Add __version__ to python bindings (#739)

---------

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

.github/workflows/rust.yml

@@ -575,7 +575,7 @@ jobs:
       - name: Install Anvil
         run: cargo install --git https://github.com/foundry-rs/foundry --rev b320f350156a0fb15c2eb13dc380deb2367c4474 --profile local --locked anvil --force
       - name: Build python ezkl
-        run: source .env/bin/activate; maturin develop --features python-bindings --release
+        run: source .env/bin/activate; unset CONDA_PREFIX; maturin develop --features python-bindings --release
       - name: Run pytest
         run: source .env/bin/activate; pytest -vv
 
@@ -599,7 +599,7 @@ jobs:
       - name: Setup Virtual Env and Install python dependencies
         run: python -m venv .env; source .env/bin/activate; pip install -r requirements.txt;
       - name: Build python ezkl
-        run: source .env/bin/activate; maturin develop --features python-bindings --release
+        run: source .env/bin/activate; unset CONDA_PREFIX; maturin develop --features python-bindings --release
       - name: Div rebase
         run: source .env/bin/activate; cargo nextest run --release --verbose tests::accuracy_measurement_div_rebase_
       - name: Public inputs
@@ -634,7 +634,7 @@ jobs:
       - name: Setup Virtual Env and Install python dependencies
         run: python -m venv .env; source .env/bin/activate; pip install -r requirements.txt;
       - name: Build python ezkl
-        run: source .env/bin/activate; maturin develop --features python-bindings --release
+        run: source .env/bin/activate; unset CONDA_PREFIX; maturin develop --features python-bindings --release
       # - name: authenticate-kaggle-cli
       #   shell: bash
       #   env:

examples/notebooks/mnist_gan_proof_splitting.ipynb

@@ -343,7 +343,6 @@
                 "         compiled_model_path,\n",
                 "         vk_path,\n",
                 "         pk_path,\n",
-                "         compress_selectors=True,\n",
                 "      )\n",
                 "\n",
                 "    assert res == True\n",

examples/onnx/gather_nd/gen.py

@@ -0,0 +1,48 @@
+from torch import nn
+import json
+import numpy as np
+import tf2onnx
+
+
+import tensorflow as tf
+from tensorflow.keras.layers import *
+from tensorflow.keras.models import Model
+
+
+# gather_nd in tf then export to onnx
+
+
+
+
+x = in1 = Input((15, 18,))
+w = in2 = Input((15, 1), dtype=tf.int32)
+x = tf.gather_nd(x, w, batch_dims=1)
+tm = Model((in1, in2), x )
+tm.summary()
+tm.compile(optimizer='adam', loss='mse')
+
+shape = [1, 15, 18]
+index_shape = [1, 15, 1]
+# After training, export to onnx (network.onnx) and create a data file (input.json)
+x = 0.1*np.random.rand(1,*shape)
+# w = random int tensor
+w = np.random.randint(0, 10, index_shape)
+
+spec = tf.TensorSpec(shape, tf.float32, name='input_0')
+index_spec = tf.TensorSpec(index_shape, tf.int32, name='input_1')
+
+model_path = "network.onnx"
+
+tf2onnx.convert.from_keras(tm, input_signature=[spec, index_spec], inputs_as_nchw=['input_0', 'input_1'], opset=12, output_path=model_path)
+
+
+d = x.reshape([-1]).tolist()
+d1 = w.reshape([-1]).tolist()
+
+
+data = dict(
+    input_data=[d, d1],
+)
+
+# Serialize data into file:
+json.dump(data, open("input.json", 'w'))

examples/onnx/scatter_nd/gen.py

@@ -0,0 +1,76 @@
+import torch
+import torch.nn as nn
+import sys
+import json
+
+sys.path.append("..")
+
+class Model(nn.Module):
+    """
+    Just one Linear layer
+    """
+    def __init__(self, configs):
+        super(Model, self).__init__()
+        self.seq_len = configs.seq_len
+        self.pred_len = configs.pred_len
+
+        # Use this line if you want to visualize the weights
+        # self.Linear.weight = nn.Parameter((1/self.seq_len)*torch.ones([self.pred_len,self.seq_len]))
+        self.channels = configs.enc_in
+        self.individual = configs.individual
+        if self.individual:
+            self.Linear = nn.ModuleList()
+            for i in range(self.channels):
+                self.Linear.append(nn.Linear(self.seq_len,self.pred_len))
+        else:
+            self.Linear = nn.Linear(self.seq_len, self.pred_len)
+
+    def forward(self, x):
+        # x: [Batch, Input length, Channel]
+        if self.individual:
+            output = torch.zeros([x.size(0),self.pred_len,x.size(2)],dtype=x.dtype).to(x.device)
+            for i in range(self.channels):
+                output[:,:,i] = self.Linear[i](x[:,:,i])
+            x = output
+        else:
+            x = self.Linear(x.permute(0,2,1)).permute(0,2,1)
+        return x # [Batch, Output length, Channel]
+
+class Configs:
+    def __init__(self, seq_len, pred_len, enc_in=321, individual=True):
+      self.seq_len = seq_len
+      self.pred_len = pred_len
+      self.enc_in = enc_in
+      self.individual = individual
+
+model = 'Linear'
+seq_len = 10
+pred_len = 4
+enc_in = 3
+
+configs = Configs(seq_len, pred_len, enc_in, True)
+circuit = Model(configs)
+
+x = torch.randn(1, seq_len, pred_len)
+
+
+torch.onnx.export(circuit, x, "network.onnx",
+                  export_params=True,        # store the trained parameter weights inside the model file
+                  opset_version=15,          # the ONNX version to export the model to
+                  do_constant_folding=True,  # whether to execute constant folding for optimization
+                  # the model's input names
+                  input_names=['input'],
+                  output_names=['output'],  # the model's output names
+                  dynamic_axes={'input': {0: 'batch_size'},  # variable length axes
+                                'output': {0: 'batch_size'}})
+
+
+d1 = ((x).detach().numpy()).reshape([-1]).tolist()
+
+
+data = dict(
+    input_data=[d1],
+)
+
+# Serialize data into file:
+json.dump(data, open("input.json", 'w'))

examples/onnx/scatter_nd/input.json

@@ -0,0 +1 @@
+{"input_data": [[0.1874287724494934, 1.0498261451721191, 0.22384068369865417, 1.048445224761963, -0.5670360326766968, -0.38653188943862915, 0.12878702580928802, -2.3675858974456787, 0.5800458192825317, -0.43653929233551025, -0.2511898875236511, 0.3324051797389984, 0.27960312366485596, 0.4763695001602173, 0.3796705901622772, 1.1334782838821411, -0.87981778383255, -1.2451434135437012, 0.7672272324562073, -0.24404007196426392, -0.6875824928283691, 0.3619358539581299, -0.10131897777318954, 0.7169521450996399, 1.6585893630981445, -0.5451845526695251, 0.429487019777298, 0.7426952123641968, -0.2543637454509735, 0.06546942889690399, 0.7939824461936951, 0.1579471379518509, -0.043604474514722824, -0.8621711730957031, -0.5344759821891785, -0.05880478024482727, -0.17351101338863373, 0.5095029473304749, -0.7864817976951599, -0.449171245098114]]}

src/circuit/ops/base.rs

@@ -12,15 +12,11 @@ pub enum BaseOp {
     DotInit,
     CumProdInit,
     CumProd,
-    Identity,
     Add,
     Mult,
     Sub,
     SumInit,
     Sum,
-    Neg,
-    Range { tol: i32 },
-    IsZero,
     IsBoolean,
 }
 
@@ -36,12 +32,8 @@ impl BaseOp {
         let (a, b) = inputs;
         match &self {
             BaseOp::Add => a + b,
-            BaseOp::Identity => b,
-            BaseOp::Neg => -b,
             BaseOp::Sub => a - b,
             BaseOp::Mult => a * b,
-            BaseOp::Range { .. } => b,
-            BaseOp::IsZero => b,
             BaseOp::IsBoolean => b,
             _ => panic!("nonaccum_f called on accumulating operation"),
         }
@@ -73,19 +65,15 @@ impl BaseOp {
     /// display func
     pub fn as_str(&self) -> &'static str {
         match self {
-            BaseOp::Identity => "IDENTITY",
             BaseOp::Dot => "DOT",
             BaseOp::DotInit => "DOTINIT",
             BaseOp::CumProdInit => "CUMPRODINIT",
             BaseOp::CumProd => "CUMPROD",
             BaseOp::Add => "ADD",
-            BaseOp::Neg => "NEG",
             BaseOp::Sub => "SUB",
             BaseOp::Mult => "MULT",
             BaseOp::Sum => "SUM",
             BaseOp::SumInit => "SUMINIT",
-            BaseOp::Range { .. } => "RANGE",
-            BaseOp::IsZero => "ISZERO",
             BaseOp::IsBoolean => "ISBOOLEAN",
         }
     }
@@ -93,8 +81,6 @@ impl BaseOp {
     /// Returns the range of the query offset for this operation.
     pub fn query_offset_rng(&self) -> (i32, usize) {
         match self {
-            BaseOp::Identity => (0, 1),
-            BaseOp::Neg => (0, 1),
             BaseOp::DotInit => (0, 1),
             BaseOp::Dot => (-1, 2),
             BaseOp::CumProd => (-1, 2),
@@ -104,8 +90,6 @@ impl BaseOp {
             BaseOp::Mult => (0, 1),
             BaseOp::Sum => (-1, 2),
             BaseOp::SumInit => (0, 1),
-            BaseOp::Range { .. } => (0, 1),
-            BaseOp::IsZero => (0, 1),
             BaseOp::IsBoolean => (0, 1),
         }
     }
@@ -113,8 +97,6 @@ impl BaseOp {
     /// Returns the number of inputs for this operation.
     pub fn num_inputs(&self) -> usize {
         match self {
-            BaseOp::Identity => 1,
-            BaseOp::Neg => 1,
             BaseOp::DotInit => 2,
             BaseOp::Dot => 2,
             BaseOp::CumProdInit => 1,
@@ -124,8 +106,6 @@ impl BaseOp {
             BaseOp::Mult => 2,
             BaseOp::Sum => 1,
             BaseOp::SumInit => 1,
-            BaseOp::Range { .. } => 1,
-            BaseOp::IsZero => 0,
             BaseOp::IsBoolean => 0,
         }
     }
@@ -133,19 +113,15 @@ impl BaseOp {
     /// Returns the number of outputs for this operation.
     pub fn constraint_idx(&self) -> usize {
         match self {
-            BaseOp::Identity => 0,
-            BaseOp::Neg => 0,
             BaseOp::DotInit => 0,
             BaseOp::Dot => 1,
             BaseOp::Add => 0,
             BaseOp::Sub => 0,
             BaseOp::Mult => 0,
-            BaseOp::Range { .. } => 0,
             BaseOp::Sum => 1,
             BaseOp::SumInit => 0,
             BaseOp::CumProd => 1,
             BaseOp::CumProdInit => 0,
-            BaseOp::IsZero => 0,
             BaseOp::IsBoolean => 0,
         }
     }

src/circuit/ops/chip.rs

@@ -188,38 +188,158 @@ impl<'source> FromPyObject<'source> for Tolerance {
     }
 }
 
