;
+
+        // Verify all indices were used exactly once
+        if !used_positions.values().all(|&x| x) {
+            return Err(CircuitError::MissingShuffleElement);
+        }
+
+        Tensor::from(index_output.into_iter()).into()
+    } else {
+        Tensor::new(
+            Some(&vec![Value::<F>::unknown(); output_len]),
+            &[output_len],
+        )?
+        .into()
+    };
+
+    region.assign_shuffle(&config.shuffles.outputs[2], &claimed_index_output)?;
+
+    // the incrementing index is the set of numbered values for the input tensor 0...n, and is FIXED
+    let incrementing_index: ValTensor<F> =
+        Tensor::from((0..output.len() as u64).map(|x| ValType::Constant(F::from(x)))).into();
+    region.assign(&config.shuffles.inputs[2], &incrementing_index)?;
 
     let mut shuffle_block = 0;
 
     if !region.is_dummy() {
-        (0..reference_len)
+        (0..output_len)
             .map(|i| {
-                let (x, _, z) = config.shuffles.references[0]
+                let (x, _, z) = config.shuffles.outputs[0]
                     .cartesian_coord(region.combined_dynamic_shuffle_coord() + i + flush_len_ref);
                 shuffle_block = x;
-                let ref_selector = config.shuffles.reference_selectors[shuffle_block];
+                let ref_selector = config.shuffles.output_selectors[shuffle_block];
                 region.enable(Some(&ref_selector), z)?;
                 Ok(())
             })
@@ -1220,7 +1305,7 @@ pub(crate) fn shuffles<F: PrimeField + TensorType + PartialOrd + std::hash::Hash
 
     if !region.is_dummy() {
         // Enable the selectors
-        (0..reference_len)
+        (0..output_len)
             .map(|i| {
                 let (x, y, z) =
                     config.custom_gates.inputs[0].cartesian_coord(region.linear_coord() + i);
@@ -1237,11 +1322,11 @@ pub(crate) fn shuffles<F: PrimeField + TensorType + PartialOrd + std::hash::Hash
             .collect::<Result<Vec<_>, CircuitError>>()?;
     }
 
-    region.increment_shuffle_col_coord(reference_len + flush_len_ref);
+    region.increment_shuffle_col_coord(output_len + flush_len_ref);
     region.increment_shuffle_index(1);
-    region.increment(reference_len);
+    region.increment(output_len);
 
-    Ok(input)
+    Ok(output)
 }
 
 /// One hot accumulated layout
@@ -1698,6 +1783,8 @@ pub(crate) fn get_missing_set_elements<
     // assign the claimed output
     claimed_output = region.assign(&config.custom_gates.output, &claimed_output)?;
 
+    region.increment(claimed_output.len());
+
     // input and claimed output should be the shuffles of fullset
     // concatentate input and claimed output
     let input_and_claimed_output = input.concat(claimed_output.clone())?;
@@ -1891,7 +1978,7 @@ pub(crate) fn scatter_nd<F: PrimeField + TensorType + PartialOrd + std::hash::Ha
 /// use ezkl::tensor::ValTensor;
 ///
 /// let dummy_config = BaseConfig::dummy(12, 2);
-/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(128,2));
+/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(65536, 4));
 ///
 /// let x = ValTensor::from_integer_rep_tensor(Tensor::<IntegerRep>::new(
 ///     Some(&[2, 15, 2, 1, 1, 0]),
@@ -1989,7 +2076,7 @@ pub fn sum<F: PrimeField + TensorType + PartialOrd + std::hash::Hash>(
 /// use ezkl::tensor::ValTensor;
 ///
 /// let dummy_config = BaseConfig::dummy(12, 2);
-/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(128,2));
+/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(65536, 4));
 ///
 /// let x = ValTensor::from_integer_rep_tensor(Tensor::<IntegerRep>::new(
 ///     Some(&[2, 15, 2, 1, 1, 0]),
@@ -2144,7 +2231,7 @@ fn axes_wise_op<F: PrimeField + TensorType + PartialOrd + std::hash::Hash>(
 /// use ezkl::tensor::ValTensor;
 ///
 /// let dummy_config = BaseConfig::dummy(12, 2);
-/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(128,2));
+/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(65536, 4));
 ///
 /// let x = ValTensor::from_integer_rep_tensor(Tensor::<IntegerRep>::new(
 ///     Some(&[2, 15, 2, 1, 1, 0]),
@@ -2180,7 +2267,7 @@ pub fn prod_axes<F: PrimeField + TensorType + PartialOrd + std::hash::Hash>(
 /// use ezkl::tensor::ValTensor;
 ///
 /// let dummy_config = BaseConfig::dummy(12, 2);
-/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(128,2));
+/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(65536, 4));
 ///
 /// let x = ValTensor::from_integer_rep_tensor(Tensor::<IntegerRep>::new(
 ///     Some(&[2, 15, 2, 1, 1, 0]),
@@ -2216,7 +2303,7 @@ pub fn sum_axes<F: PrimeField + TensorType + PartialOrd + std::hash::Hash>(
 /// use ezkl::tensor::ValTensor;
 ///
 /// let dummy_config = BaseConfig::dummy(12, 2);
-/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(128,2));
+/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(65536, 4));
 ///
 /// let x = ValTensor::from_integer_rep_tensor(Tensor::<IntegerRep>::new(
 ///     Some(&[2, 15, 2, 1, 1, 0]),
@@ -2258,7 +2345,7 @@ pub fn argmax_axes<F: PrimeField + TensorType + PartialOrd + std::hash::Hash>(
 /// use ezkl::tensor::ValTensor;
 ///
 /// let dummy_config = BaseConfig::dummy(12, 2);
-/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(128,2));
+/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(65536, 4));
 /// let x = ValTensor::from_integer_rep_tensor(Tensor::<IntegerRep>::new(
 ///     Some(&[2, 15, 2, 1, 1, 0]),
 ///     &[2, 3],
@@ -2294,7 +2381,7 @@ pub fn max_axes<F: PrimeField + TensorType + PartialOrd + std::hash::Hash>(
 /// use ezkl::tensor::ValTensor;
 ///
 /// let dummy_config = BaseConfig::dummy(12, 2);
-/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(128,2));
+/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(65536, 4));
 ///
 /// let x = ValTensor::from_integer_rep_tensor(Tensor::<IntegerRep>::new(
 ///     Some(&[2, 15, 2, 1, 1, 0]),
@@ -2340,7 +2427,7 @@ pub fn argmin_axes<F: PrimeField + TensorType + PartialOrd + std::hash::Hash>(
 /// use ezkl::tensor::ValTensor;
 ///
 /// let dummy_config = BaseConfig::dummy(12, 2);
-/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(128,2));
+/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(65536, 4));
 ///
 /// let x = ValTensor::from_integer_rep_tensor(Tensor::<IntegerRep>::new(
 ///     Some(&[2, 15, 2, 1, 1, 0]),
@@ -2381,25 +2468,6 @@ pub(crate) fn pairwise<F: PrimeField + TensorType + PartialOrd + std::hash::Hash
     lhs.expand(&broadcasted_shape)?;
     rhs.expand(&broadcasted_shape)?;
 
-    // original values
-    let orig_lhs = lhs.clone();
-    let orig_rhs = rhs.clone();
-
-    let first_zero_indices = HashSet::from_iter(lhs.get_const_zero_indices());
-    let second_zero_indices = HashSet::from_iter(rhs.get_const_zero_indices());
-
-    let removal_indices = match op {
-        BaseOp::Add | BaseOp::Mult => {
-            // join the zero indices
-            first_zero_indices
-                .union(&second_zero_indices)
-                .cloned()
-                .collect()
-        }
-        BaseOp::Sub => second_zero_indices.clone(),
-        _ => return Err(CircuitError::UnsupportedOp),
-    };
-
     if lhs.len() != rhs.len() {
         return Err(CircuitError::DimMismatch(format!(
             "pairwise {} layout",
@@ -2411,11 +2479,7 @@ pub(crate) fn pairwise<F: PrimeField + TensorType + PartialOrd + std::hash::Hash
         .iter()
         .enumerate()
         .map(|(i, input)| {
-            let res = region.assign_with_omissions(
-                &config.custom_gates.inputs[i],
-                input,
-                &removal_indices,
-            )?;
+            let res = region.assign(&config.custom_gates.inputs[i], input)?;
 
             Ok(res.get_inner()?)
         })
@@ -2434,12 +2498,8 @@ pub(crate) fn pairwise<F: PrimeField + TensorType + PartialOrd + std::hash::Hash
         halo2_proofs::plonk::Error::Synthesis
     })?;
 
-    let assigned_len = op_result.len() - removal_indices.len();
-    let mut output = region.assign_with_omissions(
-        &config.custom_gates.output,
-        &op_result.into(),
-        &removal_indices,
-    )?;
+    let assigned_len = op_result.len();
+    let mut output = region.assign(&config.custom_gates.output, &op_result.into())?;
 
     // Enable the selectors
     if !region.is_dummy() {
@@ -2457,44 +2517,6 @@ pub(crate) fn pairwise<F: PrimeField + TensorType + PartialOrd + std::hash::Hash
     }
     region.increment(assigned_len);
 
