# saashajoshi/piqture architecture

> Architecture diagram of the GitHub repository saashajoshi/piqture, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-10-01). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/saashajoshi/piqture
- Repository: https://github.com/saashajoshi/piqture

## Overview

piQture is a Python quantum machine-learning toolkit for image processing. Its documented user workflow loads and transforms MNIST images, encodes an image into a Qiskit circuit, and lets users compose QCNN layers into a model circuit. Other major branches provide several image encodings, tensor-network circuits, and variational classifiers. The graph keeps those capabilities distinct; unsampled implementations are represented only at paths supported by the repository tree, and edges are limited to documented workflow steps or references shown in the source index.

Main entry points: [`piqture/neural_networks/qcnn.py`](https://github.com/saashajoshi/piqture/blob/main/piqture/neural_networks/qcnn.py) (QCNN composer), [`piqture/gates/two_qubit_unitary.py`](https://github.com/saashajoshi/piqture/blob/main/piqture/gates/two_qubit_unitary.py) (Two-qubit unitaries), [`piqture/neural_networks/layers/base_layer.py`](https://github.com/saashajoshi/piqture/blob/main/piqture/neural_networks/layers/base_layer.py) (Layer base), [`piqture/embeddings/image_embeddings`](https://github.com/saashajoshi/piqture/tree/main/piqture/embeddings/image_embeddings) (Image encodings)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_data["Image data"]
  node_mnist["MNIST loader"]
  node_transforms["Image transforms<br/>[transforms.py]"]
end

subgraph group_embedding["Image encoding"]
  node_embedding_base["Embedding base<br/>[image_embedding.py]"]
  node_angle["Angle encoding<br/>[angle_encoding.py]"]
  node_amplitude["Amplitude encoding"]
  node_image_encodings["Image encodings"]
  node_mixin["Image embedding mixin"]
end

subgraph group_models["Quantum models"]
  node_qcnn["QCNN composer<br/>[qcnn.py]"]
  node_qnn_base["QNN base"]
  node_layer_base["Layer base<br/>[base_layer.py]"]
  node_conv_layer["Convolution layer"]
  node_pool_layer["Pooling layers<br/>[pooling_layer.py]"]
  node_fc_layer["Fully connected layer"]
  node_autoencoder["Quantum autoencoder"]
  node_classifiers["Variational classifiers"]
end

subgraph group_tensor["Tensor networks"]
  node_base_tensor["Tensor network base"]
  node_mera["MERA<br/>[mera.py]"]
  node_mps["MPS<br/>[mps.py]"]
  node_peps["PEPS<br/>[peps.py]"]
  node_ttn["TTN<br/>[ttn.py]"]
end

subgraph group_circuits["Circuit foundations"]
  node_unitary["Two-qubit unitaries"]
  node_unitary_block["Unitary block interface<br/>[unitary_block.py]"]
end

node_user(("User"))

node_user -->|"loads images"| node_mnist
node_mnist -->|"uses transforms"| node_transforms
node_user -->|"encodes image"| node_image_encodings
node_image_encodings -->|"extends base"| node_embedding_base
node_image_encodings -->|"uses mixin"| node_mixin
node_angle -->|"uses base"| node_embedding_base
node_user -->|"composes model"| node_qcnn
node_qcnn -->|"initializes"| node_qnn_base
node_qcnn -->|"validates layers"| node_layer_base
node_qcnn -.->|"builds layer"| node_conv_layer
node_qcnn -.->|"builds layer"| node_pool_layer
node_qcnn -.->|"builds layer"| node_fc_layer
node_conv_layer -->|"inherits from"| node_layer_base
node_pool_layer -->|"inherits from"| node_layer_base
node_fc_layer -->|"inherits from"| node_layer_base
node_conv_layer -->|"uses MERA"| node_mera
node_user -.->|"selects classifier"| node_classifiers
node_user -.->|"selects model"| node_autoencoder
node_mera -->|"initializes base"| node_base_tensor
node_mps -->|"uses base"| node_base_tensor
node_ttn -->|"initializes base"| node_base_tensor
node_mera -->|"uses gates"| node_unitary
node_mps -->|"uses gates"| node_unitary
node_ttn -->|"uses gates"| node_unitary
node_unitary -->|"implements interface"| node_unitary_block