+/// A struct representing the selectors for the dynamic lookup tables
+#[derive(Clone, Debug, Default)]
+pub struct DynamicLookups {
+    /// [Selector]s generated when configuring the layer. We use a [BTreeMap] as we expect to configure many dynamic lookup ops.
+    pub lookup_selectors: BTreeMap<(usize, usize), Selector>,
+    /// Selectors for the dynamic lookup tables
+    pub table_selectors: Vec<Selector>,
+    /// Inputs:
+    pub inputs: Vec<VarTensor>,
+    /// tables
+    pub tables: Vec<VarTensor>,
+}
+
+impl DynamicLookups {
+    /// Returns a new [DynamicLookups] with no inputs, no selectors, and no tables.
+    pub fn dummy(col_size: usize, num_inner_cols: usize) -> Self {
+        let dummy_var = VarTensor::dummy(col_size, num_inner_cols);
+        let single_col_dummy_var = VarTensor::dummy(col_size, 1);
+
+        Self {
+            lookup_selectors: BTreeMap::new(),
+            table_selectors: vec![],
+            inputs: vec![dummy_var.clone(), dummy_var.clone(), dummy_var.clone()],
+            tables: vec![
+                single_col_dummy_var.clone(),
+                single_col_dummy_var.clone(),
+                single_col_dummy_var.clone(),
+            ],
+        }
+    }
+}
+
+/// 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), Selector>,
+    /// Selectors for the dynamic lookup tables
+    pub reference_selectors: Vec<Selector>,
+    /// Inputs:
+    pub inputs: Vec<VarTensor>,
+    /// tables
+    pub references: Vec<VarTensor>,
+}
+
+impl Shuffles {
+    /// Returns a new [DynamicLookups] with no inputs, no selectors, and no tables.
+    pub fn dummy(col_size: usize, num_inner_cols: usize) -> Self {
+        let dummy_var = VarTensor::dummy(col_size, num_inner_cols);
+        let single_col_dummy_var = VarTensor::dummy(col_size, 1);
+
+        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()],
+        }
+    }
+}
+
+/// A struct representing the selectors for the static lookup tables
+#[derive(Clone, Debug, Default)]
+pub struct StaticLookups<F: PrimeField + TensorType + PartialOrd> {
+    /// [Selector]s generated when configuring the layer. We use a [BTreeMap] as we expect to configure many dynamic lookup ops.
+    pub selectors: BTreeMap<(LookupOp, usize, usize), Selector>,
+    /// Selectors for the dynamic lookup tables
+    pub tables: BTreeMap<LookupOp, Table<F>>,
+    ///
+    pub index: VarTensor,
+    ///
+    pub output: VarTensor,
+    ///
+    pub input: VarTensor,
+}
+
+impl<F: PrimeField + TensorType + PartialOrd> StaticLookups<F> {
+    /// Returns a new [StaticLookups] with no inputs, no selectors, and no tables.
+    pub fn dummy(col_size: usize, num_inner_cols: usize) -> Self {
+        let dummy_var = VarTensor::dummy(col_size, num_inner_cols);
+
+        Self {
+            selectors: BTreeMap::new(),
+            tables: BTreeMap::new(),
+            index: dummy_var.clone(),
+            output: dummy_var.clone(),
+            input: dummy_var,
+        }
+    }
+}
+
+/// A struct representing the selectors for custom gates
+#[derive(Clone, Debug, Default)]
+pub struct CustomGates {
+    /// the inputs to the accumulated operations.
+    pub inputs: Vec<VarTensor>,
+    /// the (currently singular) output of the accumulated operations.
+    pub output: VarTensor,
+    /// selector
+    pub selectors: BTreeMap<(BaseOp, usize, usize), Selector>,
+}
+
+impl CustomGates {
+    /// Returns a new [CustomGates] with no inputs, no selectors, and no tables.
+    pub fn dummy(col_size: usize, num_inner_cols: usize) -> Self {
+        let dummy_var = VarTensor::dummy(col_size, num_inner_cols);
+        Self {
+            inputs: vec![dummy_var.clone(), dummy_var.clone()],
+            output: dummy_var,
+            selectors: BTreeMap::new(),
+        }
+    }
+}
+
+/// A struct representing the selectors for the range checks
+#[derive(Clone, Debug, Default)]
+pub struct RangeChecks<F: PrimeField + TensorType + PartialOrd> {
+    /// [Selector]s generated when configuring the layer. We use a [BTreeMap] as we expect to configure many dynamic lookup ops.
+    pub selectors: BTreeMap<(Range, usize, usize), Selector>,
+    /// Selectors for the dynamic lookup tables
+    pub ranges: BTreeMap<Range, RangeCheck<F>>,
+    ///
+    pub index: VarTensor,
+    ///
+    pub input: VarTensor,
+}
+
+impl<F: PrimeField + TensorType + PartialOrd> RangeChecks<F> {
+    /// Returns a new [RangeChecks] with no inputs, no selectors, and no tables.
+    pub fn dummy(col_size: usize, num_inner_cols: usize) -> Self {
+        let dummy_var = VarTensor::dummy(col_size, num_inner_cols);
+        Self {
+            selectors: BTreeMap::new(),
+            ranges: BTreeMap::new(),
+            index: dummy_var.clone(),
+            input: dummy_var,
+        }
+    }
+}
+
 /// Configuration for an accumulated arg.
 #[derive(Clone, Debug, Default)]
 pub struct BaseConfig<F: PrimeField + TensorType + PartialOrd> {
-    /// the inputs to the accumulated operations.
-    pub inputs: Vec<VarTensor>,
-    /// the VarTensor reserved for lookup operations (could be an element of inputs)
-    /// Note that you should be careful to ensure that the lookup_input is not simultaneously assigned to by other non-lookup operations eg. in the case of composite ops.
-    pub lookup_input: VarTensor,
-    /// the (currently singular) output of the accumulated operations.
-    pub output: VarTensor,
-    /// The VarTensor reserved for dynamic lookup operations (could be an element of inputs or the same as output)
-    /// Note that you should be careful to ensure that the lookup_output is not simultaneously assigned to by other non-lookup operations eg. in the case of composite ops.
-    pub dynamic_lookup_tables: Vec<VarTensor>,
-    /// the VarTensor reserved for lookup operations (could be an element of inputs or the same as output)
-    /// Note that you should be careful to ensure that the lookup_output is not simultaneously assigned to by other non-lookup operations eg. in the case of composite ops.
-    pub lookup_output: VarTensor,
-    ///
-    pub lookup_index: VarTensor,
-    /// [Selector]s generated when configuring the layer. We use a [BTreeMap] as we expect to configure [BaseOp].
-    pub selectors: BTreeMap<(BaseOp, usize, usize), Selector>,
-    /// [Selector]s generated when configuring the layer. We use a [BTreeMap] as we expect to configure many lookup ops.
-    pub lookup_selectors: BTreeMap<(LookupOp, usize, usize), Selector>,
-    /// [Selector]s generated when configuring the layer. We use a [BTreeMap] as we expect to configure many dynamic lookup ops.
-    pub dynamic_lookup_selectors: BTreeMap<(usize, usize), Vec<Selector>>,
-    ///
-    pub dynamic_table_selectors: Vec<Selector>,
-    ///
-    pub tables: BTreeMap<LookupOp, Table<F>>,
-    ///
-    pub range_checks: BTreeMap<Range, RangeCheck<F>>,
-    /// [Selector]s generated when configuring the layer. We use a [BTreeMap] as we expect to configure many lookup ops.
-    pub range_check_selectors: BTreeMap<(Range, usize, usize), Selector>,
+    /// Custom gates
+    pub custom_gates: CustomGates,
+    /// StaticLookups
+    pub static_lookups: StaticLookups<F>,
+    /// [Selector]s for the dynamic lookup tables
+    pub dynamic_lookups: DynamicLookups,
+    /// [Selector]s for the range checks
+    pub range_checks: RangeChecks<F>,
+    /// [Selector]s for the shuffles
+    pub shuffles: Shuffles,
     /// Activate sanity checks
     pub check_mode: CheckMode,
     _marker: PhantomData<F>,
@@ -228,22 +348,12 @@ pub struct BaseConfig<F: PrimeField + TensorType + PartialOrd> {
 impl<F: PrimeField + TensorType + PartialOrd> BaseConfig<F> {
     /// Returns a new [BaseConfig] with no inputs, no selectors, and no tables.
     pub fn dummy(col_size: usize, num_inner_cols: usize) -> Self {
-        let dummy_var = VarTensor::dummy(col_size, num_inner_cols);
-
         Self {
-            inputs: vec![dummy_var.clone(), dummy_var.clone()],
-            lookup_input: dummy_var.clone(),
-            output: dummy_var.clone(),
-            lookup_output: dummy_var.clone(),
-            dynamic_lookup_tables: vec![VarTensor::dummy(col_size, 2), dummy_var.clone()],
-            lookup_index: dummy_var,
-            selectors: BTreeMap::new(),
-            lookup_selectors: BTreeMap::new(),
-            dynamic_lookup_selectors: BTreeMap::new(),
-            dynamic_table_selectors: vec![],
-            range_check_selectors: BTreeMap::new(),
-            tables: BTreeMap::new(),
-            range_checks: BTreeMap::new(),
+            custom_gates: CustomGates::dummy(col_size, num_inner_cols),
+            static_lookups: StaticLookups::dummy(col_size, num_inner_cols),
+            dynamic_lookups: DynamicLookups::dummy(col_size, num_inner_cols),
+            shuffles: Shuffles::dummy(col_size, num_inner_cols),
+            range_checks: RangeChecks::dummy(col_size, num_inner_cols),
             check_mode: CheckMode::SAFE,
             _marker: PhantomData,
         }
@@ -276,10 +386,7 @@ impl<F: PrimeField + TensorType + PartialOrd> BaseConfig<F> {
             for j in 0..output.num_inner_cols() {
                 nonaccum_selectors.insert((BaseOp::Add, i, j), meta.selector());
                 nonaccum_selectors.insert((BaseOp::Sub, i, j), meta.selector());
-                nonaccum_selectors.insert((BaseOp::Neg, i, j), meta.selector());
                 nonaccum_selectors.insert((BaseOp::Mult, i, j), meta.selector());
-                nonaccum_selectors.insert((BaseOp::IsZero, i, j), meta.selector());
-                nonaccum_selectors.insert((BaseOp::Identity, i, j), meta.selector());
                 nonaccum_selectors.insert((BaseOp::IsBoolean, i, j), meta.selector());
             }
         }
@@ -324,12 +431,6 @@ impl<F: PrimeField + TensorType + PartialOrd> BaseConfig<F> {
 
                         vec![(output.clone()) * (output.clone() - Expression::Constant(F::from(1)))]
                     }
-                    BaseOp::IsZero => {
-                        let expected_output: Tensor<Expression<F>> = output
-                            .query_rng(meta, *block_idx, *inner_col_idx, 0, 1)
-                            .expect("non accum: output query failed");
-                        vec![expected_output[base_op.constraint_idx()].clone()]
-                    }
                     _ => {
                         let expected_output: Tensor<Expression<F>> = output
                             .query_rng(meta, *block_idx, *inner_col_idx, rotation_offset, rng)
@@ -383,19 +484,15 @@ impl<F: PrimeField + TensorType + PartialOrd> BaseConfig<F> {
             .collect();
 
         Self {
-            selectors,
-            lookup_selectors: BTreeMap::new(),
-            range_check_selectors: BTreeMap::new(),
-            dynamic_lookup_selectors: BTreeMap::new(),
-            inputs: inputs.to_vec(),
-            lookup_input: VarTensor::Empty,
-            lookup_output: VarTensor::Empty,
-            lookup_index: VarTensor::Empty,
-            dynamic_table_selectors: vec![],
-            dynamic_lookup_tables: vec![],
-            tables: BTreeMap::new(),
-            range_checks: BTreeMap::new(),
-            output: output.clone(),
+            custom_gates: CustomGates {
+                inputs: inputs.to_vec(),
+                output: output.clone(),
+                selectors,
+            },
+            static_lookups: StaticLookups::default(),
+            dynamic_lookups: DynamicLookups::default(),
+            shuffles: Shuffles::default(),
+            range_checks: RangeChecks::default(),
             check_mode,
             _marker: PhantomData,
         }
@@ -428,9 +525,9 @@ impl<F: PrimeField + TensorType + PartialOrd> BaseConfig<F> {
 
         // we borrow mutably twice so we need to do this dance
 
-        let table = if !self.tables.contains_key(nl) {
+        let table = if !self.static_lookups.tables.contains_key(nl) {
             // as all tables have the same input we see if there's another table who's input we can reuse
-            let table = if let Some(table) = self.tables.values().next() {
+            let table = if let Some(table) = self.static_lookups.tables.values().next() {
                 Table::<F>::configure(
                     cs,
                     lookup_range,
@@ -441,7 +538,7 @@ impl<F: PrimeField + TensorType + PartialOrd> BaseConfig<F> {
             } else {
                 Table::<F>::configure(cs, lookup_range, logrows, nl, None)
             };
-            self.tables.insert(nl.clone(), table.clone());
+            self.static_lookups.tables.insert(nl.clone(), table.clone());
             table
         } else {
             return Ok(());
@@ -525,22 +622,23 @@ impl<F: PrimeField + TensorType + PartialOrd> BaseConfig<F> {
                         res
                     });
                 }
-                self.lookup_selectors
+                self.static_lookups
+                    .selectors
                     .insert((nl.clone(), x, y), multi_col_selector);
             }
         }
         // if we haven't previously initialized the input/output, do so now
-        if let VarTensor::Empty = self.lookup_input {
+        if let VarTensor::Empty = self.static_lookups.input {
             debug!("assigning lookup input");
-            self.lookup_input = input.clone();
+            self.static_lookups.input = input.clone();
         }
-        if let VarTensor::Empty = self.lookup_output {
+        if let VarTensor::Empty = self.static_lookups.output {
             debug!("assigning lookup output");
-            self.lookup_output = output.clone();
+            self.static_lookups.output = output.clone();
         }
-        if let VarTensor::Empty = self.lookup_index {
+        if let VarTensor::Empty = self.static_lookups.index {
             debug!("assigning lookup index");
-            self.lookup_index = index.clone();
+            self.static_lookups.index = index.clone();
         }
         Ok(())
     }
@@ -550,8 +648,8 @@ impl<F: PrimeField + TensorType + PartialOrd> BaseConfig<F> {
     pub fn configure_dynamic_lookup(
         &mut self,
         cs: &mut ConstraintSystem<F>,
-        lookups: &[VarTensor; 2],
-        tables: &[VarTensor; 2],
+        lookups: &[VarTensor; 3],
+        tables: &[VarTensor; 3],
     ) -> Result<(), Box<dyn Error>>
     where
         F: Field,
@@ -607,18 +705,105 @@ impl<F: PrimeField + TensorType + PartialOrd> BaseConfig<F> {
 
                     expression
                 });
-                self.dynamic_lookup_selectors
+                self.dynamic_lookups
+                    .lookup_selectors
                     .entry((x, y))
-                    .or_default()
-                    .push(s_lookup);
+                    .or_insert(s_lookup);
             }
         }
-        self.dynamic_table_selectors.push(s_ltable);
+        self.dynamic_lookups.table_selectors.push(s_ltable);
 
         // if we haven't previously initialized the input/output, do so now
-        if self.dynamic_lookup_tables.is_empty() {
+        if self.dynamic_lookups.tables.is_empty() {
             debug!("assigning dynamic lookup table");
-            self.dynamic_lookup_tables = tables.to_vec();
+            self.dynamic_lookups.tables = tables.to_vec();
+        }
+        if self.dynamic_lookups.inputs.is_empty() {
+            debug!("assigning dynamic lookup input");
+            self.dynamic_lookups.inputs = lookups.to_vec();
+        }
+
+        Ok(())
+    }
+
+    /// Configures and creates lookup selectors
+    #[allow(clippy::too_many_arguments)]
+    pub fn configure_shuffles(
+        &mut self,
+        cs: &mut ConstraintSystem<F>,
+        inputs: &[VarTensor; 2],
+        references: &[VarTensor; 2],
+    ) -> Result<(), Box<dyn Error>>
+    where
+        F: Field,
+    {
+        for l in inputs.iter() {
+            if !l.is_advice() {
+                return Err("wrong input type for dynamic lookup".into());
+            }
+        }
+
+        for t in references.iter() {
+            if !t.is_advice() || t.num_blocks() > 1 || t.num_inner_cols() > 1 {
+                return Err("wrong table type for dynamic lookup".into());
+            }
+        }
+
+        let one = Expression::Constant(F::ONE);
+
+        let s_reference = 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| {
+                    let s_inputq = cs.query_selector(s_input);
+                    let mut expression = vec![];
+                    let s_referenceq = cs.query_selector(s_reference);
+                    let mut input_queries = vec![one.clone()];
+
+                    for input in inputs {
+                        input_queries.push(match input {
+                            VarTensor::Advice { inner: advices, .. } => {
+                                cs.query_advice(advices[x][y], Rotation(0))
+                            }
+                            _ => unreachable!(),
+                        });
+                    }
+
+                    let mut ref_queries = vec![one.clone()];
+                    for reference in references {
+                        ref_queries.push(match reference {
+                            VarTensor::Advice { inner: advices, .. } => {
+                                cs.query_advice(advices[0][0], Rotation(0))
+                            }
+                            _ => unreachable!(),
+                        });
+                    }
+
+                    let lhs = input_queries.into_iter().map(|c| c * s_inputq.clone());
+                    let rhs = ref_queries.into_iter().map(|c| c * s_referenceq.clone());
+                    expression.extend(lhs.zip(rhs));
+
+                    expression
+                });
+                self.shuffles
+                    .input_selectors
+                    .entry((x, y))
+                    .or_insert(s_input);
+            }
+        }
+        self.shuffles.reference_selectors.push(s_reference);
+
+        // 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.inputs.is_empty() {
+            debug!("assigning shuffles input");
+            self.shuffles.inputs = inputs.to_vec();
         }
 