-    let a_tensor = orig_lhs.get_inner_tensor()?;
-    let b_tensor = orig_rhs.get_inner_tensor()?;
-
-    // infill the zero indices with the correct values from values[0] or values[1]
-    if !removal_indices.is_empty() {
-        output
-            .get_inner_tensor_mut()?
-            .par_enum_map_mut_filtered(&removal_indices, |i| {
-                let val = match op {
-                    BaseOp::Add => {
-                        let a_is_null = first_zero_indices.contains(&i);
-                        let b_is_null = second_zero_indices.contains(&i);
-
-                        if a_is_null && b_is_null {
-                            ValType::Constant(F::ZERO)
-                        } else if a_is_null {
-                            b_tensor[i].clone()
-                        } else {
-                            a_tensor[i].clone()
-                        }
-                    }
-                    BaseOp::Sub => {
-                        let a_is_null = first_zero_indices.contains(&i);
-                        // by default b is null in this case for sub
-                        if a_is_null {
-                            ValType::Constant(F::ZERO)
-                        } else {
-                            a_tensor[i].clone()
-                        }
-                    }
-                    BaseOp::Mult => ValType::Constant(F::ZERO),
-                    // can safely panic as the prior check ensures this is not called
-                    _ => unreachable!(),
-                };
-                Ok::<_, TensorError>(val)
-            })?;
-    }
-
     output.reshape(&broadcasted_shape)?;
 
     let end = global_start.elapsed();
@@ -2522,7 +2544,7 @@ pub(crate) fn pairwise<F: PrimeField + TensorType + PartialOrd + std::hash::Hash
 /// use ezkl::tensor::ValTensor;
 ///
 /// let dummy_config = BaseConfig::dummy(12, 2);
-/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(128,2));
+/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(65536, 4));
 ///
 /// let x = ValTensor::from_integer_rep_tensor(Tensor::<IntegerRep>::new(
 /// Some(&[2, 15, 2, 1, 1, 0]),
@@ -2579,7 +2601,7 @@ pub(crate) fn expand<F: PrimeField + TensorType + PartialOrd + std::hash::Hash>(
 /// use ezkl::tensor::ValTensor;
 ///
 /// let dummy_config = BaseConfig::dummy(12, 2);
-/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(128, 2));
+/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(65536, 4));
 ///
 /// let a = ValTensor::from_integer_rep_tensor(Tensor::<IntegerRep>::new(
 ///   Some(&[1, 12, 6, 4, 5, 6]),
@@ -2607,7 +2629,8 @@ pub fn greater<F: PrimeField + TensorType + PartialOrd + std::hash::Hash>(
 
     let diff = pairwise(config, region, &[lhs, rhs], BaseOp::Sub)?;
     let sign = sign(config, region, &[diff])?;
-    equals(config, region, &[sign, create_unit_tensor(1)])
+    let eq = equals(config, region, &[sign, create_unit_tensor(1)])?;
+    Ok(eq)
 }
 
 /// Greater equals than operation.
@@ -2626,7 +2649,7 @@ pub fn greater<F: PrimeField + TensorType + PartialOrd + std::hash::Hash>(
 /// use ezkl::tensor::ValTensor;
 ///
 /// let dummy_config = BaseConfig::dummy(12, 2);
-/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(128, 2));
+/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(65536, 4));
 ///
 ///
 /// let a = ValTensor::from_integer_rep_tensor(Tensor::<IntegerRep>::new(
@@ -2676,7 +2699,7 @@ pub fn greater_equal<F: PrimeField + TensorType + PartialOrd + std::hash::Hash>(
 /// use ezkl::tensor::ValTensor;
 ///
 /// let dummy_config = BaseConfig::dummy(12, 2);
-/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(128, 2));
+/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(65536, 4));
 ///
 /// let a = ValTensor::from_integer_rep_tensor(Tensor::<IntegerRep>::new(
 ///  Some(&[1, 0, 5, 4, 5, 1]),
@@ -2717,7 +2740,7 @@ pub fn less<F: PrimeField + TensorType + PartialOrd + std::hash::Hash>(
 /// use ezkl::tensor::ValTensor;
 ///
 /// let dummy_config = BaseConfig::dummy(12, 2);
-/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(128,2));
+/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(65536, 4));
 ///
 /// let a = ValTensor::from_integer_rep_tensor(Tensor::<IntegerRep>::new(
 ///  Some(&[1, 0, 5, 4, 5, 1]),
@@ -2758,7 +2781,7 @@ pub fn less_equal<F: PrimeField + TensorType + PartialOrd + std::hash::Hash>(
 /// use ezkl::tensor::ValTensor;
 ///
 /// let dummy_config = BaseConfig::dummy(12, 2);
-/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(128,2));
+/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(65536, 4));
 ///
 /// let a = ValTensor::from_integer_rep_tensor(Tensor::<IntegerRep>::new(
 ///  Some(&[1, 1, 1, 1, 1, 0]),
@@ -2802,7 +2825,7 @@ pub fn and<F: PrimeField + TensorType + PartialOrd + std::hash::Hash>(
 /// use ezkl::tensor::ValTensor;
 ///
 /// let dummy_config = BaseConfig::dummy(12, 2);
-/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(128,2));
+/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(65536, 4));
 ///
 /// let a = ValTensor::from_integer_rep_tensor(Tensor::<IntegerRep>::new(
 ///   Some(&[1, 1, 1, 1, 1, 0]),
@@ -2850,7 +2873,7 @@ pub fn or<F: PrimeField + TensorType + PartialOrd + std::hash::Hash>(
 /// use ezkl::tensor::ValTensor;
 ///
 /// let dummy_config = BaseConfig::dummy(12, 2);
-/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(128,2));
+/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(65536, 4));
 ///
 /// let a = ValTensor::from_integer_rep_tensor(Tensor::<IntegerRep>::new(
 /// Some(&[1, 1, 1, 1, 1, 0]),
@@ -2881,6 +2904,7 @@ pub(crate) fn equals_zero<F: PrimeField + TensorType + PartialOrd + std::hash::H
 ) -> Result<ValTensor<F>, CircuitError> {
     let values = values[0].clone();
     let values_inverse = values.inverse()?;
+
     let product_values_and_invert = pairwise(
         config,
         region,
@@ -2924,7 +2948,7 @@ pub(crate) fn equals_zero<F: PrimeField + TensorType + PartialOrd + std::hash::H
 /// use ezkl::tensor::ValTensor;
 ///
 /// let dummy_config = BaseConfig::dummy(12, 2);
-/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(128,2));
+/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(65536, 4));
 ///
 /// let a = ValTensor::from_integer_rep_tensor(Tensor::<IntegerRep>::new(
 ///  Some(&[1, 1, 1, 1, 1, 0]),
@@ -2980,7 +3004,7 @@ pub fn xor<F: PrimeField + TensorType + PartialOrd + std::hash::Hash>(
 /// use ezkl::tensor::ValTensor;
 ///
 /// let dummy_config = BaseConfig::dummy(12, 2);
-/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(128,2));
+/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(65536, 4));
 ///
 /// let x = ValTensor::from_integer_rep_tensor(Tensor::<IntegerRep>::new(
 ///    Some(&[1, 1, 1, 1, 1, 0]),
@@ -3023,7 +3047,7 @@ pub fn not<F: PrimeField + TensorType + PartialOrd + std::hash::Hash>(
 /// use ezkl::tensor::ValTensor;
 ///
 /// let dummy_config = BaseConfig::dummy(12, 2);
-/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(128,2));
+/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(65536, 4));
 ///
 /// let mask = ValTensor::from_integer_rep_tensor(Tensor::<IntegerRep>::new(
 ///    Some(&[1, 0, 1, 0, 1, 0]),
@@ -3081,7 +3105,7 @@ pub fn iff<F: PrimeField + TensorType + PartialOrd + std::hash::Hash>(
 /// use ezkl::tensor::ValTensor;
 ///
 /// let dummy_config = BaseConfig::dummy(12, 2);
-/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(128,2));
+/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(65536, 4));
 ///
 /// let x = ValTensor::from_integer_rep_tensor(Tensor::<IntegerRep>::new(
 ///     Some(&[2, 1, 2, 1, 1, 1]),
@@ -3113,7 +3137,7 @@ pub fn neg<F: PrimeField + TensorType + PartialOrd + std::hash::Hash>(
 /// use ezkl::tensor::ValTensor;
 ///
 /// let dummy_config = BaseConfig::dummy(12, 2);
-/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(128,2));
+/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(65536, 4));
 ///
 ///
 /// let x = ValTensor::from_integer_rep_tensor(Tensor::<IntegerRep>::new(
@@ -3195,7 +3219,7 @@ pub fn sumpool<F: PrimeField + TensorType + PartialOrd + std::hash::Hash>(
 /// use ezkl::tensor::ValTensor;
 ///
 /// let dummy_config = BaseConfig::dummy(12, 2);
-/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(128,2));
+/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(65536, 4));
 ///
 ///
 /// let x = ValTensor::from_integer_rep_tensor(Tensor::<IntegerRep>::new(
@@ -3249,32 +3273,30 @@ pub fn max_pool<F: PrimeField + TensorType + PartialOrd + std::hash::Hash>(
         .multi_cartesian_product()
         .collect::<Vec<_>>();
 
-    output
-        .iter_mut()
-        .enumerate()
-        .map(|(flat_index, o)| {
-            let coord = &cartesian_coord[flat_index];
-            let (b, i) = (coord[0], coord[1]);
+    let inner_loop_function = |idx: usize, region: &mut RegionCtx<F>| {
+        let coord = &cartesian_coord[idx];
+        let (b, i) = (coord[0], coord[1]);
 
-            let mut slice = vec![b..b + 1, i..i + 1];
-            slice.extend(
-                coord[2..]
-                    .iter()
-                    .zip(stride.iter())
-                    .zip(pool_dims.iter())
-                    .map(|((c, s), k)| {
-                        let start = c * s;
-                        let end = start + k;
-                        start..end
-                    }),
-            );
+        let mut slice = vec![b..b + 1, i..i + 1];
+        slice.extend(
+            coord[2..]
+                .iter()
+                .zip(stride.iter())
+                .zip(pool_dims.iter())
+                .map(|((c, s), k)| {
+                    let start = c * s;
+                    let end = start + k;
+                    start..end
+                }),
+        );
 
-            let slice = padded_image.get_slice(&slice)?;
-            let max_w = max(config, region, &[slice])?;
-            *o = max_w.get_inner_tensor()?[0].clone();
-            Ok(())
-        })
-        .collect::<Result<Vec<_>, CircuitError>>()?;
+        let slice = padded_image.get_slice(&slice)?;
+        let max_w = max(config, region, &[slice])?;
+
+        Ok::<_, CircuitError>(max_w.get_inner_tensor()?[0].clone())
+    };
+
+    region.apply_in_loop(&mut output, inner_loop_function)?;
 
     let res: ValTensor<F> = output.into();
 