         Ok(())
@@ -643,15 +828,16 @@ impl<F: PrimeField + TensorType + PartialOrd> BaseConfig<F> {
 
         // we borrow mutably twice so we need to do this dance
 
-        let range_check =
-            if let std::collections::btree_map::Entry::Vacant(e) = self.range_checks.entry(range) {
-                // as all tables have the same input we see if there's another table who's input we can reuse
-                let range_check = RangeCheck::<F>::configure(cs, range, logrows);
-                e.insert(range_check.clone());
-                range_check
-            } else {
-                return Ok(());
-            };
+        let range_check = if let std::collections::btree_map::Entry::Vacant(e) =
+            self.range_checks.ranges.entry(range)
+        {
+            // as all tables have the same input we see if there's another table who's input we can reuse
+            let range_check = RangeCheck::<F>::configure(cs, range, logrows);
+            e.insert(range_check.clone());
+            range_check
+        } else {
+            return Ok(());
+        };
 
         for x in 0..input.num_blocks() {
             for y in 0..input.num_inner_cols() {
@@ -708,19 +894,20 @@ impl<F: PrimeField + TensorType + PartialOrd> BaseConfig<F> {
                         res
                     });
                 }
-                self.range_check_selectors
+                self.range_checks
+                    .selectors
                     .insert((range, x, y), multi_col_selector);
             }
         }
         // if we haven't previously initialized the input/output, do so now
-        if let VarTensor::Empty = self.lookup_input {
-            debug!("assigning lookup input");
-            self.lookup_input = input.clone();
+        if let VarTensor::Empty = self.range_checks.input {
+            debug!("assigning range check input");
+            self.range_checks.input = input.clone();
         }
 
-        if let VarTensor::Empty = self.lookup_index {
-            debug!("assigning lookup index");
-            self.lookup_index = index.clone();
+        if let VarTensor::Empty = self.range_checks.index {
+            debug!("assigning range check index");
+            self.range_checks.index = index.clone();
         }
 
         Ok(())
@@ -728,7 +915,7 @@ impl<F: PrimeField + TensorType + PartialOrd> BaseConfig<F> {
 
     /// layout_tables must be called before layout.
     pub fn layout_tables(&mut self, layouter: &mut impl Layouter<F>) -> Result<(), Box<dyn Error>> {
-        for (i, table) in self.tables.values_mut().enumerate() {
+        for (i, table) in self.static_lookups.tables.values_mut().enumerate() {
             if !table.is_assigned {
                 debug!(
                     "laying out table for {}",
@@ -749,7 +936,7 @@ impl<F: PrimeField + TensorType + PartialOrd> BaseConfig<F> {
         &mut self,
         layouter: &mut impl Layouter<F>,
     ) -> Result<(), Box<dyn Error>> {
-        for range_check in self.range_checks.values_mut() {
+        for range_check in self.range_checks.ranges.values_mut() {
             if !range_check.is_assigned {
                 debug!("laying out range check for {:?}", range_check.range);
                 range_check.layout(layouter)?;

src/circuit/ops/poly.rs

@@ -14,10 +14,17 @@ pub enum PolyOp {
         dim: usize,
         constant_idx: Option<Tensor<usize>>,
     },
+    GatherND {
+        batch_dims: usize,
+        indices: Option<Tensor<usize>>,
+    },
     ScatterElements {
         dim: usize,
         constant_idx: Option<Tensor<usize>>,
     },
+    ScatterND {
+        constant_idx: Option<Tensor<usize>>,
+    },
     MultiBroadcastTo {
         shape: Vec<usize>,
     },
@@ -60,8 +67,6 @@ pub enum PolyOp {
         len_prod: usize,
     },
     Pow(u32),
-    Pack(u32, u32),
-    GlobalSumPool,
     Concat {
         axis: usize,
     },
@@ -91,7 +96,9 @@ impl<F: PrimeField + TensorType + PartialOrd + Serialize + for<'de> Deserialize<
     fn as_string(&self) -> String {
         match &self {
             PolyOp::GatherElements { dim, .. } => format!("GATHERELEMENTS (dim={})", dim),
+            PolyOp::GatherND { batch_dims, .. } => format!("GATHERND (batch_dims={})", batch_dims),
             PolyOp::ScatterElements { dim, .. } => format!("SCATTERELEMENTS (dim={})", dim),
+            PolyOp::ScatterND { .. } => "SCATTERND".into(),
             PolyOp::MultiBroadcastTo { shape } => format!("MULTIBROADCASTTO (shape={:?})", shape),
             PolyOp::MoveAxis { .. } => "MOVEAXIS".into(),
             PolyOp::Downsample { .. } => "DOWNSAMPLE".into(),
@@ -110,8 +117,6 @@ impl<F: PrimeField + TensorType + PartialOrd + Serialize + for<'de> Deserialize<
             PolyOp::Sum { .. } => "SUM".into(),
             PolyOp::Prod { .. } => "PROD".into(),
             PolyOp::Pow(_) => "POW".into(),
-            PolyOp::Pack(_, _) => "PACK".into(),
-            PolyOp::GlobalSumPool => "GLOBALSUMPOOL".into(),
             PolyOp::Conv { .. } => "CONV".into(),
             PolyOp::DeConv { .. } => "DECONV".into(),
             PolyOp::Concat { axis } => format!("CONCAT (axis={})", axis),
@@ -181,13 +186,6 @@ impl<F: PrimeField + TensorType + PartialOrd + Serialize + for<'de> Deserialize<
                 output_padding,
                 stride,
             } => tensor::ops::deconv(&inputs, *padding, *output_padding, *stride),
-            PolyOp::Pack(base, scale) => {
-                if 1 != inputs.len() {
-                    return Err(TensorError::DimMismatch("pack inputs".to_string()));
-                }
-
-                tensor::ops::pack(&inputs[0], F::from(*base as u64), *scale)
-            }
             PolyOp::Pow(u) => {
                 if 1 != inputs.len() {
                     return Err(TensorError::DimMismatch("pow inputs".to_string()));
@@ -206,7 +204,6 @@ impl<F: PrimeField + TensorType + PartialOrd + Serialize + for<'de> Deserialize<
                 }
                 tensor::ops::prod_axes(&inputs[0], axes)
             }
-            PolyOp::GlobalSumPool => unreachable!(),
             PolyOp::Concat { axis } => {
                 tensor::ops::concat(&inputs.iter().collect::<Vec<_>>(), *axis)
             }
@@ -225,6 +222,18 @@ impl<F: PrimeField + TensorType + PartialOrd + Serialize + for<'de> Deserialize<
                 };
                 tensor::ops::gather_elements(&x, &y, *dim)
             }
+            PolyOp::GatherND {
+                indices,
+                batch_dims,
+            } => {
+                let x = inputs[0].clone();
+                let y = if let Some(idx) = indices {
+                    idx.clone()
+                } else {
+                    inputs[1].clone().map(|x| felt_to_i128(x) as usize)
+                };
+                tensor::ops::gather_nd(&x, &y, *batch_dims)
+            }
             PolyOp::ScatterElements { dim, constant_idx } => {
                 let x = inputs[0].clone();
 
@@ -241,6 +250,21 @@ impl<F: PrimeField + TensorType + PartialOrd + Serialize + for<'de> Deserialize<
                 };
                 tensor::ops::scatter(&x, &idx, &src, *dim)
             }
+
+            PolyOp::ScatterND { constant_idx } => {
+                let x = inputs[0].clone();
+                let idx = if let Some(idx) = constant_idx {
+                    idx.clone()
+                } else {
+                    inputs[1].clone().map(|x| felt_to_i128(x) as usize)
+                };
+                let src = if constant_idx.is_some() {
+                    inputs[1].clone()
+                } else {
+                    inputs[2].clone()
+                };
+                tensor::ops::scatter_nd(&x, &idx, &src)
+            }
         }?;
 
         Ok(ForwardResult { output: res })
@@ -288,7 +312,17 @@ impl<F: PrimeField + TensorType + PartialOrd + Serialize + for<'de> Deserialize<
                 if let Some(idx) = constant_idx {
                     tensor::ops::gather_elements(values[0].get_inner_tensor()?, idx, *dim)?.into()
                 } else {
-                    layouts::gather_elements(config, region, values[..].try_into()?, *dim)?
+                    layouts::gather_elements(config, region, values[..].try_into()?, *dim)?.0
+                }
+            }
+            PolyOp::GatherND {
+                batch_dims,
+                indices,
+            } => {
+                if let Some(idx) = indices {
+                    tensor::ops::gather_nd(values[0].get_inner_tensor()?, idx, *batch_dims)?.into()
+                } else {
+                    layouts::gather_nd(config, region, values[..].try_into()?, *batch_dims)?.0
                 }
             }
             PolyOp::ScatterElements { dim, constant_idx } => {
@@ -304,6 +338,18 @@ impl<F: PrimeField + TensorType + PartialOrd + Serialize + for<'de> Deserialize<
                     layouts::scatter_elements(config, region, values[..].try_into()?, *dim)?
                 }
             }
+            PolyOp::ScatterND { constant_idx } => {
+                if let Some(idx) = constant_idx {
+                    tensor::ops::scatter_nd(
+                        values[0].get_inner_tensor()?,
+                        idx,
+                        values[1].get_inner_tensor()?,
+                    )?
+                    .into()
+                } else {
+                    layouts::scatter_nd(config, region, values[..].try_into()?)?
+                }
+            }
             PolyOp::DeConv {
                 padding,
                 output_padding,
@@ -334,10 +380,6 @@ impl<F: PrimeField + TensorType + PartialOrd + Serialize + for<'de> Deserialize<
                 input
             }
             PolyOp::Pow(exp) => layouts::pow(config, region, values[..].try_into()?, *exp)?,
-            PolyOp::Pack(base, scale) => {
-                layouts::pack(config, region, values[..].try_into()?, *base, *scale)?
-            }
-            PolyOp::GlobalSumPool => unreachable!(),
             PolyOp::Concat { axis } => layouts::concat(values[..].try_into()?, axis)?,
             PolyOp::Slice { axis, start, end } => {
                 layouts::slice(config, region, values[..].try_into()?, axis, start, end)?
@@ -405,7 +447,9 @@ impl<F: PrimeField + TensorType + PartialOrd + Serialize + for<'de> Deserialize<
             vec![1, 2]
         } else if matches!(self, PolyOp::Concat { .. }) {
             (0..100).collect()
-        } else if matches!(self, PolyOp::ScatterElements { .. }) {
+        } else if matches!(self, PolyOp::ScatterElements { .. })
+            | matches!(self, PolyOp::ScatterND { .. })
+        {
             vec![0, 2]
         } else {
             vec![]

src/circuit/ops/region.rs

@@ -23,28 +23,61 @@ use super::lookup::LookupOp;
 /// Dynamic lookup index
 #[derive(Clone, Debug, Default)]
 pub struct DynamicLookupIndex {
-    lookup_index: usize,
+    index: usize,
     col_coord: usize,
 }
 
 impl DynamicLookupIndex {
     /// Create a new dynamic lookup index
-    pub fn new(lookup_index: usize, col_coord: usize) -> DynamicLookupIndex {
-        DynamicLookupIndex {
-            lookup_index,
-            col_coord,
-        }
+    pub fn new(index: usize, col_coord: usize) -> DynamicLookupIndex {
+        DynamicLookupIndex { index, col_coord }
     }
 
     /// Get the lookup index
-    pub fn lookup_index(&self) -> usize {
-        self.lookup_index
+    pub fn index(&self) -> usize {
+        self.index
     }
 
     /// Get the column coord
     pub fn col_coord(&self) -> usize {
         self.col_coord
     }
+
+    /// update with another dynamic lookup index
+    pub fn update(&mut self, other: &DynamicLookupIndex) {
+        self.index += other.index;
+        self.col_coord += other.col_coord;
+    }
+}
+
+/// Dynamic lookup index
+#[derive(Clone, Debug, Default)]
+pub struct ShuffleIndex {
+    index: usize,
+    col_coord: usize,
+}
+
+impl ShuffleIndex {
+    /// Create a new dynamic lookup index
+    pub fn new(index: usize, col_coord: usize) -> ShuffleIndex {
+        ShuffleIndex { index, col_coord }
+    }
+
+    /// Get the lookup index
+    pub fn index(&self) -> usize {
+        self.index
+    }
+
+    /// Get the column coord
+    pub fn col_coord(&self) -> usize {
+        self.col_coord
+    }
+
+    /// update with another shuffle index
+    pub fn update(&mut self, other: &ShuffleIndex) {
+        self.index += other.index;
+        self.col_coord += other.col_coord;
+    }
 }
 
 /// Region error
@@ -94,6 +127,7 @@ pub struct RegionCtx<'a, F: PrimeField + TensorType + PartialOrd> {
     num_inner_cols: usize,
     total_constants: usize,
     dynamic_lookup_index: DynamicLookupIndex,
+    shuffle_index: ShuffleIndex,
     used_lookups: HashSet<LookupOp>,
     used_range_checks: HashSet<Range>,
     max_lookup_inputs: i128,
@@ -110,7 +144,7 @@ impl<'a, F: PrimeField + TensorType + PartialOrd> RegionCtx<'a, F> {
 
     ///
     pub fn increment_dynamic_lookup_index(&mut self, n: usize) {
-        self.dynamic_lookup_index.lookup_index += n;
+        self.dynamic_lookup_index.index += n;
     }
 
     ///
@@ -118,6 +152,16 @@ impl<'a, F: PrimeField + TensorType + PartialOrd> RegionCtx<'a, F> {
         self.dynamic_lookup_index.col_coord += n;
     }
 
+    ///
+    pub fn increment_shuffle_index(&mut self, n: usize) {
+        self.shuffle_index.index += n;
+    }
+
+    ///
+    pub fn increment_shuffle_col_coord(&mut self, n: usize) {
+        self.shuffle_index.col_coord += n;
+    }
+
     ///
     pub fn throw_range_check_error(&self) -> bool {
         self.throw_range_check_error
@@ -135,6 +179,7 @@ impl<'a, F: PrimeField + TensorType + PartialOrd> RegionCtx<'a, F> {
             linear_coord,
             total_constants: 0,
             dynamic_lookup_index: DynamicLookupIndex::default(),
+            shuffle_index: ShuffleIndex::default(),
             used_lookups: HashSet::new(),
             used_range_checks: HashSet::new(),
             max_lookup_inputs: 0,
@@ -149,6 +194,7 @@ impl<'a, F: PrimeField + TensorType + PartialOrd> RegionCtx<'a, F> {
         row: usize,
         num_inner_cols: usize,
         dynamic_lookup_index: DynamicLookupIndex,
+        shuffle_index: ShuffleIndex,
     ) -> RegionCtx<'a, F> {
         let linear_coord = row * num_inner_cols;
         RegionCtx {
@@ -158,6 +204,7 @@ impl<'a, F: PrimeField + TensorType + PartialOrd> RegionCtx<'a, F> {
             row,
             total_constants: 0,
             dynamic_lookup_index,
+            shuffle_index,
             used_lookups: HashSet::new(),
             used_range_checks: HashSet::new(),
             max_lookup_inputs: 0,
@@ -183,6 +230,7 @@ impl<'a, F: PrimeField + TensorType + PartialOrd> RegionCtx<'a, F> {
             row,
             total_constants: 0,
             dynamic_lookup_index: DynamicLookupIndex::default(),
+            shuffle_index: ShuffleIndex::default(),
             used_lookups: HashSet::new(),
             used_range_checks: HashSet::new(),
             max_lookup_inputs: 0,
@@ -198,9 +246,6 @@ impl<'a, F: PrimeField + TensorType + PartialOrd> RegionCtx<'a, F> {
         linear_coord: usize,
         total_constants: usize,
         num_inner_cols: usize,
-        dynamic_lookup_index: DynamicLookupIndex,
-        used_lookups: HashSet<LookupOp>,
-        used_range_checks: HashSet<Range>,
         throw_range_check_error: bool,
     ) -> RegionCtx<'a, F> {
         let region = None;
@@ -210,9 +255,10 @@ impl<'a, F: PrimeField + TensorType + PartialOrd> RegionCtx<'a, F> {
             linear_coord,
             row,
             total_constants,
-            dynamic_lookup_index,
-            used_lookups,
-            used_range_checks,
+            dynamic_lookup_index: DynamicLookupIndex::default(),
+            shuffle_index: ShuffleIndex::default(),
+            used_lookups: HashSet::new(),
+            used_range_checks: HashSet::new(),
             max_lookup_inputs: 0,
             min_lookup_inputs: 0,
             max_range_size: 0,
@@ -272,6 +318,7 @@ impl<'a, F: PrimeField + TensorType + PartialOrd> RegionCtx<'a, F> {
         let lookups = Arc::new(Mutex::new(self.used_lookups.clone()));
         let range_checks = Arc::new(Mutex::new(self.used_range_checks.clone()));
         let dynamic_lookup_index = Arc::new(Mutex::new(self.dynamic_lookup_index.clone()));
+        let shuffle_index = Arc::new(Mutex::new(self.shuffle_index.clone()));
 
         *output = output
             .par_enum_map(|idx, _| {
@@ -287,9 +334,6 @@ impl<'a, F: PrimeField + TensorType + PartialOrd> RegionCtx<'a, F> {
                     starting_linear_coord,
                     starting_constants,
                     self.num_inner_cols,
-                    DynamicLookupIndex::default(),
-                    HashSet::new(),
-                    HashSet::new(),
                     self.throw_range_check_error,
                 );
                 let res = inner_loop_function(idx, &mut local_reg);
@@ -309,18 +353,19 @@ impl<'a, F: PrimeField + TensorType + PartialOrd> RegionCtx<'a, F> {
                 // update the lookups
                 let mut lookups = lookups.lock().unwrap();
                 lookups.extend(local_reg.used_lookups());
+                // update the range checks
                 let mut range_checks = range_checks.lock().unwrap();
                 range_checks.extend(local_reg.used_range_checks());
+                // update the dynamic lookup index
                 let mut dynamic_lookup_index = dynamic_lookup_index.lock().unwrap();
-                dynamic_lookup_index.lookup_index += local_reg.dynamic_lookup_index.lookup_index;
-                dynamic_lookup_index.col_coord += local_reg.dynamic_lookup_index.col_coord;
+                dynamic_lookup_index.update(&local_reg.dynamic_lookup_index);
+                // update the shuffle index
+                let mut shuffle_index = shuffle_index.lock().unwrap();
+                shuffle_index.update(&local_reg.shuffle_index);
 
                 res
             })
-            .map_err(|e| {
-                log::error!("dummy_loop: {:?}", e);
-                Error::Synthesis
-            })?;
+            .map_err(|e| RegionError::from(format!("dummy_loop: {:?}", e)))?;
         self.total_constants = constants.into_inner();
         self.linear_coord = linear_coord.into_inner();
         #[allow(trivial_numeric_casts)]
@@ -357,6 +402,14 @@ impl<'a, F: PrimeField + TensorType + PartialOrd> RegionCtx<'a, F> {
                     e
                 ))
             })?;
+        self.shuffle_index = Arc::try_unwrap(shuffle_index)
+            .map_err(|e| {
+                RegionError::from(format!("dummy_loop: failed to get shuffle index: {:?}", e))
+            })?
+            .into_inner()
+            .map_err(|e| {
+                RegionError::from(format!("dummy_loop: failed to get shuffle index: {:?}", e))
+            })?;
 
         Ok(())
     }
@@ -429,7 +482,7 @@ impl<'a, F: PrimeField + TensorType + PartialOrd> RegionCtx<'a, F> {
 
     /// Get the dynamic lookup index
     pub fn dynamic_lookup_index(&self) -> usize {
-        self.dynamic_lookup_index.lookup_index
+        self.dynamic_lookup_index.index
     }
 
     /// Get the dynamic lookup column coordinate
@@ -437,6 +490,16 @@ impl<'a, F: PrimeField + TensorType + PartialOrd> RegionCtx<'a, F> {
         self.dynamic_lookup_index.col_coord
     }
 
+    /// Get the shuffle index
+    pub fn shuffle_index(&self) -> usize {
+        self.shuffle_index.index
+    }
+
+    /// Get the shuffle column coordinate
+    pub fn shuffle_col_coord(&self) -> usize {
+        self.shuffle_index.col_coord
+    }
+
     /// get used lookups
     pub fn used_lookups(&self) -> HashSet<LookupOp> {
         self.used_lookups.clone()
@@ -486,6 +549,11 @@ impl<'a, F: PrimeField + TensorType + PartialOrd> RegionCtx<'a, F> {
         }
     }
 
+    ///
+    pub fn combined_dynamic_shuffle_coord(&self) -> usize {
+        self.dynamic_lookup_col_coord() + self.shuffle_col_coord()
+    }
+
     /// Assign a valtensor to a vartensor
     pub fn assign_dynamic_lookup(
         &mut self,
@@ -496,7 +564,7 @@ impl<'a, F: PrimeField + TensorType + PartialOrd> RegionCtx<'a, F> {
         if let Some(region) = &self.region {
             var.assign(
                 &mut region.borrow_mut(),
-                self.dynamic_lookup_col_coord(),
+                self.combined_dynamic_shuffle_coord(),
                 values,
             )
         } else {
@@ -504,6 +572,15 @@ impl<'a, F: PrimeField + TensorType + PartialOrd> RegionCtx<'a, F> {
         }
     }
 
+    /// Assign a valtensor to a vartensor
+    pub fn assign_shuffle(
+        &mut self,
+        var: &VarTensor,
+        values: &ValTensor<F>,
+    ) -> Result<ValTensor<F>, Error> {
+        self.assign_dynamic_lookup(var, values)
+    }
+
     /// Assign a valtensor to a vartensor
     pub fn assign_with_omissions(
         &mut self,

src/circuit/tests.rs

@@ -357,8 +357,6 @@ mod matmul_col_ultra_overflow_double_col {
         );
 
         assert!(result.is_ok());
-
-        println!("done.");
     }
 }
 
@@ -476,8 +474,6 @@ mod matmul_col_ultra_overflow {
         );
 
         assert!(result.is_ok());
-
-        println!("done.");
     }
 }
 
@@ -1280,8 +1276,6 @@ mod conv_col_ultra_overflow {
         );
 
         assert!(result.is_ok());
-
-        println!("done.");
     }
 }
 
@@ -1435,8 +1429,6 @@ mod conv_relu_col_ultra_overflow {
         );
 
         assert!(result.is_ok());
-
-        println!("done.");
     }
 }
 
@@ -1605,15 +1597,19 @@ mod dynamic_lookup {
 
             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);
-            for _ in 0..NUM_LOOP {
-                config
-                    .configure_dynamic_lookup(cs, &[a.clone(), b.clone()], &[d.clone(), e.clone()])
-                    .unwrap();
-            }
-
+            config
+                .configure_dynamic_lookup(
+                    cs,
+                    &[a.clone(), b.clone(), c.clone()],
+                    &[d.clone(), e.clone(), f.clone()],
+                )
+                .unwrap();
             config
         }
 
@@ -1690,11 +1686,18 @@ mod dynamic_lookup {
 
         let lookups = (0..NUM_LOOP)
             .map(|loop_idx| {
+                let prev_idx = if loop_idx == 0 {
+                    NUM_LOOP - 1
+                } else {
+                    loop_idx - 1
+                };
                 [
                     ValTensor::from(Tensor::from(
-                        (0..2).map(|i| Value::known(F::from((i * loop_idx) as u64 + 1))),
+                        (0..3).map(|i| Value::known(F::from((i * prev_idx) as u64 + 1))),
+                    )),
+                    ValTensor::from(Tensor::from(
+                        (0..3).map(|i| Value::known(F::from((prev_idx * i * i) as u64 + 1))),
                     )),
-                    ValTensor::from(Tensor::from((0..2).map(|_| Value::known(F::from(10000))))),
                 ]
             })
             .collect::<Vec<_>>();
@@ -1710,6 +1713,133 @@ mod dynamic_lookup {
     }
 }
 
+#[cfg(test)]
+mod shuffle {
+    use super::*;
+
+    const K: usize = 6;
+    const LEN: usize = 4;
+    const NUM_LOOP: usize = 5;
+
+    #[derive(Clone)]
+    struct MyCircuit<F: PrimeField + TensorType + PartialOrd> {
+        inputs: [[ValTensor<F>; 1]; NUM_LOOP],
+        references: [[ValTensor<F>; 1]; NUM_LOOP],
+        _marker: PhantomData<F>,
+    }
+
+    impl Circuit<F> for MyCircuit<F> {
+        type Config = BaseConfig<F>;
+        type FloorPlanner = SimpleFloorPlanner;
+        type Params = TestParams;
+
+        fn without_witnesses(&self) -> Self {
+            self.clone()
+        }
+
+        fn configure(cs: &mut ConstraintSystem<F>) -> Self::Config {
+            let a = VarTensor::new_advice(cs, K, 2, LEN);
+            let b = VarTensor::new_advice(cs, K, 2, LEN);
+            let c: VarTensor = VarTensor::new_advice(cs, K, 2, LEN);
+
+            let d = VarTensor::new_advice(cs, K, 1, LEN);
+            let e = 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()])
+                .unwrap();
+            config
+        }
+
+        fn synthesize(
+            &self,
+            config: Self::Config,
+            mut layouter: impl Layouter<F>,
+        ) -> Result<(), Error> {
+            layouter
+                .assign_region(
+                    || "",
+                    |region| {
+                        let mut region = RegionCtx::new(region, 0, 1);
+                        for i in 0..NUM_LOOP {
+                            layouts::shuffles(
+                                &config,
+                                &mut region,
+                                &self.inputs[i],
+                                &self.references[i],
+                            )
+                            .map_err(|_| Error::Synthesis)?;
+                        }
+                        assert_eq!(
+                            region.shuffle_col_coord(),
+                            NUM_LOOP * self.references[0][0].len()
+                        );
+                        assert_eq!(region.shuffle_index(), NUM_LOOP);
+
+                        Ok(())
+                    },
+                )
+                .unwrap();
+            Ok(())
+        }
+    }
+
+    #[test]
+    fn shufflecircuit() {
+        // parameters
+        let references = (0..NUM_LOOP)
+            .map(|loop_idx| {
+                [ValTensor::from(Tensor::from((0..LEN).map(|i| {
+                    Value::known(F::from((i * loop_idx) as u64 + 1))
+                })))]
+            })
+            .collect::<Vec<_>>();
+
+        let inputs = (0..NUM_LOOP)
+            .map(|loop_idx| {
+                [ValTensor::from(Tensor::from((0..LEN).rev().map(|i| {
+                    Value::known(F::from((i * loop_idx) as u64 + 1))
+                })))]
+            })
+            .collect::<Vec<_>>();
+
+        let circuit = MyCircuit::<F> {
+            references: references.clone().try_into().unwrap(),
+            inputs: inputs.try_into().unwrap(),
+            _marker: PhantomData,
+        };
+
+        let prover = MockProver::run(K as u32, &circuit, vec![]).unwrap();
+        prover.assert_satisfied();
+
+        let inputs = (0..NUM_LOOP)
+            .map(|loop_idx| {
+                let prev_idx = if loop_idx == 0 {
+                    NUM_LOOP - 1
+                } else {
+                    loop_idx - 1
+                };
+                [ValTensor::from(Tensor::from((0..LEN).rev().map(|i| {
+                    Value::known(F::from((i * prev_idx) as u64 + 1))
+                })))]
+            })
+            .collect::<Vec<_>>();
+
+        let circuit = MyCircuit::<F> {
+            references: references.try_into().unwrap(),
+            inputs: inputs.try_into().unwrap(),
+            _marker: PhantomData,
+        };
+
+        let prover = MockProver::run(K as u32, &circuit, vec![]).unwrap();
+        assert!(prover.verify().is_err());
+    }
+}
+
 #[cfg(test)]
 mod add_with_overflow {
     use super::*;
@@ -2113,75 +2243,6 @@ mod pow {
     }
 }
 