@@ -3296,7 +3318,7 @@ pub fn max_pool<F: PrimeField + TensorType + PartialOrd + std::hash::Hash>(
 /// use ezkl::tensor::ValTensor;
 ///
 /// let dummy_config = BaseConfig::dummy(12, 2);
-/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(128,2));
+/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(65536, 4));
 ///
 /// let c = ValTensor::from_integer_rep_tensor(Tensor::<IntegerRep>::new(Some(&[6, 0, 12, 4, 0, 8, 0, 0, 3, 0, 0, 2]), &[1, 2, 2, 3]).unwrap());
 /// let x = ValTensor::from_integer_rep_tensor(Tensor::<IntegerRep>::new(
@@ -3524,7 +3546,7 @@ pub fn deconv<
 /// use ezkl::circuit::BaseConfig;
 ///
 /// let dummy_config = BaseConfig::dummy(12, 2);
-/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(128,2));
+/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(65536, 4));
 ///
 /// let x = ValTensor::from_integer_rep_tensor(Tensor::<IntegerRep>::new(
 ///     Some(&[5, 2, 3, 0, 4, -1, 3, 1, 6]),
@@ -4248,7 +4270,7 @@ pub(crate) fn argmin<F: PrimeField + TensorType + PartialOrd + std::hash::Hash>(
 /// use ezkl::circuit::region::RegionSettings;
 /// use ezkl::circuit::BaseConfig;
 /// let dummy_config = BaseConfig::dummy(12, 2);
-/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(128,2));
+/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(65536, 4));
 /// let x = ValTensor::from_integer_rep_tensor(Tensor::<IntegerRep>::new(
 ///    Some(&[5, 2, 3, 0]),
 ///   &[1, 1, 2, 2],
@@ -4301,7 +4323,7 @@ pub fn max_comp<F: PrimeField + TensorType + PartialOrd + std::hash::Hash>(
 /// use ezkl::circuit::region::RegionSettings;
 /// use ezkl::circuit::BaseConfig;
 /// let dummy_config = BaseConfig::dummy(12, 2);
-/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(128,2));
+/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(65536, 4));
 /// let x = ValTensor::from_integer_rep_tensor(Tensor::<IntegerRep>::new(
 ///   Some(&[5, 2, 3, 0]),
 ///  &[1, 1, 2, 2],
@@ -4340,10 +4362,7 @@ pub(crate) fn max<F: PrimeField + TensorType + PartialOrd + std::hash::Hash>(
     region: &mut RegionCtx<F>,
     values: &[ValTensor<F>; 1],
 ) -> Result<ValTensor<F>, CircuitError> {
-    let input_len = values[0].len();
-    _sort_ascending(config, region, values)?
-        .get_slice(&[input_len - 1..input_len])
-        .map_err(|e| e.into())
+    Ok(_sort_ascending(config, region, values)?.last()?)
 }
 
 /// min layout
@@ -4352,9 +4371,7 @@ pub(crate) fn min<F: PrimeField + TensorType + PartialOrd + std::hash::Hash>(
     region: &mut RegionCtx<F>,
     values: &[ValTensor<F>; 1],
 ) -> Result<ValTensor<F>, CircuitError> {
-    _sort_ascending(config, region, values)?
-        .get_slice(&[0..1])
-        .map_err(|e| e.into())
+    Ok(_sort_ascending(config, region, values)?.first()?)
 }
 
 /// floor layout
@@ -4377,7 +4394,7 @@ pub(crate) fn min<F: PrimeField + TensorType + PartialOrd + std::hash::Hash>(
 /// use ezkl::circuit::region::RegionSettings;
 /// use ezkl::circuit::BaseConfig;
 /// let dummy_config = BaseConfig::dummy(12, 2);
-/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(128,2));
+/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(65536, 4));
 /// let x = ValTensor::from_integer_rep_tensor(Tensor::<IntegerRep>::new(
 /// Some(&[3, -2, -3, 1]),
 /// &[1, 1, 2, 2],
@@ -4489,7 +4506,7 @@ pub fn floor<F: PrimeField + TensorType + PartialOrd + std::hash::Hash>(
 /// use ezkl::circuit::region::RegionSettings;
 /// use ezkl::circuit::BaseConfig;
 /// let dummy_config = BaseConfig::dummy(12, 2);
-/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(128,2));
+/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(65536, 4));
 /// let x = ValTensor::from_integer_rep_tensor(Tensor::<IntegerRep>::new(
 ///  Some(&[3, -2, 3, 1]),
 /// &[1, 1, 2, 2],
@@ -4602,7 +4619,7 @@ pub fn ceil<F: PrimeField + TensorType + PartialOrd + std::hash::Hash>(
 /// use ezkl::circuit::region::RegionSettings;
 /// use ezkl::circuit::BaseConfig;
 /// let dummy_config = BaseConfig::dummy(12, 2);
-/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(128,2));
+/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(65536, 4));
 /// let x = ValTensor::from_integer_rep_tensor(Tensor::<IntegerRep>::new(
 /// Some(&[3, 2, 3, 1]),
 /// &[1, 1, 2, 2],
@@ -4889,7 +4906,7 @@ pub fn ln<F: PrimeField + TensorType + PartialOrd + std::hash::Hash>(
 /// use ezkl::circuit::region::RegionSettings;
 /// use ezkl::circuit::BaseConfig;
 /// let dummy_config = BaseConfig::dummy(12, 2);
-/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(128,2));
+/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(65536, 4));
 /// let x = ValTensor::from_integer_rep_tensor(Tensor::<IntegerRep>::new(
 /// Some(&[3, -2, 3, 1]),
 /// &[1, 1, 2, 2],
@@ -5032,7 +5049,7 @@ pub fn round<F: PrimeField + TensorType + PartialOrd + std::hash::Hash>(
 /// use ezkl::circuit::region::RegionSettings;
 /// use ezkl::circuit::BaseConfig;
 /// let dummy_config = BaseConfig::dummy(12, 2);
-/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(128,2));
+/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(65536, 4));
 /// let x = ValTensor::from_integer_rep_tensor(Tensor::<IntegerRep>::new(
 /// Some(&[3, -2, 3, 1]),
 /// &[1, 1, 2, 2],
@@ -5267,7 +5284,24 @@ pub(crate) fn decompose<F: PrimeField + TensorType + PartialOrd + std::hash::Has
     let rest = claimed_output.get_slice(&rest_slice)?;
 
     let sign = range_check(config, region, &[sign], &(-1, 1))?;
-    let rest = range_check(config, region, &[rest], &(0, (*base - 1) as i128))?;
+
+    // isZero(input) * sign == 0.
+    {
+        let is_zero = equals_zero(config, region, &[input.clone()])?;
+        // take the product of the sign and is_zero
+        let sign_is_zero = pairwise(config, region, &[sign.clone(), is_zero], BaseOp::Mult)?;
+        // constrain the sign_is_zero to be 0
+        enforce_equality(
+            config,
+            region,
+            &[
+                sign_is_zero.clone(),
+                create_constant_tensor(F::ZERO, sign_is_zero.len()),
+            ],
+        )?;
+    }
+
+    let rest = range_check(config, region, &[rest], &(0, (*base - 1) as IntegerRep))?;
 
     // equation needs to be constructed as ij,ij->i but for arbitrary n dims we need to construct this dynamically
     // indices should map in order of the alphabet
@@ -5531,7 +5565,7 @@ pub(crate) fn percent<F: PrimeField + TensorType + PartialOrd + std::hash::Hash>
 /// use ezkl::circuit::BaseConfig;
 ///
 /// let dummy_config = BaseConfig::dummy(12, 2);
-/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(128,2));
+/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(65536, 4));
 ///
 /// let x = ValTensor::from_integer_rep_tensor(Tensor::<IntegerRep>::new(
 ///     Some(&[2, 2, 3, 2, 2, 0]),
@@ -5582,7 +5616,7 @@ pub fn softmax<F: PrimeField + TensorType + PartialOrd + std::hash::Hash>(
 /// use ezkl::circuit::BaseConfig;
 ///
 /// let dummy_config = BaseConfig::dummy(12, 2);
-/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(128,2));
+/// let mut dummy_region = RegionCtx::new_dummy(0,2,RegionSettings::all_true(65536, 4));
 ///
 /// let x = ValTensor::from_integer_rep_tensor(Tensor::<IntegerRep>::new(
 ///     Some(&[100, 200, 300, 400, 500, 600]),

src/circuit/table.rs

@@ -132,21 +132,16 @@ impl<F: PrimeField + TensorType + PartialOrd + std::hash::Hash> Table<F> {
         (first_element, op_f.output[0])
     }
 
-    ///
+    /// calculates the column size given the number of rows and reserved blinding rows
     pub fn cal_col_size(logrows: usize, reserved_blinding_rows: usize) -> usize {
         2usize.pow(logrows as u32) - reserved_blinding_rows
     }
-
-    ///
-    pub fn cal_bit_range(bits: usize, reserved_blinding_rows: usize) -> usize {
-        2usize.pow(bits as u32) - reserved_blinding_rows
-    }
 }
 
 ///
 pub fn num_cols_required(range_len: IntegerRep, col_size: usize) -> usize {
     // number of cols needed to store the range
-    (range_len / (col_size as IntegerRep)) as usize + 1
+    (range_len / col_size as IntegerRep) as usize + 1
 }
 
 impl<F: PrimeField + TensorType + PartialOrd + std::hash::Hash> Table<F> {
@@ -355,16 +350,11 @@ impl<F: PrimeField + TensorType + PartialOrd + std::hash::Hash> RangeCheck<F> {
         integer_rep_to_felt(chunk * (self.col_size as IntegerRep) + self.range.0)
     }
 
-    ///
+    /// calculates the column size
     pub fn cal_col_size(logrows: usize, reserved_blinding_rows: usize) -> usize {
         2usize.pow(logrows as u32) - reserved_blinding_rows
     }
 
-    ///
-    pub fn cal_bit_range(bits: usize, reserved_blinding_rows: usize) -> usize {
-        2usize.pow(bits as u32) - reserved_blinding_rows
-    }
-
     /// get column index given input
     pub fn get_col_index(&self, input: F) -> F {
         //    range is split up into chunks of size col_size, find the chunk that input is in

src/circuit/tests.rs

@@ -1040,6 +1040,10 @@ mod conv {
             let a = VarTensor::new_advice(cs, K, 1, (LEN + 1) * LEN);
             let b = VarTensor::new_advice(cs, K, 1, (LEN + 1) * LEN);
             let output = VarTensor::new_advice(cs, K, 1, (LEN + 1) * LEN);
+
+            // column for constants
+            let _constant = VarTensor::constant_cols(cs, K, 8, false);
+
             Self::Config::configure(cs, &[a, b], &output, CheckMode::SAFE)
         }
 
@@ -1171,7 +1175,7 @@ mod conv_col_ultra_overflow {
 
     use super::*;
 
-    const K: usize = 4;
+    const K: usize = 6;
     const LEN: usize = 10;
 
     #[derive(Clone)]
@@ -1191,9 +1195,10 @@ mod conv_col_ultra_overflow {
         }
 
         fn configure(cs: &mut ConstraintSystem<F>) -> Self::Config {
-            let a = VarTensor::new_advice(cs, K, 1, LEN * LEN * LEN);
-            let b = VarTensor::new_advice(cs, K, 1, LEN * LEN * LEN);
-            let output = VarTensor::new_advice(cs, K, 1, LEN * LEN * LEN);
+            let a = VarTensor::new_advice(cs, K, 1, LEN * LEN * LEN * LEN);
+            let b = VarTensor::new_advice(cs, K, 1, LEN * LEN * LEN * LEN);
+            let output = VarTensor::new_advice(cs, K, 1, LEN * LEN * LEN * LEN);
+            let _constant = VarTensor::constant_cols(cs, K, LEN * LEN * LEN * LEN, false);
             Self::Config::configure(cs, &[a, b], &output, CheckMode::SAFE)
         }
 
@@ -1776,13 +1781,18 @@ mod shuffle {
 
             let d = VarTensor::new_advice(cs, K, 1, LEN);
             let e = VarTensor::new_advice(cs, K, 1, LEN);
+            let f: VarTensor = VarTensor::new_advice(cs, K, 1, LEN);
 
             let _constant = VarTensor::constant_cols(cs, K, LEN * NUM_LOOP, false);
 
             let mut config =
                 Self::Config::configure(cs, &[a.clone(), b.clone()], &c, CheckMode::SAFE);
             config
-                .configure_shuffles(cs, &[a.clone(), b.clone()], &[d.clone(), e.clone()])
+                .configure_shuffles(
+                    cs,
+                    &[a.clone(), b.clone(), c.clone()],
+                    &[d.clone(), e.clone(), f.clone()],
+                )
                 .unwrap();
             config
         }

src/commands.rs

@@ -83,7 +83,7 @@ pub const DEFAULT_VK_SOL: &str = "vk.sol";
 /// Default VK abi path
 pub const DEFAULT_VK_ABI: &str = "vk.abi";
 /// Default scale rebase multipliers for calibration
-pub const DEFAULT_SCALE_REBASE_MULTIPLIERS: &str = "1,2,10";
+pub const DEFAULT_SCALE_REBASE_MULTIPLIERS: &str = "1,10";
 /// Default use reduced srs for verification
 pub const DEFAULT_USE_REDUCED_SRS_FOR_VERIFICATION: &str = "false";
 /// Default only check for range check rebase

src/fieldutils.rs

@@ -5,7 +5,7 @@ use halo2curves::ff::PrimeField;
 /// Integer representation of a PrimeField element.
 pub type IntegerRep = i128;
 
-/// Converts an i64 to a PrimeField element.
+/// Converts an integer rep to a PrimeField element.
 pub fn integer_rep_to_felt<F: PrimeField>(x: IntegerRep) -> F {
     if x >= 0 {
         F::from_u128(x as u128)
@@ -69,7 +69,7 @@ mod test {
     fn felttointegerrep() {
         for x in -(2_i128.pow(16))..(2_i128.pow(16)) {
             let fieldx: F = integer_rep_to_felt::<F>(x);
-            let xf: i128 = felt_to_integer_rep::<F>(fieldx);
+            let xf: IntegerRep = felt_to_integer_rep::<F>(fieldx);
             assert_eq!(x, xf);
         }
     }

src/graph/errors.rs

@@ -11,6 +11,12 @@ pub enum GraphError {
     /// Shape mismatch in circuit construction
     #[error("invalid dimensions used for node {0} ({1})")]
     InvalidDims(usize, String),
+    /// Non scalar power
+    #[error("we only support scalar powers")]
+    NonScalarPower,
+    /// Non scalar base for exponentiation
+    #[error("we only support scalar bases for exponentiation")]
+    NonScalarBase,
     /// Wrong method was called to configure an op
     #[error("wrong method was called to configure node {0} ({1})")]
     WrongMethod(usize, String),
@@ -143,4 +149,7 @@ pub enum GraphError {
     /// Invalid RunArg
     #[error("invalid RunArgs: {0}")]
     InvalidRunArgs(String),
+    /// Only nearest neighbor interpolation is supported
+    #[error("only nearest neighbor interpolation is supported")]
+    InvalidInterpolation,
 }

src/graph/input.rs

@@ -14,7 +14,6 @@ use pyo3::prelude::*;
 use pyo3::types::PyDict;
 #[cfg(feature = "python-bindings")]
 use pyo3::ToPyObject;
-use serde::ser::SerializeStruct;
 use serde::{Deserialize, Deserializer, Serialize, Serializer};
 use std::io::BufReader;
 use std::io::BufWriter;
@@ -515,7 +514,7 @@ impl<'de> Deserialize<'de> for DataSource {
 
 /// Input to graph as a datasource
 /// Always use JSON serialization for GraphData. Seriously.
-#[derive(Clone, Debug, Deserialize, Default, PartialEq)]
+#[derive(Clone, Debug, Deserialize, Default, PartialEq, Serialize)]
 pub struct GraphData {
     /// Inputs to the model / computational graph (can be empty vectors if inputs are coming from on-chain).
     pub input_data: DataSource,
@@ -769,18 +768,6 @@ impl ToPyObject for FileSourceInner {
     }
 }
 
-impl Serialize for GraphData {
-    fn serialize<S>(&self, serializer: S) -> Result<S::Ok, S::Error>
-    where
-        S: Serializer,
-    {
-        let mut state = serializer.serialize_struct("GraphData", 4)?;
-        state.serialize_field("input_data", &self.input_data)?;
-        state.serialize_field("output_data", &self.output_data)?;
-        state.end()
-    }
-}
-
 #[cfg(test)]
 mod tests {
     use super::*;

src/graph/model.rs

@@ -1045,8 +1045,8 @@ impl Model {
         if settings.requires_shuffle() {
             base_gate.configure_shuffles(
                 meta,
-                vars.advices[0..2].try_into()?,
-                vars.advices[3..5].try_into()?,
+                vars.advices[0..3].try_into()?,
+                vars.advices[3..6].try_into()?,
             )?;
         }
 

src/graph/utilities.rs

@@ -44,11 +44,10 @@ use tract_onnx::tract_hir::{
     tract_core::ops::cnn::{conv::KernelFormat, MaxPool, SumPool},
 };
 
-/// Quantizes an iterable of f32s to a [Tensor] of i32s using a fixed point representation.
+/// Quantizes an iterable of f64 to a [Tensor] of IntegerRep using a fixed point representation.
 /// Arguments
 ///
-/// * `vec` - the vector to quantize.
-/// * `dims` - the dimensionality of the resulting [Tensor].
+/// * `elem` - the element to quantize.
 /// * `shift` - offset used in the fixed point representation.
 /// * `scale` - `2^scale` used in the fixed point representation.
 pub fn quantize_float(
@@ -85,7 +84,7 @@ pub fn scale_to_multiplier(scale: crate::Scale) -> f64 {
     f64::powf(2., scale as f64)
 }
 
-/// Converts a scale (log base 2) to a fixed point multiplier.
+/// Converts a fixed point multiplier to a scale (log base 2).
 pub fn multiplier_to_scale(mult: f64) -> crate::Scale {
     mult.log2().round() as crate::Scale
 }
@@ -312,6 +311,9 @@ pub fn new_op_from_onnx(
     let mut deleted_indices = vec![];
     let node = match node.op().name().as_ref() {
         "ShiftLeft" => {
+            if inputs.len() != 2 {
+                return Err(GraphError::InvalidDims(idx, "shift left".to_string()));
+            };
             // load shift amount
             if let Some(c) = inputs[1].opkind().get_mutable_constant() {
                 inputs[1].decrement_use();
@@ -324,10 +326,13 @@ pub fn new_op_from_onnx(
                     out_scale: Some(input_scales[0] - raw_values[0] as i32),
                 })
             } else {
-                return Err(GraphError::OpMismatch(idx, "ShiftLeft".to_string()));
+                return Err(GraphError::OpMismatch(idx, "shift left".to_string()));
             }
         }
         "ShiftRight" => {
+            if inputs.len() != 2 {
+                return Err(GraphError::InvalidDims(idx, "shift right".to_string()));
+            };
             // load shift amount
             if let Some(c) = inputs[1].opkind().get_mutable_constant() {
                 inputs[1].decrement_use();
@@ -340,7 +345,7 @@ pub fn new_op_from_onnx(
                     out_scale: Some(input_scales[0] + raw_values[0] as i32),
                 })
             } else {
-                return Err(GraphError::OpMismatch(idx, "ShiftRight".to_string()));
+                return Err(GraphError::OpMismatch(idx, "shift right".to_string()));
             }
         }
         "MultiBroadcastTo" => {
@@ -363,7 +368,10 @@ pub fn new_op_from_onnx(
                 }
             }
 