-#[cfg(test)]
-mod pack {
-    use super::*;
-
-    const K: usize = 8;
-    const LEN: usize = 4;
-
-    #[derive(Clone)]
-    struct MyCircuit<F: PrimeField + TensorType + PartialOrd> {
-        inputs: [ValTensor<F>; 1],
-        _marker: PhantomData<F>,
-    }
-
-    impl Circuit<F> for MyCircuit<F> {
-        type Config = BaseConfig<F>;
-        type FloorPlanner = SimpleFloorPlanner;
-        type Params = TestParams;
-
-        fn without_witnesses(&self) -> Self {
-            self.clone()
-        }
-
-        fn configure(cs: &mut ConstraintSystem<F>) -> Self::Config {
-            let a = VarTensor::new_advice(cs, K, 1, LEN);
-            let b = VarTensor::new_advice(cs, K, 1, LEN);
-            let output = VarTensor::new_advice(cs, K, 1, LEN);
-
-            Self::Config::configure(cs, &[a, b], &output, CheckMode::SAFE)
-        }
-
-        fn synthesize(
-            &self,
-            mut config: Self::Config,
-            mut layouter: impl Layouter<F>,
-        ) -> Result<(), Error> {
-            layouter
-                .assign_region(
-                    || "",
-                    |region| {
-                        let mut region = RegionCtx::new(region, 0, 1);
-                        config
-                            .layout(
-                                &mut region,
-                                &self.inputs.clone(),
-                                Box::new(PolyOp::Pack(2, 1)),
-                            )
-                            .map_err(|_| Error::Synthesis)
-                    },
-                )
-                .unwrap();
-            Ok(())
-        }
-    }
-
-    #[test]
-    fn packcircuit() {
-        // parameters
-        let a = Tensor::from((0..LEN).map(|i| Value::known(F::from(i as u64 + 1))));
-
-        let circuit = MyCircuit::<F> {
-            inputs: [ValTensor::from(a)],
-            _marker: PhantomData,
-        };
-
-        let prover = MockProver::run(K as u32, &circuit, vec![]).unwrap();
-        prover.assert_satisfied();
-    }
-}
-
 #[cfg(test)]
 mod matmul_relu {
     use super::*;
@@ -2469,7 +2530,5 @@ mod lookup_ultra_overflow {
         );
 
         assert!(result.is_ok());
-
-        println!("done.");
     }
 }

src/commands.rs

@@ -77,7 +77,7 @@ pub const DEFAULT_CALIBRATION_FILE: &str = "calibration.json";
 /// Default lookup safety margin
 pub const DEFAULT_LOOKUP_SAFETY_MARGIN: &str = "2";
 /// Default Compress selectors
-pub const DEFAULT_COMPRESS_SELECTORS: &str = "false";
+pub const DEFAULT_DISABLE_SELECTOR_COMPRESSION: &str = "false";
 /// Default render vk seperately
 pub const DEFAULT_RENDER_VK_SEPERATELY: &str = "false";
 /// Default VK sol path
@@ -402,9 +402,6 @@ pub enum Commands {
         /// Number of logrows to use for srs. Overrides settings_path if specified.
         #[arg(long, default_value = None)]
         logrows: Option<u32>,
-        /// Check mode for SRS. Verifies downloaded srs is valid. Set to unsafe for speed.
-        #[arg(long, default_value = DEFAULT_CHECKMODE)]
-        check: CheckMode,
     },
     /// Loads model and input and runs mock prover (for testing)
     Mock {
@@ -450,8 +447,8 @@ pub enum Commands {
         #[arg(long, default_value = DEFAULT_SPLIT)]
         split_proofs: bool,
         /// compress selectors
-        #[arg(long, default_value = DEFAULT_COMPRESS_SELECTORS)]
-        compress_selectors: bool,
+        #[arg(long, default_value = DEFAULT_DISABLE_SELECTOR_COMPRESSION)]
+        disable_selector_compression: bool,
     },
     /// Aggregates proofs :)
     Aggregate {
@@ -515,8 +512,8 @@ pub enum Commands {
         #[arg(short = 'W', long)]
         witness: Option<PathBuf>,
         /// compress selectors
-        #[arg(long, default_value = DEFAULT_COMPRESS_SELECTORS)]
-        compress_selectors: bool,
+        #[arg(long, default_value = DEFAULT_DISABLE_SELECTOR_COMPRESSION)]
+        disable_selector_compression: bool,
     },
 
     #[cfg(not(target_arch = "wasm32"))]
@@ -540,8 +537,8 @@ pub enum Commands {
         #[arg(long, default_value = DEFAULT_FUZZ_RUNS)]
         num_runs: usize,
         /// compress selectors
-        #[arg(long, default_value = DEFAULT_COMPRESS_SELECTORS)]
-        compress_selectors: bool,
+        #[arg(long, default_value = DEFAULT_DISABLE_SELECTOR_COMPRESSION)]
+        disable_selector_compression: bool,
     },
     #[cfg(not(target_arch = "wasm32"))]
     /// Deploys a test contact that the data attester reads from and creates a data attestation formatted input.json file that contains call data information

src/execute.rs

@@ -23,6 +23,7 @@ use crate::pfsys::{create_proof_circuit_kzg, verify_proof_circuit_kzg};
 use crate::pfsys::{save_vk, srs::*};
 use crate::tensor::TensorError;
 use crate::RunArgs;
+#[cfg(unix)]
 use gag::Gag;
 use halo2_proofs::dev::VerifyFailure;
 use halo2_proofs::poly::commitment::Params;
@@ -63,7 +64,11 @@ use std::process::Command;
 use std::sync::atomic::{AtomicBool, AtomicI64, Ordering};
 #[cfg(not(target_arch = "wasm32"))]
 use std::sync::OnceLock;
+
 #[cfg(not(target_arch = "wasm32"))]
+use crate::EZKL_BUF_CAPACITY;
+#[cfg(not(target_arch = "wasm32"))]
+use std::io::BufWriter;
 use std::time::Duration;
 use tabled::Tabled;
 use thiserror::Error;
@@ -140,13 +145,13 @@ pub async fn run(command: Commands) -> Result<String, Box<dyn Error>> {
             compiled_circuit,
             transcript,
             num_runs,
-            compress_selectors,
+            disable_selector_compression,
         } => fuzz(
             compiled_circuit,
             witness,
             transcript,
             num_runs,
-            compress_selectors,
+            disable_selector_compression,
         ),
         Commands::GenSrs { srs_path, logrows } => gen_srs_cmd(srs_path, logrows as u32),
         #[cfg(not(target_arch = "wasm32"))]
@@ -154,8 +159,7 @@ pub async fn run(command: Commands) -> Result<String, Box<dyn Error>> {
             srs_path,
             settings_path,
             logrows,
-            check,
-        } => get_srs_cmd(srs_path, settings_path, logrows, check).await,
+        } => get_srs_cmd(srs_path, settings_path, logrows).await,
         Commands::Table { model, args } => table(model, args),
         #[cfg(feature = "render")]
         Commands::RenderCircuit {
@@ -260,14 +264,14 @@ pub async fn run(command: Commands) -> Result<String, Box<dyn Error>> {
             vk_path,
             pk_path,
             witness,
-            compress_selectors,
+            disable_selector_compression,
         } => setup(
             compiled_circuit,
             srs_path,
             vk_path,
             pk_path,
             witness,
-            compress_selectors,
+            disable_selector_compression,
         ),
         #[cfg(not(target_arch = "wasm32"))]
         Commands::SetupTestEvmData {
@@ -331,7 +335,7 @@ pub async fn run(command: Commands) -> Result<String, Box<dyn Error>> {
             srs_path,
             logrows,
             split_proofs,
-            compress_selectors,
+            disable_selector_compression,
         } => setup_aggregate(
             sample_snarks,
             vk_path,
@@ -339,7 +343,7 @@ pub async fn run(command: Commands) -> Result<String, Box<dyn Error>> {
             srs_path,
             logrows,
             split_proofs,
-            compress_selectors,
+            disable_selector_compression,
         ),
         Commands::Aggregate {
             proof_path,
@@ -487,16 +491,28 @@ async fn fetch_srs(uri: &str) -> Result<Vec<u8>, Box<dyn Error>> {
 }
 
 #[cfg(not(target_arch = "wasm32"))]
-fn check_srs_hash(logrows: u32, srs_path: Option<PathBuf>) -> Result<String, Box<dyn Error>> {
+pub(crate) fn get_file_hash(path: &PathBuf) -> Result<String, Box<dyn Error>> {
     use std::io::Read;
-
-    let path = get_srs_path(logrows, srs_path);
-    let file = std::fs::File::open(path.clone())?;
+    let file = std::fs::File::open(path)?;
+    let mut reader = std::io::BufReader::new(file);
     let mut buffer = vec![];
-    let bytes_read = std::io::BufReader::new(file).read_to_end(&mut buffer)?;
-    debug!("read {} bytes from SRS file", bytes_read);
+    let bytes_read = reader.read_to_end(&mut buffer)?;
+    info!(
+        "read {} bytes from file (vector of len = {})",
+        bytes_read,
+        buffer.len()
+    );
+
     let hash = sha256::digest(buffer);
-    info!("SRS hash: {}", hash);
+    info!("file hash: {}", hash);
+
+    Ok(hash)
+}
+
+#[cfg(not(target_arch = "wasm32"))]
+fn check_srs_hash(logrows: u32, srs_path: Option<PathBuf>) -> Result<String, Box<dyn Error>> {
+    let path = get_srs_path(logrows, srs_path);
+    let hash = get_file_hash(&path)?;
 
     let predefined_hash = match { crate::srs_sha::PUBLIC_SRS_SHA256_HASHES.get(&logrows) } {
         Some(h) => h,
@@ -520,7 +536,6 @@ pub(crate) async fn get_srs_cmd(
     srs_path: Option<PathBuf>,
     settings_path: Option<PathBuf>,
     logrows: Option<u32>,
-    check_mode: CheckMode,
 ) -> Result<String, Box<dyn Error>> {
     // logrows overrides settings
 
@@ -548,18 +563,21 @@ pub(crate) async fn get_srs_cmd(
         let srs_uri = format!("{}{}", PUBLIC_SRS_URL, k);
         let mut reader = Cursor::new(fetch_srs(&srs_uri).await?);
         // check the SRS
-        if matches!(check_mode, CheckMode::SAFE) {
-            #[cfg(not(target_arch = "wasm32"))]
-            let pb = init_spinner();
-            #[cfg(not(target_arch = "wasm32"))]
-            pb.set_message("Validating SRS (this may take a while) ...");
-            ParamsKZG::<Bn256>::read(&mut reader)?;
-            #[cfg(not(target_arch = "wasm32"))]
-            pb.finish_with_message("SRS validated");
-        }
+        #[cfg(not(target_arch = "wasm32"))]
+        let pb = init_spinner();
+        #[cfg(not(target_arch = "wasm32"))]
+        pb.set_message("Validating SRS (this may take a while) ...");
+        let params = ParamsKZG::<Bn256>::read(&mut reader)?;
+        #[cfg(not(target_arch = "wasm32"))]
+        pb.finish_with_message("SRS validated.");
 
+        info!("Saving SRS to disk...");
         let mut file = std::fs::File::create(get_srs_path(k, srs_path.clone()))?;
-        file.write_all(reader.get_ref())?;
+        let mut buffer = BufWriter::with_capacity(*EZKL_BUF_CAPACITY, &mut file);
+        params.write(&mut buffer)?;
+
+        info!("Saved SRS to disk.");
+
         info!("SRS downloaded");
     } else {
         info!("SRS already exists at that path");
@@ -806,7 +824,6 @@ pub(crate) fn calibrate(
     let settings = GraphSettings::load(&settings_path)?;
     // now retrieve the run args
     // we load the model to get the input and output shapes
-    // check if gag already exists
 
     let model = Model::from_run_args(&settings.run_args, &model_path)?;
 
@@ -887,7 +904,12 @@ pub(crate) fn calibrate(
             input_scale, param_scale, scale_rebase_multiplier, div_rebasing
         ));
 
-        let key = (input_scale, param_scale, scale_rebase_multiplier);
+        let key = (
+            input_scale,
+            param_scale,
+            scale_rebase_multiplier,
+            div_rebasing,
+        );
         forward_pass_res.insert(key, vec![]);
 
         let local_run_args = RunArgs {
@@ -898,6 +920,18 @@ pub(crate) fn calibrate(
             ..settings.run_args.clone()
         };
 
+        // if unix get a gag
+        #[cfg(unix)]
+        let _r = match Gag::stdout() {
+            Ok(g) => Some(g),
+            _ => None,
+        };
+        #[cfg(unix)]
+        let _g = match Gag::stderr() {
+            Ok(g) => Some(g),
+            _ => None,
+        };
+
         let mut circuit = match GraphCircuit::from_run_args(&local_run_args, &model_path) {
             Ok(c) => c,
             Err(e) => {
@@ -938,29 +972,27 @@ pub(crate) fn calibrate(
             }
         }
 
-        let min_lookup_range = forward_pass_res
-            .get(&key)
-            .unwrap()
+        // drop the gag
+        #[cfg(unix)]
+        drop(_r);
+        #[cfg(unix)]
+        drop(_g);
+
+        let result = forward_pass_res.get(&key).ok_or("key not found")?;
+
+        let min_lookup_range = result
             .iter()
             .map(|x| x.min_lookup_inputs)
             .min()
             .unwrap_or(0);
 
-        let max_lookup_range = forward_pass_res
-            .get(&key)
-            .unwrap()
+        let max_lookup_range = result
             .iter()
             .map(|x| x.max_lookup_inputs)
             .max()
             .unwrap_or(0);
 
-        let max_range_size = forward_pass_res
-            .get(&key)
-            .unwrap()
-            .iter()
-            .map(|x| x.max_range_size)
-            .max()
-            .unwrap_or(0);
+        let max_range_size = result.iter().map(|x| x.max_range_size).max().unwrap_or(0);
 
         let res = circuit.calc_min_logrows(
             (min_lookup_range, max_lookup_range),
@@ -1081,6 +1113,7 @@ pub(crate) fn calibrate(
             best_params.run_args.input_scale,
             best_params.run_args.param_scale,
             best_params.run_args.scale_rebase_multiplier,
+            best_params.run_args.div_rebasing,
         ))
         .ok_or("no params found")?
         .iter()
@@ -1499,7 +1532,7 @@ pub(crate) fn setup(
     vk_path: PathBuf,
     pk_path: PathBuf,
     witness: Option<PathBuf>,
-    compress_selectors: bool,
+    disable_selector_compression: bool,
 ) -> Result<String, Box<dyn Error>> {
     // these aren't real values so the sanity checks are mostly meaningless
     let mut circuit = GraphCircuit::load(compiled_circuit)?;
@@ -1513,7 +1546,7 @@ pub(crate) fn setup(
     let pk = create_keys::<KZGCommitmentScheme<Bn256>, Fr, GraphCircuit>(
         &circuit,
         &params,
-        compress_selectors,
+        disable_selector_compression,
     )
     .map_err(Box::<dyn Error>::from)?;
 
@@ -1654,7 +1687,7 @@ pub(crate) fn fuzz(
     data_path: PathBuf,
     transcript: TranscriptType,
     num_runs: usize,
-    compress_selectors: bool,
+    disable_selector_compression: bool,
 ) -> Result<String, Box<dyn Error>> {
     check_solc_requirement();
     let passed = AtomicBool::new(true);
@@ -1664,7 +1697,7 @@ pub(crate) fn fuzz(
     let logrows = circuit.settings().run_args.logrows;
 
     info!("setting up tests");
-
+    #[cfg(unix)]
     let _r = Gag::stdout()?;
     let params = gen_srs::<KZGCommitmentScheme<Bn256>>(logrows);
 