-            assert_eq!(input_ops.len(), 3, "Range requires 3 inputs");
+            if input_ops.len() != 3 {
+                return Err(GraphError::InvalidDims(idx, "range".to_string()));
+            }
+
             let input_ops = input_ops
                 .iter()
                 .map(|x| x.get_constant().ok_or(GraphError::NonConstantRange))
@@ -419,6 +427,10 @@ pub fn new_op_from_onnx(
             if let Some(c) = inputs[1].opkind().get_mutable_constant() {
                 inputs[1].decrement_use();
                 deleted_indices.push(inputs.len() - 1);
+                if inputs[0].out_dims().is_empty() || inputs[0].out_dims()[0].len() <= axis {
+                    return Err(GraphError::InvalidDims(idx, "gather".to_string()));
+                }
+
                 op = SupportedOp::Hybrid(crate::circuit::ops::hybrid::HybridOp::Gather {
                     dim: axis,
                     constant_idx: Some(c.raw_values.map(|x| {
@@ -447,8 +459,17 @@ pub fn new_op_from_onnx(
         "Topk" => {
             let op = load_op::<Topk>(node.op(), idx, node.op().name().to_string())?;
             let axis = op.axis;
+
+            if inputs.len() != 2 {
+                return Err(GraphError::InvalidDims(idx, "topk".to_string()));
+            };
+
             // if param_visibility.is_public() {
             let k = if let Some(c) = inputs[1].opkind().get_mutable_constant() {
+                if c.raw_values.len() != 1 {
+                    return Err(GraphError::InvalidDims(idx, "topk".to_string()));
+                }
+
                 inputs[1].decrement_use();
                 deleted_indices.push(inputs.len() - 1);
                 c.raw_values.map(|x| x as usize)[0]
@@ -488,6 +509,10 @@ pub fn new_op_from_onnx(
             if let Some(c) = inputs[1].opkind().get_mutable_constant() {
                 inputs[1].decrement_use();
                 deleted_indices.push(1);
+                if c.raw_values.is_empty() {
+                    return Err(GraphError::InvalidDims(idx, "scatter elements".to_string()));
+                }
+
                 op = SupportedOp::Linear(crate::circuit::ops::poly::PolyOp::ScatterElements {
                     dim: axis,
                     constant_idx: Some(c.raw_values.map(|x| x as usize)),
@@ -522,6 +547,9 @@ pub fn new_op_from_onnx(
             if let Some(c) = inputs[1].opkind().get_mutable_constant() {
                 inputs[1].decrement_use();
                 deleted_indices.push(1);
+                if c.raw_values.is_empty() {
+                    return Err(GraphError::InvalidDims(idx, "scatter nd".to_string()));
+                }
                 op = SupportedOp::Linear(crate::circuit::ops::poly::PolyOp::ScatterND {
                     constant_idx: Some(c.raw_values.map(|x| x as usize)),
                 })
@@ -555,6 +583,9 @@ pub fn new_op_from_onnx(
             if let Some(c) = inputs[1].opkind().get_mutable_constant() {
                 inputs[1].decrement_use();
                 deleted_indices.push(1);
+                if c.raw_values.is_empty() {
+                    return Err(GraphError::InvalidDims(idx, "gather nd".to_string()));
+                }
                 op = SupportedOp::Linear(crate::circuit::ops::poly::PolyOp::GatherND {
                     batch_dims,
                     indices: Some(c.raw_values.map(|x| x as usize)),
@@ -589,6 +620,9 @@ pub fn new_op_from_onnx(
             if let Some(c) = inputs[1].opkind().get_mutable_constant() {
                 inputs[1].decrement_use();
                 deleted_indices.push(1);
+                if c.raw_values.is_empty() {
+                    return Err(GraphError::InvalidDims(idx, "gather elements".to_string()));
+                }
                 op = SupportedOp::Linear(crate::circuit::ops::poly::PolyOp::GatherElements {
                     dim: axis,
                     constant_idx: Some(c.raw_values.map(|x| x as usize)),
@@ -684,7 +718,9 @@ pub fn new_op_from_onnx(
             };
             let op = load_op::<Reduce>(node.op(), idx, node.op().name().to_string())?;
             let axes: Vec<usize> = op.axes.into_iter().collect();
-            assert_eq!(axes.len(), 1, "only support argmax over one axis");
+            if axes.len() != 1 {
+                return Err(GraphError::InvalidDims(idx, "argmax".to_string()));
+            }
 
             SupportedOp::Hybrid(HybridOp::ReduceArgMax { dim: axes[0] })
         }
@@ -694,7 +730,9 @@ pub fn new_op_from_onnx(
             };
             let op = load_op::<Reduce>(node.op(), idx, node.op().name().to_string())?;
             let axes: Vec<usize> = op.axes.into_iter().collect();
-            assert_eq!(axes.len(), 1, "only support argmin over one axis");
+            if axes.len() != 1 {
+                return Err(GraphError::InvalidDims(idx, "argmin".to_string()));
+            }
 
             SupportedOp::Hybrid(HybridOp::ReduceArgMin { dim: axes[0] })
         }
@@ -803,6 +841,9 @@ pub fn new_op_from_onnx(
             }
         }
         "Recip" => {
+            if inputs.len() != 1 {
+                return Err(GraphError::InvalidDims(idx, "recip".to_string()));
+            };
             let in_scale = input_scales[0];
             let max_scale = std::cmp::max(scales.get_max(), in_scale);
             // If the input scale is larger than the params scale
@@ -846,6 +887,9 @@ pub fn new_op_from_onnx(
             scale: scale_to_multiplier(input_scales[0]).into(),
         }),
         "Rsqrt" => {
+            if input_scales.len() != 1 {
+                return Err(GraphError::InvalidDims(idx, "rsqrt".to_string()));
+            };
             let in_scale = input_scales[0];
             let max_scale = std::cmp::max(scales.get_max(), in_scale);
             SupportedOp::Hybrid(HybridOp::Rsqrt {
@@ -933,7 +977,9 @@ pub fn new_op_from_onnx(
             let op = load_op::<Cast>(node.op(), idx, node.op().name().to_string())?;
             let dt = op.to;
 
-            assert_eq!(input_scales.len(), 1);
+            if input_scales.len() != 1 {
+                return Err(GraphError::InvalidDims(idx, "cast".to_string()));
+            };
 
             match dt {
                 DatumType::Bool
@@ -983,6 +1029,11 @@ pub fn new_op_from_onnx(
 
             if const_idx.len() == 1 {
                 let const_idx = const_idx[0];
+
+                if inputs.len() <= const_idx {
+                    return Err(GraphError::InvalidDims(idx, "mul".to_string()));
+                }
+
                 if let Some(c) = inputs[const_idx].opkind().get_mutable_constant() {
                     if c.raw_values.len() == 1 && c.raw_values[0] < 1. {
                         // if not divisible by 2 then we need to add a range check
@@ -1057,6 +1108,9 @@ pub fn new_op_from_onnx(
                     return Err(GraphError::OpMismatch(idx, "softmax".to_string()));
                 }
             };
+            if input_scales.len() != 1 {
+                return Err(GraphError::InvalidDims(idx, "softmax".to_string()));
+            }
 
             let in_scale = input_scales[0];
             let max_scale = std::cmp::max(scales.get_max(), in_scale);
@@ -1096,22 +1150,42 @@ pub fn new_op_from_onnx(
                 pool_dims: kernel_shape.to_vec(),
             })
         }
-        "Ceil" => SupportedOp::Hybrid(HybridOp::Ceil {
-            scale: scale_to_multiplier(input_scales[0]).into(),
-            legs: run_args.decomp_legs,
-        }),
-        "Floor" => SupportedOp::Hybrid(HybridOp::Floor {
-            scale: scale_to_multiplier(input_scales[0]).into(),
-            legs: run_args.decomp_legs,
-        }),
-        "Round" => SupportedOp::Hybrid(HybridOp::Round {
-            scale: scale_to_multiplier(input_scales[0]).into(),
-            legs: run_args.decomp_legs,
-        }),
-        "RoundHalfToEven" => SupportedOp::Hybrid(HybridOp::RoundHalfToEven {
-            scale: scale_to_multiplier(input_scales[0]).into(),
-            legs: run_args.decomp_legs,
-        }),
+        "Ceil" => {
+            if input_scales.len() != 1 {
+                return Err(GraphError::InvalidDims(idx, "ceil".to_string()));
+            }
+            SupportedOp::Hybrid(HybridOp::Ceil {
+                scale: scale_to_multiplier(input_scales[0]).into(),
+                legs: run_args.decomp_legs,
+            })
+        }
+        "Floor" => {
+            if input_scales.len() != 1 {
+                return Err(GraphError::InvalidDims(idx, "floor".to_string()));
+            }
+            SupportedOp::Hybrid(HybridOp::Floor {
+                scale: scale_to_multiplier(input_scales[0]).into(),
+                legs: run_args.decomp_legs,
+            })
+        }
+        "Round" => {
+            if input_scales.len() != 1 {
+                return Err(GraphError::InvalidDims(idx, "round".to_string()));
+            }
+            SupportedOp::Hybrid(HybridOp::Round {
+                scale: scale_to_multiplier(input_scales[0]).into(),
+                legs: run_args.decomp_legs,
+            })
+        }
+        "RoundHalfToEven" => {
+            if input_scales.len() != 1 {
+                return Err(GraphError::InvalidDims(idx, "roundhalftoeven".to_string()));
+            }
+            SupportedOp::Hybrid(HybridOp::RoundHalfToEven {
+                scale: scale_to_multiplier(input_scales[0]).into(),
+                legs: run_args.decomp_legs,
+            })
+        }
         "Sign" => SupportedOp::Linear(PolyOp::Sign),
         "Pow" => {
             // Extract the slope layer hyperparams from a const
@@ -1121,7 +1195,9 @@ pub fn new_op_from_onnx(
                 inputs[1].decrement_use();
                 deleted_indices.push(1);
                 if c.raw_values.len() > 1 {
-                    unimplemented!("only support scalar pow")
+                    return Err(GraphError::NonScalarPower);
+                } else if c.raw_values.is_empty() {
+                    return Err(GraphError::InvalidDims(idx, "pow".to_string()));
                 }
 
                 let exponent = c.raw_values[0];
@@ -1138,7 +1214,9 @@ pub fn new_op_from_onnx(
                 inputs[0].decrement_use();
                 deleted_indices.push(0);
                 if c.raw_values.len() > 1 {
-                    unimplemented!("only support scalar base")
+                    return Err(GraphError::NonScalarBase);
+                } else if c.raw_values.is_empty() {
+                    return Err(GraphError::InvalidDims(idx, "pow".to_string()));
                 }
 
                 let base = c.raw_values[0];
@@ -1148,10 +1226,14 @@ pub fn new_op_from_onnx(
                     base: base.into(),
                 })
             } else {
-                unimplemented!("only support constant base or pow for now")
+                return Err(GraphError::InvalidDims(idx, "pow".to_string()));
             }
         }
         "Div" => {
+            if inputs.len() != 2 {
+                return Err(GraphError::InvalidDims(idx, "div".to_string()));
+            }
+
             let const_idx = inputs
                 .iter()
                 .enumerate()
@@ -1159,14 +1241,15 @@ pub fn new_op_from_onnx(
                 .map(|(i, _)| i)
                 .collect::<Vec<_>>();
 