@@ -1673,7 +1706,7 @@ pub(crate) fn fuzz(
     let pk = create_keys::<KZGCommitmentScheme<Bn256>, Fr, GraphCircuit>(
         &circuit,
         &params,
-        compress_selectors,
+        disable_selector_compression,
     )
     .map_err(Box::<dyn Error>::from)?;
 
@@ -1682,6 +1715,7 @@ pub(crate) fn fuzz(
     let public_inputs = circuit.prepare_public_inputs(&data)?;
 
     let strategy = KZGSingleStrategy::new(&params);
+    #[cfg(unix)]
     std::mem::drop(_r);
 
     info!("starting fuzzing");
@@ -1694,7 +1728,7 @@ pub(crate) fn fuzz(
         let bad_pk = create_keys::<KZGCommitmentScheme<Bn256>, Fr, GraphCircuit>(
             &circuit,
             &new_params,
-            compress_selectors,
+            disable_selector_compression,
         )
         .map_err(|_| ())?;
 
@@ -1772,7 +1806,7 @@ pub(crate) fn fuzz(
         let bad_pk = create_keys::<KZGCommitmentScheme<Bn256>, Fr, GraphCircuit>(
             &circuit,
             &new_params,
-            compress_selectors,
+            disable_selector_compression,
         )
         .map_err(|_| ())?;
 
@@ -1872,6 +1906,7 @@ pub(crate) fn run_fuzz_fn(
     passed: &AtomicBool,
 ) {
     let num_failures = AtomicI64::new(0);
+    #[cfg(unix)]
     let _r = Gag::stdout().unwrap();
 
     let pb = init_bar(num_runs as u64);
@@ -1885,6 +1920,7 @@ pub(crate) fn run_fuzz_fn(
         pb.inc(1);
     });
     pb.finish_with_message("Done.");
+    #[cfg(unix)]
     std::mem::drop(_r);
     info!(
         "num failures: {} out of {}",
@@ -1951,7 +1987,7 @@ pub(crate) fn setup_aggregate(
     srs_path: Option<PathBuf>,
     logrows: u32,
     split_proofs: bool,
-    compress_selectors: bool,
+    disable_selector_compression: bool,
 ) -> Result<String, Box<dyn Error>> {
     // the K used for the aggregation circuit
     let params = load_params_cmd(srs_path, logrows)?;
@@ -1965,7 +2001,7 @@ pub(crate) fn setup_aggregate(
     let agg_pk = create_keys::<KZGCommitmentScheme<Bn256>, Fr, AggregationCircuit>(
         &agg_circuit,
         &params,
-        compress_selectors,
+        disable_selector_compression,
     )?;
 
     let agg_vk = agg_pk.get_vk();
@@ -2042,7 +2078,8 @@ pub(crate) fn verify(
     let circuit_settings = GraphSettings::load(&settings_path)?;
 
     let params = if reduced_srs {
-        load_params_cmd(srs_path, circuit_settings.log2_total_instances())?
+        // only need G_0 for the verification with shplonk
+        load_params_cmd(srs_path, 1)?
     } else {
         load_params_cmd(srs_path, circuit_settings.run_args.logrows)?
     };

src/graph/mod.rs

@@ -12,6 +12,8 @@ pub mod utilities;
 pub mod vars;
 #[cfg(not(target_arch = "wasm32"))]
 use colored_json::ToColoredJson;
+#[cfg(unix)]
+use gag::Gag;
 use halo2_proofs::plonk::VerifyingKey;
 use halo2_proofs::poly::kzg::commitment::ParamsKZG;
 pub use input::DataSource;
@@ -453,6 +455,10 @@ pub struct GraphSettings {
     pub total_dynamic_col_size: usize,
     /// number of dynamic lookups
     pub num_dynamic_lookups: usize,
+    /// number of shuffles
+    pub num_shuffles: usize,
+    /// total shuffle column size
+    pub total_shuffle_col_size: usize,
     /// the shape of public inputs to the model (in order of appearance)
     pub model_instance_shapes: Vec<Vec<usize>>,
     /// model output scales
@@ -482,8 +488,14 @@ impl GraphSettings {
             .ceil() as u32
     }
 
-    fn dynamic_lookup_logrows(&self) -> u32 {
-        (self.total_dynamic_col_size as f64).log2().ceil() as u32
+    fn dynamic_lookup_and_shuffle_logrows(&self) -> u32 {
+        (self.total_dynamic_col_size as f64 + self.total_shuffle_col_size as f64)
+            .log2()
+            .ceil() as u32
+    }
+
+    fn dynamic_lookup_and_shuffle_col_size(&self) -> usize {
+        self.total_dynamic_col_size + self.total_shuffle_col_size
     }
 
     fn module_constraint_logrows(&self) -> u32 {
@@ -583,6 +595,11 @@ impl GraphSettings {
         self.num_dynamic_lookups > 0
     }
 
+    /// requires dynamic shuffle
+    pub fn requires_shuffle(&self) -> bool {
+        self.num_shuffles > 0
+    }
+
     /// any kzg visibility
     pub fn module_requires_kzg(&self) -> bool {
         self.run_args.input_visibility.is_kzgcommit()
@@ -1097,7 +1114,7 @@ impl GraphCircuit {
         // These are hard lower limits, we can't overflow instances or modules constraints
         let instance_logrows = self.settings().log2_total_instances();
         let module_constraint_logrows = self.settings().module_constraint_logrows();
-        let dynamic_lookup_logrows = self.settings().dynamic_lookup_logrows();
+        let dynamic_lookup_logrows = self.settings().dynamic_lookup_and_shuffle_logrows();
         min_logrows = std::cmp::max(
             min_logrows,
             // max of the instance logrows and the module constraint logrows and the dynamic lookup logrows is the lower limit
@@ -1181,9 +1198,9 @@ impl GraphCircuit {
         max_range_size: i128,
     ) -> bool {
         // if num cols is too large then the extended k is too large
-        if Self::calc_num_cols(safe_lookup_range.1 - safe_lookup_range.0, k) > MAX_NUM_LOOKUP_COLS {
-            return false;
-        } else if Self::calc_num_cols(max_range_size, k) > MAX_NUM_LOOKUP_COLS {
+        if Self::calc_num_cols(safe_lookup_range.1 - safe_lookup_range.0, k) > MAX_NUM_LOOKUP_COLS
+            || Self::calc_num_cols(max_range_size, k) > MAX_NUM_LOOKUP_COLS
+        {
             return false;
         }
 
@@ -1192,7 +1209,26 @@ impl GraphCircuit {
         settings.run_args.logrows = k;
         settings.required_range_checks = vec![(0, max_range_size)];
         let mut cs = ConstraintSystem::default();
+        // if unix get a gag
+        #[cfg(unix)]
+        let _r = match Gag::stdout() {
+            Ok(g) => Some(g),
+            _ => None,
+        };
+        #[cfg(unix)]
+        let _g = match Gag::stderr() {
+            Ok(g) => Some(g),
+            _ => None,
+        };
+
         Self::configure_with_params(&mut cs, settings);
+
+        // drop the gag
+        #[cfg(unix)]
+        drop(_r);
+        #[cfg(unix)]
+        drop(_g);
+
         #[cfg(feature = "mv-lookup")]
         let cs = cs.chunk_lookups();
         // quotient_poly_degree * params.n - 1 is the degree of the quotient polynomial

src/graph/model.rs

@@ -103,6 +103,10 @@ pub struct DummyPassRes {
     pub num_dynamic_lookups: usize,
     /// dynamic lookup col size
     pub dynamic_lookup_col_coord: usize,
+    /// num shuffles
+    pub num_shuffles: usize,
+    /// shuffle
+    pub shuffle_col_coord: usize,
     /// linear coordinate
     pub linear_coord: usize,
     /// total const size
@@ -546,6 +550,8 @@ impl Model {
             model_input_scales: self.graph.get_input_scales(),
             num_dynamic_lookups: res.num_dynamic_lookups,
             total_dynamic_col_size: res.dynamic_lookup_col_coord,
+            num_shuffles: res.num_shuffles,
+            total_shuffle_col_size: res.shuffle_col_coord,
             total_const_size: res.total_const_size,
             check_mode,
             version: env!("CARGO_PKG_VERSION").to_string(),
@@ -1019,7 +1025,6 @@ impl Model {
 
         let lookup_range = settings.run_args.lookup_range;
         let logrows = settings.run_args.logrows as usize;
-        let num_dynamic_lookups = settings.num_dynamic_lookups;
         let required_lookups = settings.required_lookups.clone();
         let required_range_checks = settings.required_range_checks.clone();
 
@@ -1041,8 +1046,16 @@ impl Model {
             base_gate.configure_range_check(meta, input, index, range, logrows)?;
         }
 
-        for _ in 0..num_dynamic_lookups {
+        if settings.requires_dynamic_lookup() {
             base_gate.configure_dynamic_lookup(
+                meta,
+                vars.advices[0..3].try_into()?,
+                vars.advices[3..6].try_into()?,
+            )?;
+        }
+
+        if settings.requires_shuffle() {
+            base_gate.configure_shuffles(
                 meta,
                 vars.advices[0..2].try_into()?,
                 vars.advices[3..5].try_into()?,
@@ -1456,6 +1469,8 @@ impl Model {
             max_range_size: region.max_range_size(),
             num_dynamic_lookups: region.dynamic_lookup_index(),
             dynamic_lookup_col_coord: region.dynamic_lookup_col_coord(),
+            num_shuffles: region.shuffle_index(),
+            shuffle_col_coord: region.shuffle_col_coord(),
             outputs,
         };
 

src/graph/utilities.rs

@@ -23,7 +23,10 @@ use std::sync::Arc;
 use tract_onnx::prelude::{DatumType, Node as OnnxNode, TypedFact, TypedOp};
 #[cfg(not(target_arch = "wasm32"))]
 use tract_onnx::tract_core::ops::{
-    array::{Gather, GatherElements, MultiBroadcastTo, OneHot, ScatterElements, Slice, Topk},
+    array::{
+        Gather, GatherElements, GatherNd, MultiBroadcastTo, OneHot, ScatterElements, ScatterNd,
+        Slice, Topk,
+    },
     change_axes::AxisOp,
     cnn::{Conv, Deconv},
     einsum::EinSum,
@@ -467,6 +470,78 @@ pub fn new_op_from_onnx(
 
             // Extract the max value
         }
+        "ScatterNd" => {
+            if inputs.len() != 3 {
+                return Err(Box::new(GraphError::InvalidDims(
+                    idx,
+                    "scatter nd".to_string(),
+                )));
+            };
+            // just verify it deserializes correctly
+            let _op = load_op::<ScatterNd>(node.op(), idx, node.op().name().to_string())?;
+
+            let mut op = SupportedOp::Linear(crate::circuit::ops::poly::PolyOp::ScatterND {
+                constant_idx: None,
+            });
+
+            // if param_visibility.is_public() {
+            if let Some(c) = inputs[1].opkind().get_mutable_constant() {
+                inputs[1].decrement_use();
+                deleted_indices.push(inputs.len() - 1);
+                op = SupportedOp::Linear(crate::circuit::ops::poly::PolyOp::ScatterND {
+                    constant_idx: Some(c.raw_values.map(|x| x as usize)),
+                })
+            }
+            // }
+
+            if inputs[1].opkind().is_input() {
+                inputs[1].replace_opkind(SupportedOp::Input(crate::circuit::ops::Input {
+                    scale: 0,
+                    datum_type: InputType::TDim,
+                }));
+                inputs[1].bump_scale(0);
+            }
+
+            op
+        }
+
+        "GatherNd" => {
+            if inputs.len() != 2 {
+                return Err(Box::new(GraphError::InvalidDims(
+                    idx,
+                    "gather nd".to_string(),
+                )));
+            };
+            let op = load_op::<GatherNd>(node.op(), idx, node.op().name().to_string())?;
+            let batch_dims = op.batch_dims;
+
+            let mut op = SupportedOp::Linear(crate::circuit::ops::poly::PolyOp::GatherND {
+                batch_dims,
+                indices: None,
+            });
+
+            // if param_visibility.is_public() {
+            if let Some(c) = inputs[1].opkind().get_mutable_constant() {
+                inputs[1].decrement_use();
+                deleted_indices.push(inputs.len() - 1);
+                op = SupportedOp::Linear(crate::circuit::ops::poly::PolyOp::GatherND {
+                    batch_dims,
+                    indices: Some(c.raw_values.map(|x| x as usize)),
+                })
+            }
+            // }
+
+            if inputs[1].opkind().is_input() {
+                inputs[1].replace_opkind(SupportedOp::Input(crate::circuit::ops::Input {
+                    scale: 0,
+                    datum_type: InputType::TDim,
+                }));
+                inputs[1].bump_scale(0);
+            }
+
+            op
+        }
+
         "GatherElements" => {
             if inputs.len() != 2 {
                 return Err(Box::new(GraphError::InvalidDims(

src/graph/vars.rs

@@ -429,17 +429,20 @@ impl<F: PrimeField + TensorType + PartialOrd> ModelVars<F> {
         let num_constants = params.total_const_size;
         let module_requires_fixed = params.module_requires_fixed();
         let requires_dynamic_lookup = params.requires_dynamic_lookup();
-        let dynamic_lookup_size = params.total_dynamic_col_size;
+        let requires_shuffle = params.requires_shuffle();
+        let dynamic_lookup_and_shuffle_size = params.dynamic_lookup_and_shuffle_col_size();
 
         let mut advices = (0..3)
             .map(|_| VarTensor::new_advice(cs, logrows, num_inner_cols, var_len))
             .collect_vec();
 