-            if const_idx.len() > 1 {
+            if const_idx.len() > 1 || const_idx.is_empty() {
                 return Err(GraphError::InvalidDims(idx, "div".to_string()));
             }
 
             let const_idx = const_idx[0];
-
             if const_idx != 1 {
-                unimplemented!("only support div with constant as second input")
+                return Err(GraphError::MisformedParams(
+                    "only support div with constant as second input".to_string(),
+                ));
             }
 
             if let Some(c) = inputs[const_idx].opkind().get_mutable_constant() {
@@ -1180,10 +1263,14 @@ pub fn new_op_from_onnx(
                         denom: denom.into(),
                     })
                 } else {
-                    unimplemented!("only support non zero divisors of size 1")
+                    return Err(GraphError::MisformedParams(
+                        "only support non zero divisors of size 1".to_string(),
+                    ));
                 }
             } else {
-                unimplemented!("only support div with constant as second input")
+                return Err(GraphError::MisformedParams(
+                    "only support div with constant as second input".to_string(),
+                ));
             }
         }
         "Cube" => SupportedOp::Linear(PolyOp::Pow(3)),
@@ -1323,7 +1410,7 @@ pub fn new_op_from_onnx(
             if !resize_node.contains("interpolator: Nearest")
                 && !resize_node.contains("nearest: Floor")
             {
-                unimplemented!("Only nearest neighbor interpolation is supported")
+                return Err(GraphError::InvalidInterpolation);
             }
             // check if optional scale factor is present
             if inputs.len() != 2 && inputs.len() != 3 {
@@ -1427,6 +1514,10 @@ pub fn new_op_from_onnx(
             SupportedOp::Linear(PolyOp::Reshape(output_shape))
         }
         "Flatten" => {
+            if inputs.len() != 1 || inputs[0].out_dims().is_empty() {
+                return Err(GraphError::InvalidDims(idx, "flatten".to_string()));
+            };
+
             let new_dims: Vec<usize> = vec![inputs[0].out_dims()[0].iter().product::<usize>()];
             SupportedOp::Linear(PolyOp::Flatten(new_dims))
         }
@@ -1546,6 +1637,7 @@ pub fn homogenize_input_scales(
 }
 
 #[cfg(test)]
+/// tests for the utility module
 pub mod tests {
 
     use super::*;

src/graph/vars.rs

@@ -435,7 +435,7 @@ impl<F: PrimeField + TensorType + PartialOrd + std::hash::Hash> ModelVars<F> {
             .collect_vec();
 
         if requires_dynamic_lookup || requires_shuffle {
-            let num_cols = if requires_dynamic_lookup { 3 } else { 2 };
+            let num_cols = 3;
             for _ in 0..num_cols {
                 let dynamic_lookup =
                     VarTensor::new_advice(cs, logrows, 1, dynamic_lookup_and_shuffle_size);

src/tensor/mod.rs

@@ -24,9 +24,6 @@ use std::path::PathBuf;
 pub use val::*;
 pub use var::*;
 
-#[cfg(feature = "metal")]
-use instant::Instant;
-
 use crate::{
     circuit::utils,
     fieldutils::{integer_rep_to_felt, IntegerRep},
@@ -40,8 +37,6 @@ use halo2_proofs::{
     poly::Rotation,
 };
 use itertools::Itertools;
-#[cfg(feature = "metal")]
-use metal::{Device, MTLResourceOptions, MTLSize};
 use std::error::Error;
 use std::fmt::Debug;
 use std::io::Read;
@@ -49,31 +44,6 @@ use std::iter::Iterator;
 use std::ops::{Add, Deref, DerefMut, Div, Mul, Neg, Range, Sub};
 use std::{cmp::max, ops::Rem};
 
-#[cfg(feature = "metal")]
-use std::collections::HashMap;
-
-#[cfg(feature = "metal")]
-const LIB_DATA: &[u8] = include_bytes!("metal/tensor_ops.metallib");
-
-#[cfg(feature = "metal")]
-lazy_static::lazy_static! {
-    static ref DEVICE: Device = Device::system_default().expect("no device found");
-
-    static ref LIB: metal::Library = DEVICE.new_library_with_data(LIB_DATA).unwrap();
-
-    static ref QUEUE: metal::CommandQueue = DEVICE.new_command_queue();
-
-    static ref PIPELINES: HashMap<String, metal::ComputePipelineState> = {
-        let mut map = HashMap::new();
-        for name in ["add", "sub", "mul"] {
-            let function = LIB.get_function(name, None).unwrap();
-            let pipeline = DEVICE.new_compute_pipeline_state_with_function(&function).unwrap();
-            map.insert(name.to_string(), pipeline);
-        }
-        map
-    };
-}
-
 /// The (inner) type of tensor elements.
 pub trait TensorType: Clone + Debug + 'static {
     /// Returns the zero value.
@@ -1404,10 +1374,6 @@ impl<T: TensorType + Add<Output = T> + std::marker::Send + std::marker::Sync> Ad
         let lhs = self.expand(&broadcasted_shape).unwrap();
         let rhs = rhs.expand(&broadcasted_shape).unwrap();
 
-        #[cfg(feature = "metal")]
-        let res = metal_tensor_op(&lhs, &rhs, "add");
-
-        #[cfg(not(feature = "metal"))]
         let res = {
             let mut res: Tensor<T> = lhs
                 .par_iter()
@@ -1505,10 +1471,6 @@ impl<T: TensorType + Sub<Output = T> + std::marker::Send + std::marker::Sync> Su
         let lhs = self.expand(&broadcasted_shape).unwrap();
         let rhs = rhs.expand(&broadcasted_shape).unwrap();
 
-        #[cfg(feature = "metal")]
-        let res = metal_tensor_op(&lhs, &rhs, "sub");
-
-        #[cfg(not(feature = "metal"))]
         let res = {
             let mut res: Tensor<T> = lhs
                 .par_iter()
@@ -1576,10 +1538,6 @@ impl<T: TensorType + Mul<Output = T> + std::marker::Send + std::marker::Sync> Mu
         let lhs = self.expand(&broadcasted_shape).unwrap();
         let rhs = rhs.expand(&broadcasted_shape).unwrap();
 
-        #[cfg(feature = "metal")]
-        let res = metal_tensor_op(&lhs, &rhs, "mul");
-
-        #[cfg(not(feature = "metal"))]
         let res = {
             let mut res: Tensor<T> = lhs
                 .par_iter()
@@ -1811,66 +1769,4 @@ mod tests {
         let b = Tensor::<IntegerRep>::new(Some(&[1, 4]), &[2, 1]).unwrap();
         assert_eq!(a.get_slice(&[0..2, 0..1]).unwrap(), b);
     }
-
-    #[test]
-    #[cfg(feature = "metal")]
-    fn tensor_metal_int() {
-        let a = Tensor::<i64>::new(Some(&[1, 2, 3, 4]), &[2, 2]).unwrap();
-        let b = Tensor::<i64>::new(Some(&[1, 2, 3, 4]), &[2, 2]).unwrap();
-        let c = metal_tensor_op(&a, &b, "add");
-        assert_eq!(c, Tensor::new(Some(&[2, 4, 6, 8]), &[2, 2]).unwrap());
-
-        let c = metal_tensor_op(&a, &b, "sub");
-        assert_eq!(c, Tensor::new(Some(&[0, 0, 0, 0]), &[2, 2]).unwrap());
-
-        let c = metal_tensor_op(&a, &b, "mul");
-        assert_eq!(c, Tensor::new(Some(&[1, 4, 9, 16]), &[2, 2]).unwrap());
-    }
-
-    #[test]
-    #[cfg(feature = "metal")]
-    fn tensor_metal_felt() {
-        use halo2curves::bn256::Fr;
-
-        let a = Tensor::<Fr>::new(
-            Some(&[Fr::from(1), Fr::from(2), Fr::from(3), Fr::from(4)]),
-            &[2, 2],
-        )
-        .unwrap();
-        let b = Tensor::<Fr>::new(
-            Some(&[Fr::from(1), Fr::from(2), Fr::from(3), Fr::from(4)]),
-            &[2, 2],
-        )
-        .unwrap();
-
-        let c = metal_tensor_op(&a, &b, "add");
-        assert_eq!(
-            c,
-            Tensor::<Fr>::new(
-                Some(&[Fr::from(2), Fr::from(4), Fr::from(6), Fr::from(8)]),
-                &[2, 2],
-            )
-            .unwrap()
-        );
-
-        let c = metal_tensor_op(&a, &b, "sub");
-        assert_eq!(
-            c,
-            Tensor::<Fr>::new(
-                Some(&[Fr::from(0), Fr::from(0), Fr::from(0), Fr::from(0)]),
-                &[2, 2],
-            )
-            .unwrap()
-        );
-
-        let c = metal_tensor_op(&a, &b, "mul");
-        assert_eq!(
-            c,
-            Tensor::<Fr>::new(
-                Some(&[Fr::from(1), Fr::from(4), Fr::from(9), Fr::from(16)]),
-                &[2, 2],
-            )
-            .unwrap()
-        );
-    }
 }

src/tensor/ops.rs

@@ -27,7 +27,7 @@ pub fn get_rep(
     n: usize,
 ) -> Result<Vec<IntegerRep>, DecompositionError> {
     // check if x is too large
-    if x.abs() > (base.pow(n as u32) as IntegerRep) - 1 {
+    if (*x).abs() > ((base as i128).pow(n as u32)) - 1 {
         return Err(DecompositionError::TooLarge(*x, base, n));
     }
     let mut rep = vec![0; n + 1];
@@ -43,8 +43,8 @@ pub fn get_rep(
     let mut x = x.abs();
     //
     for i in (1..rep.len()).rev() {
-        rep[i] = x % base as i128;
-        x /= base as i128;
+        rep[i] = x % base as IntegerRep;
+        x /= base as IntegerRep;
     }
 