-        if requires_dynamic_lookup {
-            for _ in 0..2 {
-                let dynamic_lookup = VarTensor::new_advice(cs, logrows, 1, dynamic_lookup_size);
+        if requires_dynamic_lookup || requires_shuffle {
+            let num_cols = if requires_dynamic_lookup { 3 } else { 2 };
+            for _ in 0..num_cols {
+                let dynamic_lookup =
+                    VarTensor::new_advice(cs, logrows, 1, dynamic_lookup_and_shuffle_size);
                 if dynamic_lookup.num_blocks() > 1 {
-                    panic!("dynamic lookup should only have one block");
+                    panic!("dynamic lookup or shuffle should only have one block");
                 };
                 advices.push(dynamic_lookup);
             }

src/pfsys/mod.rs

@@ -484,7 +484,7 @@ where
 pub fn create_keys<Scheme: CommitmentScheme, F: PrimeField + TensorType, C: Circuit<F>>(
     circuit: &C,
     params: &'_ Scheme::ParamsProver,
-    compress_selectors: bool,
+    disable_selector_compression: bool,
 ) -> Result<ProvingKey<Scheme::Curve>, halo2_proofs::plonk::Error>
 where
     C: Circuit<Scheme::Scalar>,
@@ -496,7 +496,7 @@ where
     // Initialize verifying key
     let now = Instant::now();
     trace!("preparing VK");
-    let vk = keygen_vk_custom(params, &empty_circuit, compress_selectors)?;
+    let vk = keygen_vk_custom(params, &empty_circuit, !disable_selector_compression)?;
     let elapsed = now.elapsed();
     info!("VK took {}.{}", elapsed.as_secs(), elapsed.subsec_millis());
 

src/python.rs

@@ -484,7 +484,6 @@ fn get_srs(
             srs_path,
             settings_path,
             logrows,
-            CheckMode::SAFE,
         ))
         .map_err(|e| {
             let err_str = format!("Failed to get srs: {}", e);
@@ -622,7 +621,7 @@ fn mock_aggregate(
     pk_path=PathBuf::from(DEFAULT_PK),
     srs_path=None,
     witness_path = None,
-    compress_selectors=DEFAULT_COMPRESS_SELECTORS.parse().unwrap(),
+    disable_selector_compression=DEFAULT_DISABLE_SELECTOR_COMPRESSION.parse().unwrap(),
 ))]
 fn setup(
     model: PathBuf,
@@ -630,7 +629,7 @@ fn setup(
     pk_path: PathBuf,
     srs_path: Option<PathBuf>,
     witness_path: Option<PathBuf>,
-    compress_selectors: bool,
+    disable_selector_compression: bool,
 ) -> Result<bool, PyErr> {
     crate::execute::setup(
         model,
@@ -638,7 +637,7 @@ fn setup(
         vk_path,
         pk_path,
         witness_path,
-        compress_selectors,
+        disable_selector_compression,
     )
     .map_err(|e| {
         let err_str = format!("Failed to run setup: {}", e);
@@ -719,7 +718,7 @@ fn verify(
     logrows=DEFAULT_AGGREGATED_LOGROWS.parse().unwrap(),
     split_proofs = false,
     srs_path = None,
-    compress_selectors=DEFAULT_COMPRESS_SELECTORS.parse().unwrap(),
+    disable_selector_compression=DEFAULT_DISABLE_SELECTOR_COMPRESSION.parse().unwrap(),
 ))]
 fn setup_aggregate(
     sample_snarks: Vec<PathBuf>,
@@ -728,7 +727,7 @@ fn setup_aggregate(
     logrows: u32,
     split_proofs: bool,
     srs_path: Option<PathBuf>,
-    compress_selectors: bool,
+    disable_selector_compression: bool,
 ) -> Result<bool, PyErr> {
     crate::execute::setup_aggregate(
         sample_snarks,
@@ -737,7 +736,7 @@ fn setup_aggregate(
         srs_path,
         logrows,
         split_proofs,
-        compress_selectors,
+        disable_selector_compression,
     )
     .map_err(|e| {
         let err_str = format!("Failed to setup aggregate: {}", e);
@@ -1104,6 +1103,7 @@ fn ezkl(_py: Python<'_>, m: &PyModule) -> PyResult<()> {
     m.add_class::<PyG1Affine>()?;
     m.add_class::<PyG1>()?;
     m.add_class::<PyTestDataSource>()?;
+    m.add("__version__", env!("CARGO_PKG_VERSION"))?;
     m.add_function(wrap_pyfunction!(felt_to_big_endian, m)?)?;
     m.add_function(wrap_pyfunction!(felt_to_int, m)?)?;
     m.add_function(wrap_pyfunction!(felt_to_float, m)?)?;

src/tensor/mod.rs

@@ -673,6 +673,68 @@ impl<T: Clone + TensorType> Tensor<T> {
         Tensor::new(Some(&res), &dims)
     }
 
+    /// Set a slice of the Tensor.
+    /// ```
+    /// use ezkl::tensor::Tensor;
+    /// let mut a = Tensor::<i32>::new(Some(&[1, 2, 3, 4, 5, 6]), &[2, 3]).unwrap();
+    /// let b = Tensor::<i32>::new(Some(&[1, 2, 3, 1, 2, 3]), &[2, 3]).unwrap();
+    /// a.set_slice(&[1..2], &Tensor::<i32>::new(Some(&[1, 2, 3]), &[1, 3]).unwrap()).unwrap();
+    /// assert_eq!(a, b);
+    /// ```
+    pub fn set_slice(
+        &mut self,
+        indices: &[Range<usize>],
+        value: &Tensor<T>,
+    ) -> Result<(), TensorError>
+    where
+        T: Send + Sync,
+    {
+        if indices.is_empty() {
+            return Ok(());
+        }
+        if self.dims.len() < indices.len() {
+            return Err(TensorError::DimError(format!(
+                "The dimensionality of the slice {:?} is greater than the tensor's {:?}",
+                indices, self.dims
+            )));
+        }
+
+        // if indices weren't specified we fill them in as required
+        let mut full_indices = indices.to_vec();
+
+        let omitted_dims = (indices.len()..self.dims.len())
+            .map(|i| self.dims[i])
+            .collect::<Vec<_>>();
+
+        for dim in &omitted_dims {
+            full_indices.push(0..*dim);
+        }
+
+        let full_dims = full_indices
+            .iter()
+            .map(|x| x.end - x.start)
+            .collect::<Vec<_>>();
+
+        // now broadcast the value to the full dims
+        let value = value.expand(&full_dims)?;
+
+        let cartesian_coord: Vec<Vec<usize>> = full_indices
+            .iter()
+            .cloned()
+            .multi_cartesian_product()
+            .collect();
+
+        let _ = cartesian_coord
+            .iter()
+            .enumerate()
+            .map(|(i, e)| {
+                self.set(e, value[i].clone());
+            })
+            .collect::<Vec<_>>();
+
+        Ok(())
+    }
+
     /// Get the array index from rows / columns indices.
     ///
     /// ```
@@ -1526,18 +1588,20 @@ pub fn get_broadcasted_shape(
     let num_dims_a = shape_a.len();
     let num_dims_b = shape_b.len();
 
-    // reewrite the below using match
-    if num_dims_a == num_dims_b {
-        let mut broadcasted_shape = Vec::with_capacity(num_dims_a);
-        for (dim_a, dim_b) in shape_a.iter().zip(shape_b.iter()) {
-            let max_dim = dim_a.max(dim_b);
-            broadcasted_shape.push(*max_dim);
+    match (num_dims_a, num_dims_b) {
+        (a, b) if a == b => {
+            let mut broadcasted_shape = Vec::with_capacity(num_dims_a);
+            for (dim_a, dim_b) in shape_a.iter().zip(shape_b.iter()) {
+                let max_dim = dim_a.max(dim_b);
+                broadcasted_shape.push(*max_dim);
+            }
+            Ok(broadcasted_shape)
         }
-        Ok(broadcasted_shape)
-    } else if num_dims_a < num_dims_b {
-        Ok(shape_b.to_vec())
-    } else {
-        Ok(shape_a.to_vec())
+        (a, b) if a < b => Ok(shape_b.to_vec()),
+        (a, b) if a > b => Ok(shape_a.to_vec()),
+        _ => Err(Box::new(TensorError::DimError(
+            "Unknown condition for broadcasting".to_string(),
+        ))),
     }
 }
 ////////////////////////

src/tensor/ops.rs

@@ -2,7 +2,7 @@ use super::TensorError;
 use crate::tensor::{Tensor, TensorType};
 use itertools::Itertools;
 use maybe_rayon::{
-    iter::IndexedParallelIterator, iter::IntoParallelRefMutIterator, iter::ParallelIterator,
+    iter::{IndexedParallelIterator, IntoParallelRefMutIterator, ParallelIterator},
     prelude::IntoParallelRefIterator,
 };
 use std::collections::{HashMap, HashSet};
@@ -1328,6 +1328,316 @@ pub fn gather_elements<T: TensorType + Send + Sync>(
     Ok(output)
 }
 
+/// Gather ND.
+/// # Arguments
+/// * `input` - Tensor
+/// * `index` - Tensor of indices to gather
+/// * `batch_dims` - Number of batch dimensions
+/// # Examples
+/// ```
+/// use ezkl::tensor::Tensor;
+/// use ezkl::tensor::ops::gather_nd;
+/// let x = Tensor::<i128>::new(
+///   Some(&[0, 1, 2, 3]),
+/// &[2, 2],
+/// ).unwrap();
+/// let index = Tensor::<usize>::new(
+/// Some(&[0, 0, 1, 1]),
+/// &[2, 2],
+/// ).unwrap();
+/// let result = gather_nd(&x, &index, 0).unwrap();
+/// let expected = Tensor::<i128>::new(Some(&[0, 3]), &[2]).unwrap();
+/// assert_eq!(result, expected);
+///
+/// let index = Tensor::<usize>::new(
+/// Some(&[1, 0]),
+/// &[2, 1],
+/// ).unwrap();
+/// let result = gather_nd(&x, &index, 0).unwrap();
+/// let expected = Tensor::<i128>::new(Some(&[2, 3, 0, 1]), &[2, 2]).unwrap();
+/// assert_eq!(result, expected);
+///
+/// let x = Tensor::<i128>::new(
+///  Some(&[0, 1, 2, 3, 4, 5, 6, 7]),
+/// &[2, 2, 2],
+/// ).unwrap();
+/// let index = Tensor::<usize>::new(
+///  Some(&[0, 1, 1, 0]),
+/// &[2, 2],
+/// ).unwrap();
+/// let result = gather_nd(&x, &index, 0).unwrap();
+/// let expected = Tensor::<i128>::new(Some(&[2, 3, 4, 5]), &[2, 2]).unwrap();
+/// assert_eq!(result, expected);
+///
+/// let index = Tensor::<usize>::new(
+///  Some(&[0, 1, 1, 0]),
+/// &[2, 1, 2],
+/// ).unwrap();
+/// let result = gather_nd(&x, &index, 0).unwrap();
+/// let expected = Tensor::<i128>::new(Some(&[2, 3, 4, 5]), &[2, 1, 2]).unwrap();
+/// assert_eq!(result, expected);
+///
+/// let index = Tensor::<usize>::new(
+/// Some(&[1, 0]),
+/// &[2, 1],
+/// ).unwrap();
+/// let result = gather_nd(&x, &index, 1).unwrap();
+/// let expected = Tensor::<i128>::new(Some(&[2, 3, 4, 5]), &[2, 2]).unwrap();
+/// assert_eq!(result, expected);
+///
+/// let index = Tensor::<usize>::new(
+///  Some(&[0, 1, 0, 0, 1, 1, 1, 0, 0, 1, 0, 1]),
+/// &[2, 2, 3],
+/// ).unwrap();
+/// let result = gather_nd(&x, &index, 0).unwrap();
+/// let expected = Tensor::<i128>::new(Some(&[2, 3, 4, 5]), &[2, 2]).unwrap();
+/// assert_eq!(result, expected);
+///
+/// let index = Tensor::<usize>::new(
+///  Some(&[0, 1, 0, 0, 1, 1, 1, 0]),
+/// &[2, 2, 2],
+/// ).unwrap();
+/// let result = gather_nd(&x, &index, 0).unwrap();
+/// let expected = Tensor::<i128>::new(Some(&[2, 3, 0, 1, 6, 7, 4, 5]), &[2, 2, 2]).unwrap();
+/// assert_eq!(result, expected);
+///
+/// let index = Tensor::<usize>::new(
+///  Some(&[0, 1, 0, 1, 1, 1]),
+/// &[2, 3],
+/// ).unwrap();
+/// let result = gather_nd(&x, &index, 0).unwrap();
+/// let expected = Tensor::<i128>::new(Some(&[2, 7]), &[2]).unwrap();
+/// assert_eq!(result, expected);
+///
+pub fn gather_nd<T: TensorType + Send + Sync>(
+    input: &Tensor<T>,
+    index: &Tensor<usize>,
+    batch_dims: usize,
+) -> Result<Tensor<T>, TensorError> {
+    // Calculate the output tensor size
+    let index_dims = index.dims().to_vec();
+    let input_dims = input.dims().to_vec();
+    let last_value = index_dims
+        .last()
+        .ok_or(TensorError::DimMismatch("gather_nd".to_string()))?;
+    if last_value > &(input_dims.len() - batch_dims) {
+        return Err(TensorError::DimMismatch("gather_nd".to_string()));
+    }
+
+    let output_size =
+    // If indices_shape[-1] == r-b, since the rank of indices is q,
+    // indices can be thought of as N (q-b-1)-dimensional tensors containing 1-D tensors of dimension r-b,
+    // where N is an integer equals to the product of 1 and all the elements in the batch dimensions of the indices_shape.
+    // Let us think of each such r-b ranked tensor as indices_slice.
+    // Each scalar value corresponding to data[0:b-1,indices_slice] is filled into
+    // the corresponding location of the (q-b-1)-dimensional tensor to form the output tensor
+     // if indices_shape[-1] < r-b, since the rank of indices is q, indices can be thought of as N (q-b-1)-dimensional tensor containing 1-D tensors of dimension < r-b.
+    // Let us think of each such tensors as indices_slice.
+    // Each tensor slice corresponding to data[0:b-1, indices_slice , :] is filled into the corresponding location of the (q-b-1)-dimensional tensor to form the output tensor
+    {
+        let output_rank = input_dims.len() + index_dims.len() - 1 - batch_dims - last_value;
+
+        let mut dims = index_dims[..index_dims.len() - 1].to_vec();
+        let input_offset = batch_dims + last_value;
+        dims.extend(input_dims[input_offset..input_dims.len()].to_vec());
+
+        assert_eq!(output_rank, dims.len());
+        dims
+
+    };
+
+    // cartesian coord over batch dims
+    let mut batch_cartesian_coord = input_dims[0..batch_dims]
+        .iter()
+        .map(|x| 0..*x)
+        .multi_cartesian_product()
+        .collect::<Vec<_>>();
+
+    if batch_cartesian_coord.is_empty() {
+        batch_cartesian_coord.push(vec![]);
+    }
+
+    let outputs = batch_cartesian_coord
+        .par_iter()
+        .map(|batch_coord| {
+            let batch_slice = batch_coord.iter().map(|x| *x..*x + 1).collect::<Vec<_>>();
+            let mut index_slice = index.get_slice(&batch_slice)?;
+            index_slice.reshape(&index.dims()[batch_dims..])?;
+            let mut input_slice = input.get_slice(&batch_slice)?;
+            input_slice.reshape(&input.dims()[batch_dims..])?;
+
+            let mut inner_cartesian_coord = index_slice.dims()[0..index_slice.dims().len() - 1]
+                .iter()
+                .map(|x| 0..*x)
+                .multi_cartesian_product()
+                .collect::<Vec<_>>();
+
+            if inner_cartesian_coord.is_empty() {
+                inner_cartesian_coord.push(vec![]);
+            }
+
+            let output = inner_cartesian_coord
+                .iter()
+                .map(|coord| {
+                    let slice = coord
+                        .iter()
+                        .map(|x| *x..*x + 1)
+                        .chain(batch_coord.iter().map(|x| *x..*x + 1))
+                        .collect::<Vec<_>>();
+
+                    let index_slice = index_slice
+                        .get_slice(&slice)
+                        .unwrap()
+                        .iter()
+                        .map(|x| *x..*x + 1)
+                        .collect::<Vec<_>>();
+
+                    input_slice.get_slice(&index_slice).unwrap()
+                })
+                .collect::<Tensor<_>>();
+
+            output.combine()
+        })
+        .collect::<Result<Vec<_>, _>>()?;
+
+    let mut outputs = outputs.into_iter().flatten().collect::<Tensor<_>>();
+
+    outputs.reshape(&output_size)?;
+
+    Ok(outputs)
+}
+
+/// Scatter ND.
+/// This operator is the inverse of GatherND.
+/// # Arguments
+/// * `input` - Tensor
+/// * `index` - Tensor of indices to scatter
+/// * `src` - Tensor of src
+/// # Examples
+/// ```
+/// use ezkl::tensor::Tensor;
+/// use ezkl::tensor::ops::scatter_nd;
+/// let x = Tensor::<i128>::new(
+///  Some(&[1, 2, 3, 4, 5, 6, 7, 8]),
+/// &[8],
+/// ).unwrap();
+///
+/// let index = Tensor::<usize>::new(
+/// Some(&[4, 3, 1, 7]),
+/// &[4, 1],
+/// ).unwrap();
+/// let src = Tensor::<i128>::new(
+/// Some(&[9, 10, 11, 12]),
+/// &[4],
+/// ).unwrap();
+/// let result = scatter_nd(&x, &index, &src).unwrap();
+/// let expected = Tensor::<i128>::new(Some(&[1, 11, 3, 10, 9, 6, 7, 12]), &[8]).unwrap();
+/// assert_eq!(result, expected);
+///
+/// let x = Tensor::<i128>::new(
+///  Some(&[1, 2, 3, 4, 5, 6, 7, 8, 8, 7, 6, 5, 4, 3, 2, 1,
+///         1, 2, 3, 4, 5, 6, 7, 8, 8, 7, 6, 5, 4, 3, 2, 1,
+///         8, 7, 6, 5, 4, 3, 2, 1, 1, 2, 3, 4, 5, 6, 7, 8,
+///         8, 7, 6, 5, 4, 3, 2, 1, 1, 2, 3, 4, 5, 6, 7, 8]),
+/// &[4, 4, 4],
+/// ).unwrap();
+///
+/// let index = Tensor::<usize>::new(
+///   Some(&[0, 2]),
+/// &[2, 1],
+/// ).unwrap();
+///
+/// let src = Tensor::<i128>::new(
+///  Some(&[5, 5, 5, 5, 6, 6, 6, 6, 7, 7, 7, 7, 8, 8, 8, 8,
+///         1, 1, 1, 1, 2, 2, 2, 2, 3, 3, 3, 3, 4, 4, 4, 4,
+///   ]),
+/// &[2, 4, 4],
+/// ).unwrap();
+///
+/// let result = scatter_nd(&x, &index, &src).unwrap();
+///
+/// let expected = Tensor::<i128>::new(
+///  Some(&[5, 5, 5, 5, 6, 6, 6, 6, 7, 7, 7, 7, 8, 8, 8, 8,
+///         1, 2, 3, 4, 5, 6, 7, 8, 8, 7, 6, 5, 4, 3, 2, 1,
+///         1, 1, 1, 1, 2, 2, 2, 2, 3, 3, 3, 3, 4, 4, 4, 4,
+///         8, 7, 6, 5, 4, 3, 2, 1, 1, 2, 3, 4, 5, 6, 7, 8]),
+/// &[4, 4, 4],
+/// ).unwrap();
+/// assert_eq!(result, expected);
+///
+/// let x = Tensor::<i128>::new(
+///  Some(&[1, 2, 3, 4, 5, 6, 7, 8]),
+/// &[2, 4],
+/// ).unwrap();
+///
+/// let index = Tensor::<usize>::new(
+/// Some(&[0, 1]),
+/// &[2, 1],
+/// ).unwrap();
+/// let src = Tensor::<i128>::new(
+/// Some(&[9, 10]),
+/// &[2],
+/// ).unwrap();
+/// let result = scatter_nd(&x, &index, &src).unwrap();
+/// let expected = Tensor::<i128>::new(Some(&[9, 9, 9, 9, 10, 10, 10, 10]), &[2, 4]).unwrap();
+/// assert_eq!(result, expected);
+///
+/// let x = Tensor::<i128>::new(
+///  Some(&[1, 2, 3, 4, 5, 6, 7, 8]),
+/// &[2, 4],
+/// ).unwrap();
+///
+/// let index = Tensor::<usize>::new(
+/// Some(&[0, 1]),
+/// &[1, 1, 2],
+/// ).unwrap();
+/// let src = Tensor::<i128>::new(
+/// Some(&[9]),
+/// &[1, 1],
+/// ).unwrap();
+/// let result = scatter_nd(&x, &index, &src).unwrap();
+/// let expected = Tensor::<i128>::new(Some(&[1, 9, 3, 4, 5, 6, 7, 8]), &[2, 4]).unwrap();
+/// assert_eq!(result, expected);
+/// ````
+///
+pub fn scatter_nd<T: TensorType + Send + Sync>(
+    input: &Tensor<T>,
+    index: &Tensor<usize>,
+    src: &Tensor<T>,
+) -> Result<Tensor<T>, TensorError> {
+    // Calculate the output tensor size
+    let index_dims = index.dims().to_vec();
+    let input_dims = input.dims().to_vec();
+    let last_value = index_dims
+        .last()
+        .ok_or(TensorError::DimMismatch("scatter_nd".to_string()))?;
+    if last_value > &input_dims.len() {
+        return Err(TensorError::DimMismatch("scatter_nd".to_string()));
+    }
+
+    let mut output = input.clone();
+
+    let cartesian_coord = index_dims[0..index_dims.len() - 1]
+        .iter()
+        .map(|x| 0..*x)
+        .multi_cartesian_product()
+        .collect::<Vec<_>>();
+
+    cartesian_coord
+        .iter()
+        .map(|coord| {
+            let slice = coord.iter().map(|x| *x..*x + 1).collect::<Vec<_>>();
+            let index_val = index.get_slice(&slice)?;
+            let index_slice = index_val.iter().map(|x| *x..*x + 1).collect::<Vec<_>>();
+            let src_val = src.get_slice(&slice)?;
+            output.set_slice(&index_slice, &src_val)?;
+            Ok(())
+        })
+        .collect::<Result<Vec<_>, _>>()?;
+
+    Ok(output)
+}
+
 fn axes_op<T: TensorType + Send + Sync>(
     a: &Tensor<T>,
     axes: &[usize],

src/tensor/val.rs

@@ -4,6 +4,37 @@ use super::{
 };
 use halo2_proofs::{arithmetic::Field, plonk::Instance};
 
+pub(crate) fn create_constant_tensor<
+    F: PrimeField + TensorType + std::marker::Send + std::marker::Sync + PartialOrd,
+>(
+    val: F,
+    len: usize,
+) -> ValTensor<F> {
+    let mut constant = Tensor::from(vec![ValType::Constant(val); len].into_iter());
+    constant.set_visibility(&crate::graph::Visibility::Fixed);
+    ValTensor::from(constant)
+}
+
+pub(crate) fn create_unit_tensor<
+    F: PrimeField + TensorType + std::marker::Send + std::marker::Sync + PartialOrd,
+>(
+    len: usize,
+) -> ValTensor<F> {
+    let mut unit = Tensor::from(vec![ValType::Constant(F::ONE); len].into_iter());
+    unit.set_visibility(&crate::graph::Visibility::Fixed);
+    ValTensor::from(unit)
+}
+
+pub(crate) fn create_zero_tensor<
+    F: PrimeField + TensorType + std::marker::Send + std::marker::Sync + PartialOrd,
+>(
+    len: usize,
+) -> ValTensor<F> {
+    let mut zero = Tensor::from(vec![ValType::Constant(F::ZERO); len].into_iter());
+    zero.set_visibility(&crate::graph::Visibility::Fixed);
+    ValTensor::from(zero)
+}
+
 #[derive(Debug, Clone)]
 /// A [ValType] is a wrapper around Halo2 value(s).
 pub enum ValType<F: PrimeField + TensorType + std::marker::Send + std::marker::Sync + PartialOrd> {
@@ -318,6 +349,19 @@ impl<F: PrimeField + TensorType + PartialOrd> ValTensor<F> {
         matches!(self, ValTensor::Instance { .. })
     }
 
+    /// reverse order of elements whilst preserving the shape
+    pub fn reverse(&mut self) -> Result<(), Box<dyn Error>> {
+        match self {
+            ValTensor::Value { inner: v, .. } => {
+                v.reverse();
+            }
+            ValTensor::Instance { .. } => {
+                return Err(Box::new(TensorError::WrongMethod));
+            }
+        };
+        Ok(())
+    }
+
     ///
     pub fn set_initial_instance_offset(&mut self, offset: usize) {
         if let ValTensor::Instance { initial_offset, .. } = self {
@@ -450,7 +494,12 @@ impl<F: PrimeField + TensorType + PartialOrd> ValTensor<F> {
             }
             _ => return Err(Box::new(TensorError::WrongMethod)),
         };
-        Ok(integer_evals.into_iter().into())
+        let mut tensor: Tensor<i128> = integer_evals.into_iter().into();
+        match tensor.reshape(self.dims()) {
+            _ => {}
+        };
+
+        Ok(tensor)
     }
 
     /// Calls `pad_to_zero_rem` on the inner tensor.

src/wasm.rs

@@ -78,6 +78,17 @@ pub fn feltToBigEndian(array: wasm_bindgen::Clamped<Vec<u8>>) -> Result<String,
     Ok(format!("{:?}", felt))
 }
 
+/// Converts a felt to a little endian string
+#[wasm_bindgen]
+#[allow(non_snake_case)]
+pub fn feltToLittleEndian(array: wasm_bindgen::Clamped<Vec<u8>>) -> Result<String, JsError> {
+    let felt: Fr = serde_json::from_slice(&array[..])
+        .map_err(|e| JsError::new(&format!("Failed to deserialize field element: {}", e)))?;
+    let repr = serde_json::to_string(&felt).unwrap();
+    let b: String = serde_json::from_str(&repr).unwrap();
+    Ok(b)
+}
+
 /// Converts a hex string to a byte array
 #[wasm_bindgen]
 #[allow(non_snake_case)]

tests/integration_tests.rs

@@ -193,7 +193,7 @@ mod native_tests {
         "1l_tiny_div",
     ];
 
-    const TESTS: [&str; 77] = [
+    const TESTS: [&str; 79] = [
         "1l_mlp", //0
         "1l_slice",
         "1l_concat",
@@ -275,6 +275,8 @@ mod native_tests {
         "ltsf",
         "remainder", //75
         "bitshift",
+        "gather_nd",
+        "scatter_nd",
     ];
 
     const WASM_TESTS: [&str; 46] = [
@@ -502,7 +504,7 @@ mod native_tests {
             }
         });
 
-            seq!(N in 0..=76 {
+            seq!(N in 0..=78 {
 
             #(#[test_case(TESTS[N])])*
             #[ignore]
@@ -589,13 +591,16 @@ mod native_tests {
 
             #(#[test_case(TESTS[N])])*
             fn mock_large_batch_public_outputs_(test: &str) {
-                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);
-                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);
-                test_dir.close().unwrap();
+                // currently variable output rank is not supported in ONNX
+                if test != "gather_nd" {
+                    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);
+                    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);
+                    test_dir.close().unwrap();
+                }
             }
 
             #(#[test_case(TESTS[N])])*
@@ -1849,6 +1854,7 @@ mod native_tests {
                 &format!("{}/{}/key.pk", test_dir, example_name),
                 "--vk-path",
                 &format!("{}/{}/key.vk", test_dir, example_name),
+                "--disable-selector-compression",
             ])
             .status()
             .expect("failed to execute process");

tests/wasm.rs

@@ -9,8 +9,8 @@ mod wasm32 {
     use ezkl::pfsys;
     use ezkl::wasm::{
         bufferToVecOfFelt, compiledCircuitValidation, encodeVerifierCalldata, feltToBigEndian,
-        feltToFloat, feltToInt, genPk, genVk, genWitness, inputValidation, pkValidation,
-        poseidonHash, proofValidation, prove, settingsValidation, srsValidation,
+        feltToFloat, feltToInt, feltToLittleEndian, genPk, genVk, genWitness, inputValidation,
+        pkValidation, poseidonHash, proofValidation, prove, settingsValidation, srsValidation,
         u8_array_to_u128_le, verify, vkValidation, witnessValidation,
     };
     use halo2_solidity_verifier::encode_calldata;
@@ -89,9 +89,16 @@ mod wasm32 {
                     .unwrap();
             assert_eq!(integer, i as i128);
 
-            let hex_string = format!("{:?}", field_element);
-            let returned_string: String = feltToBigEndian(clamped).map_err(|_| "failed").unwrap();
+            let hex_string = format!("{:?}", field_element.clone());
+            let returned_string: String = feltToBigEndian(clamped.clone())
+                .map_err(|_| "failed")
+                .unwrap();
             assert_eq!(hex_string, returned_string);
+            let repr = serde_json::to_string(&field_element).unwrap();
+            let little_endian_string: String = serde_json::from_str(&repr).unwrap();
+            let returned_string: String =
+                feltToLittleEndian(clamped).map_err(|_| "failed").unwrap();
+            assert_eq!(little_endian_string, returned_string);
         }
     }
 

tests/wasm/settings.json

@@ -29,6 +29,8 @@
     "num_rows": 16,
     "total_dynamic_col_size": 0,
     "num_dynamic_lookups": 0,
+    "num_shuffles": 0,
+    "total_shuffle_col_size": 0,
     "total_assignments": 32,
     "total_const_size": 8,
     "model_instance_shapes": [