     Ok(rep)
@@ -127,7 +127,7 @@ pub fn decompose(
         .flatten()
         .collect::<Vec<IntegerRep>>();
 
-    let output = Tensor::<i128>::new(Some(&resp), &dims)?;
+    let output = Tensor::<IntegerRep>::new(Some(&resp), &dims)?;
 
     Ok(output)
 }

tests/integration_tests.rs

@@ -187,13 +187,14 @@ mod native_tests {
 
     const PF_FAILURE_AGGR: &str = "examples/test_failure_aggr_proof.json";
 
-    const LARGE_TESTS: [&str; 6] = [
+    const LARGE_TESTS: [&str; 7] = [
         "self_attention",
         "nanoGPT",
         "multihead_attention",
         "mobilenet",
         "mnist_gan",
         "smallworm",
+        "fr_age",
     ];
 
     const ACCURACY_CAL_TESTS: [&str; 6] = [
@@ -395,29 +396,29 @@ mod native_tests {
     const TESTS_AGGR: [&str; 3] = ["1l_mlp", "1l_flatten", "1l_average"];
 
     const TESTS_EVM: [&str; 23] = [
-        "1l_mlp",
-        "1l_flatten",
-        "1l_average",
-        "1l_reshape",
-        "1l_sigmoid",
-        "1l_div",
-        "1l_sqrt",
-        "1l_prelu",
-        "1l_var",
-        "1l_leakyrelu",
-        "1l_gelu_noappx",
-        "1l_relu",
-        "1l_tanh",
-        "2l_relu_sigmoid_small",
-        "2l_relu_small",
-        "min",
-        "max",
-        "1l_max_pool",
-        "idolmodel",
-        "1l_identity",
-        "lstm",
-        "rnn",
-        "quantize_dequantize",
+        "1l_mlp",                // 0
+        "1l_flatten",            // 1
+        "1l_average",            // 2
+        "1l_reshape",            // 3
+        "1l_sigmoid",            // 4
+        "1l_div",                // 5
+        "1l_sqrt",               // 6
+        "1l_prelu",              // 7
+        "1l_var",                // 8
+        "1l_leakyrelu",          // 9
+        "1l_gelu_noappx",        // 10
+        "1l_relu",               // 11
+        "1l_tanh",               // 12
+        "2l_relu_sigmoid_small", // 13
+        "2l_relu_small",         // 14
+        "min",                   // 15
+        "max",                   // 16
+        "1l_max_pool",           // 17
+        "idolmodel",             // 18
+        "1l_identity",           // 19
+        "lstm",                  // 20
+        "rnn",                   // 21
+        "quantize_dequantize",   // 22
     ];
 
     const TESTS_EVM_AGGR: [&str; 18] = [
@@ -541,7 +542,7 @@ mod native_tests {
                 crate::native_tests::init_binary();
                 let test_dir = TempDir::new(test).unwrap();
                 let path = test_dir.path().to_str().unwrap(); crate::native_tests::mv_test_(path, test);
-                mock(path, test.to_string(), "public", "fixed", "public", 1, "accuracy", None, 0.0, false);
+                mock(path, test.to_string(), "public", "fixed", "public", 1, "accuracy", None, 0.0, false, None, None);
                 test_dir.close().unwrap();
             }
         });
@@ -606,7 +607,7 @@ mod native_tests {
                 crate::native_tests::init_binary();
                 let test_dir = TempDir::new(test).unwrap();
                 let path = test_dir.path().to_str().unwrap(); crate::native_tests::mv_test_(path, test);
-                mock(path, test.to_string(), "private", "private", "public", 1, "resources", None, 0.0, false);
+                mock(path, test.to_string(), "private", "private", "public", 1, "resources", None, 0.0, false, None, None);
                 test_dir.close().unwrap();
             }
 
@@ -616,7 +617,7 @@ mod native_tests {
                 crate::native_tests::init_binary();
                 let test_dir = TempDir::new(test).unwrap();
                 let path = test_dir.path().to_str().unwrap(); crate::native_tests::mv_test_(path, test);
-                mock(path, test.to_string(), "private", "private", "public", 1, "resources", None, 0.0, true);
+                mock(path, test.to_string(), "private", "private", "public", 1, "resources", None, 0.0, true, Some(8194), Some(4));
                 test_dir.close().unwrap();
             }
 
@@ -627,7 +628,7 @@ mod native_tests {
                 let path = test_dir.path().to_str().unwrap(); crate::native_tests::mv_test_(path, test);
                 // gen random number between 0.0 and 1.0
                 let tolerance = rand::thread_rng().gen_range(0.0..1.0) * 100.0;
-                mock(path, test.to_string(), "private", "private", "public", 1, "resources", None, tolerance, false);
+                mock(path, test.to_string(), "private", "private", "public", 1, "resources", None, tolerance, false, Some(8194), Some(5));
                 test_dir.close().unwrap();
             }
 
@@ -642,7 +643,7 @@ mod native_tests {
                     let path = test_dir.path().to_str().unwrap(); crate::native_tests::mv_test_(path, test);
                     let large_batch_dir = &format!("large_batches_{}", test);
                     crate::native_tests::mk_data_batches_(path, test, &large_batch_dir, 10);
-                    mock(path, large_batch_dir.to_string(), "private", "private", "public", 10, "resources", None, 0.0, false);
+                    mock(path, large_batch_dir.to_string(), "private", "private", "public", 10, "resources", None, 0.0, false, None, None);
                     test_dir.close().unwrap();
                 }
             }
@@ -652,7 +653,7 @@ mod native_tests {
                 crate::native_tests::init_binary();
                 let test_dir = TempDir::new(test).unwrap();
                 let path = test_dir.path().to_str().unwrap(); crate::native_tests::mv_test_(path, test);
-                mock(path, test.to_string(), "public", "private", "private", 1, "resources", None, 0.0, false);
+                mock(path, test.to_string(), "public", "private", "private", 1, "resources", None, 0.0, false, None, None);
                 test_dir.close().unwrap();
             }
 
@@ -661,7 +662,7 @@ mod native_tests {
                 crate::native_tests::init_binary();
                 let test_dir = TempDir::new(test).unwrap();
                 let path = test_dir.path().to_str().unwrap(); crate::native_tests::mv_test_(path, test);
-                mock(path, test.to_string(), "public", "hashed", "private", 1, "resources", None, 0.0, false);
+                mock(path, test.to_string(), "public", "hashed", "private", 1, "resources", None, 0.0, false, None, None);
                 test_dir.close().unwrap();
             }
 
@@ -670,7 +671,7 @@ mod native_tests {
                 crate::native_tests::init_binary();
                 let test_dir = TempDir::new(test).unwrap();
                 let path = test_dir.path().to_str().unwrap(); crate::native_tests::mv_test_(path, test);
-                mock(path, test.to_string(), "fixed", "private", "private", 1, "resources", None, 0.0, false);
+                mock(path, test.to_string(), "fixed", "private", "private", 1, "resources", None, 0.0, false, None, None);
                 test_dir.close().unwrap();
             }
 
@@ -679,7 +680,7 @@ mod native_tests {
                 crate::native_tests::init_binary();
                 let test_dir = TempDir::new(test).unwrap();
                 let path = test_dir.path().to_str().unwrap(); crate::native_tests::mv_test_(path, test);
-                mock(path, test.to_string(), "private", "private", "fixed", 1, "resources", None, 0.0, false);
+                mock(path, test.to_string(), "private", "private", "fixed", 1, "resources", None, 0.0, false, None, None);
                 test_dir.close().unwrap();
             }
 
@@ -688,7 +689,7 @@ mod native_tests {
                 crate::native_tests::init_binary();
                 let test_dir = TempDir::new(test).unwrap();
                 let path = test_dir.path().to_str().unwrap(); crate::native_tests::mv_test_(path, test);
-                mock(path, test.to_string(), "private", "fixed", "private", 1, "resources", None, 0.0, false);
+                mock(path, test.to_string(), "private", "fixed", "private", 1, "resources", None, 0.0, false, None, None);
                 test_dir.close().unwrap();
             }
 
@@ -697,7 +698,7 @@ mod native_tests {
                 crate::native_tests::init_binary();
                 let test_dir = TempDir::new(test).unwrap();
                 let path = test_dir.path().to_str().unwrap(); crate::native_tests::mv_test_(path, test);
-                mock(path, test.to_string(), "hashed", "private", "public", 1, "resources", None, 0.0, false);
+                mock(path, test.to_string(), "hashed", "private", "public", 1, "resources", None, 0.0, false, None, None);
                 test_dir.close().unwrap();
             }
 
@@ -706,7 +707,7 @@ mod native_tests {
                 crate::native_tests::init_binary();
                 let test_dir = TempDir::new(test).unwrap();
                 let path = test_dir.path().to_str().unwrap(); crate::native_tests::mv_test_(path, test);
-                mock(path, test.to_string(), "polycommit", "private", "public", 1, "resources", None, 0.0, false);
+                mock(path, test.to_string(), "polycommit", "private", "public", 1, "resources", None, 0.0, false, None, None);
                 test_dir.close().unwrap();
             }
 
@@ -716,7 +717,7 @@ mod native_tests {
                 crate::native_tests::init_binary();
                 let test_dir = TempDir::new(test).unwrap();
                 let path = test_dir.path().to_str().unwrap(); crate::native_tests::mv_test_(path, test);
-                mock(path, test.to_string(), "private", "hashed", "public", 1, "resources", None, 0.0, false);
+                mock(path, test.to_string(), "private", "hashed", "public", 1, "resources", None, 0.0, false, None, None);
                 test_dir.close().unwrap();
             }
 
@@ -726,7 +727,7 @@ mod native_tests {
                 crate::native_tests::init_binary();
                 let test_dir = TempDir::new(test).unwrap();
                 let path = test_dir.path().to_str().unwrap(); crate::native_tests::mv_test_(path, test);
-                mock(path, test.to_string(), "private", "polycommit", "public", 1, "resources", None, 0.0, false);
+                mock(path, test.to_string(), "private", "polycommit", "public", 1, "resources", None, 0.0, false, None, None);
                 test_dir.close().unwrap();
             }
 
@@ -735,7 +736,7 @@ mod native_tests {
                 crate::native_tests::init_binary();
                 let test_dir = TempDir::new(test).unwrap();
                 let path = test_dir.path().to_str().unwrap(); crate::native_tests::mv_test_(path, test);
-                mock(path, test.to_string(), "public", "private", "hashed", 1, "resources", None, 0.0, false);
+                mock(path, test.to_string(), "public", "private", "hashed", 1, "resources", None, 0.0, false, None, None);
                 test_dir.close().unwrap();
             }
 
@@ -745,7 +746,7 @@ mod native_tests {
                 crate::native_tests::init_binary();
                 let test_dir = TempDir::new(test).unwrap();
                 let path = test_dir.path().to_str().unwrap(); crate::native_tests::mv_test_(path, test);
-                mock(path, test.to_string(), "public", "private", "polycommit", 1, "resources", None, 0.0, false);
+                mock(path, test.to_string(), "public", "private", "polycommit", 1, "resources", None, 0.0, false, None, None);
                 test_dir.close().unwrap();
             }
 
@@ -754,7 +755,7 @@ mod native_tests {
                 crate::native_tests::init_binary();
                 let test_dir = TempDir::new(test).unwrap();
                 let path = test_dir.path().to_str().unwrap(); crate::native_tests::mv_test_(path, test);
-                mock(path, test.to_string(), "public", "fixed", "hashed", 1, "resources", None, 0.0, false);
+                mock(path, test.to_string(), "public", "fixed", "hashed", 1, "resources", None, 0.0, false, None, None);
                 test_dir.close().unwrap();
             }
 
@@ -764,7 +765,7 @@ mod native_tests {
                 crate::native_tests::init_binary();
                 let test_dir = TempDir::new(test).unwrap();
                 let path = test_dir.path().to_str().unwrap(); crate::native_tests::mv_test_(path, test);
-                mock(path, test.to_string(), "public", "polycommit", "hashed", 1, "resources", None, 0.0, false);
+                mock(path, test.to_string(), "public", "polycommit", "hashed", 1, "resources", None, 0.0, false, None, None);
                 test_dir.close().unwrap();
             }
 
@@ -774,7 +775,7 @@ mod native_tests {
                 crate::native_tests::init_binary();
                 let test_dir = TempDir::new(test).unwrap();
                 let path = test_dir.path().to_str().unwrap(); crate::native_tests::mv_test_(path, test);
-                mock(path, test.to_string(), "polycommit", "polycommit", "polycommit", 1, "resources", None, 0.0, false);
+                mock(path, test.to_string(), "polycommit", "polycommit", "polycommit", 1, "resources", None, 0.0, false, None, None);
                 test_dir.close().unwrap();
             }
 
@@ -784,7 +785,7 @@ mod native_tests {
                 crate::native_tests::init_binary();
                 let test_dir = TempDir::new(test).unwrap();
                 let path = test_dir.path().to_str().unwrap(); crate::native_tests::mv_test_(path, test);
-                mock(path, test.to_string(), "hashed", "private", "hashed", 1, "resources", None, 0.0, false);
+                mock(path, test.to_string(), "hashed", "private", "hashed", 1, "resources", None, 0.0, false, None, None);
                 test_dir.close().unwrap();
             }
 
@@ -794,7 +795,7 @@ mod native_tests {
                 let test_dir = TempDir::new(test).unwrap();
                 let path = test_dir.path().to_str().unwrap(); crate::native_tests::mv_test_(path, test);
                 // needs an extra row for the large model
-                mock(path, test.to_string(),"hashed", "hashed", "public", 1, "resources", None, 0.0, false);
+                mock(path, test.to_string(),"hashed", "hashed", "public", 1, "resources", None, 0.0, false, None, None);
                 test_dir.close().unwrap();
             }
 
@@ -804,7 +805,7 @@ mod native_tests {
                 let test_dir = TempDir::new(test).unwrap();
                 let path = test_dir.path().to_str().unwrap(); crate::native_tests::mv_test_(path, test);
                 // needs an extra row for the large model
-                mock(path, test.to_string(),"hashed", "hashed", "hashed", 1, "resources", None, 0.0, false);
+                mock(path, test.to_string(),"hashed", "hashed", "hashed", 1, "resources", None, 0.0, false, None, None);
                 test_dir.close().unwrap();
             }
 
@@ -963,7 +964,7 @@ mod native_tests {
 
             });
 
-            seq!(N in 0..=5 {
+            seq!(N in 0..=6 {
 
             #(#[test_case(LARGE_TESTS[N])])*
             #[ignore]
@@ -981,7 +982,7 @@ mod native_tests {
                 crate::native_tests::init_binary();
                 let test_dir = TempDir::new(test).unwrap();
                 let path = test_dir.path().to_str().unwrap(); crate::native_tests::mv_test_(path, test);
-                mock(path, test.to_string(), "private", "fixed", "public", 1, "resources", None, 0.0, false);
+                mock(path, test.to_string(), "private", "fixed", "public", 1, "resources", None, 0.0, false, None, None);
                 test_dir.close().unwrap();
             }
         });
@@ -1459,6 +1460,8 @@ mod native_tests {
         scales_to_use: Option<Vec<u32>>,
         tolerance: f32,
         bounded_lookup_log: bool,
+        decomp_base: Option<usize>,
+        decomp_legs: Option<usize>,
     ) {
         let mut tolerance = tolerance;
         gen_circuit_settings_and_witness(
@@ -1475,6 +1478,8 @@ mod native_tests {
             Commitments::KZG,
             2,
             bounded_lookup_log,
+            decomp_base,
+            decomp_legs,
         );
 
         if tolerance > 0.0 {
@@ -1616,6 +1621,8 @@ mod native_tests {
         commitment: Commitments,
         lookup_safety_margin: usize,
         bounded_lookup_log: bool,
+        decomp_base: Option<usize>,
+        decomp_legs: Option<usize>,
     ) {
         let mut args = vec![
             "gen-settings".to_string(),
@@ -1634,6 +1641,14 @@ mod native_tests {
             format!("--commitment={}", commitment),
         ];
 
+        if let Some(decomp_base) = decomp_base {
+            args.push(format!("--decomp-base={}", decomp_base));
+        }
+
+        if let Some(decomp_legs) = decomp_legs {
+            args.push(format!("--decomp-legs={}", decomp_legs));
+        }
+
         if bounded_lookup_log {
             args.push("--bounded-log-lookup".to_string());
         }
@@ -1751,6 +1766,8 @@ mod native_tests {
             Commitments::KZG,
             2,
             false,
+            None,
+            None,
         );
 
         println!(
@@ -2035,6 +2052,8 @@ mod native_tests {
             commitment,
             lookup_safety_margin,
             false,
+            None,
+            None,
         );
 
         let settings_path = format!("{}/{}/settings.json", test_dir, example_name);
@@ -2467,6 +2486,8 @@ mod native_tests {
             Commitments::KZG,
             2,
             false,
+            None,
+            None,
         );
 
         let model_path = format!("{}/{}/network.compiled", test_dir, example_name);
@@ -2774,7 +2795,10 @@ mod native_tests {
             "--features",
             "icicle",
         ];
-        #[cfg(not(feature = "icicle"))]
+        #[cfg(feature = "macos-metal")]
+        let args = ["build", "--release", "--bin", "ezkl", "--features", "macos-metal"];
+        // not macos-metal and not icicle
+        #[cfg(all(not(feature = "icicle"), not(feature = "macos-metal")))]
         let args = ["build", "--release", "--bin", "ezkl"];
         #[cfg(not(feature = "mv-lookup"))]
         let args = [

tests/py_integration_tests.rs

@@ -72,11 +72,10 @@ mod py_tests {
                     "torchtext==0.17.2",
                     "torchvision==0.17.2",
                     "pandas==2.2.1",
-                    "numpy==1.26.4",
                     "seaborn==0.13.2",
                     "notebook==7.1.2",
                     "nbconvert==7.16.3",
-                    "onnx==1.16.0",
+                    "onnx==1.17.0",
                     "kaggle==1.6.8",
                     "py-solc-x==2.0.3",
                     "web3==7.5.0",
@@ -90,12 +89,13 @@ mod py_tests {
                     "xgboost==2.0.3",
                     "hummingbird-ml==0.4.11",
                     "lightgbm==4.3.0",
+                    "numpy==1.26.4",
                 ])
                 .status()
                 .expect("failed to execute process");
             assert!(status.success());
             let status = Command::new("pip")
-                .args(["install", "numpy==1.23"])
+                .args(["install", "numpy==1.26.4"])
                 .status()
                 .expect("failed to execute process");
 

tests/python/binding_tests.py

@@ -873,6 +873,7 @@ def get_examples():
         'linear_regression',
         "mnist_gan",
         "smallworm",
+        "fr_age"
     ]
     examples = []
     for subdir, _, _ in os.walk(os.path.join(examples_path, "onnx")):