click node_mnist "https://github.com/saashajoshi/piqture/blob/main/piqture/data_loader/mnist_data_loader.py"
click node_transforms "https://github.com/saashajoshi/piqture/blob/main/piqture/transforms/transforms.py"
click node_embedding_base "https://github.com/saashajoshi/piqture/blob/main/piqture/embeddings/image_embedding.py"
click node_angle "https://github.com/saashajoshi/piqture/blob/main/piqture/embeddings/angle_encoding.py"
click node_amplitude "https://github.com/saashajoshi/piqture/blob/main/piqture/embeddings/amplitude_encoder.py"
click node_image_encodings "https://github.com/saashajoshi/piqture/tree/main/piqture/embeddings/image_embeddings"
click node_mixin "https://github.com/saashajoshi/piqture/blob/main/piqture/mixin/image_embedding_mixin.py"
click node_qcnn "https://github.com/saashajoshi/piqture/blob/main/piqture/neural_networks/qcnn.py"
click node_qnn_base "https://github.com/saashajoshi/piqture/blob/main/piqture/neural_networks/quantum_neural_network.py"
click node_layer_base "https://github.com/saashajoshi/piqture/blob/main/piqture/neural_networks/layers/base_layer.py"
click node_conv_layer "https://github.com/saashajoshi/piqture/blob/main/piqture/neural_networks/layers/convolutional_layer.py"
click node_pool_layer "https://github.com/saashajoshi/piqture/blob/main/piqture/neural_networks/layers/pooling_layer.py"
click node_fc_layer "https://github.com/saashajoshi/piqture/blob/main/piqture/neural_networks/layers/fully_connected_layer.py"
click node_autoencoder "https://github.com/saashajoshi/piqture/blob/main/piqture/neural_networks/quantum_autoencoder.py"
click node_classifiers "https://github.com/saashajoshi/piqture/tree/main/piqture/variational_classifiers"
click node_base_tensor "https://github.com/saashajoshi/piqture/blob/main/piqture/tensor_networks/base_tensor_network.py"
click node_mera "https://github.com/saashajoshi/piqture/blob/main/piqture/tensor_networks/mera.py"
click node_mps "https://github.com/saashajoshi/piqture/blob/main/piqture/tensor_networks/mps.py"
click node_peps "https://github.com/saashajoshi/piqture/blob/main/piqture/tensor_networks/peps.py"
click node_ttn "https://github.com/saashajoshi/piqture/blob/main/piqture/tensor_networks/ttn.py"
click node_unitary "https://github.com/saashajoshi/piqture/blob/main/piqture/gates/two_qubit_unitary.py"
click node_unitary_block "https://github.com/saashajoshi/piqture/blob/main/piqture/gates/unitary_block.py"

classDef toneNeutral fill:#f8fafc,stroke:#334155,stroke-width:1.5px,color:#0f172a
classDef toneBlue fill:#dbeafe,stroke:#2563eb,stroke-width:1.5px,color:#172554
classDef toneAmber fill:#fef3c7,stroke:#d97706,stroke-width:1.5px,color:#78350f
classDef toneMint fill:#dcfce7,stroke:#16a34a,stroke-width:1.5px,color:#14532d
classDef toneRose fill:#ffe4e6,stroke:#e11d48,stroke-width:1.5px,color:#881337
classDef toneIndigo fill:#e0e7ff,stroke:#4f46e5,stroke-width:1.5px,color:#312e81
classDef toneTeal fill:#ccfbf1,stroke:#0f766e,stroke-width:1.5px,color:#134e4a
class node_mnist,node_transforms,node_user toneBlue
class node_embedding_base,node_angle,node_amplitude,node_image_encodings,node_mixin toneAmber
class node_qcnn,node_qnn_base,node_layer_base,node_conv_layer,node_pool_layer,node_fc_layer,node_autoencoder,node_classifiers toneMint
class node_base_tensor,node_mera,node_mps,node_peps,node_ttn toneRose
class node_unitary,node_unitary_block toneIndigo
```

## Components

### Image data

- [MNIST loader](https://github.com/saashajoshi/piqture/blob/main/piqture/data_loader/mnist_data_loader.py): `piqture/data_loader/mnist_data_loader.py`
- [Image transforms](https://github.com/saashajoshi/piqture/blob/main/piqture/transforms/transforms.py): `piqture/transforms/transforms.py`

### Image encoding

- [Embedding base](https://github.com/saashajoshi/piqture/blob/main/piqture/embeddings/image_embedding.py): `piqture/embeddings/image_embedding.py`
- [Angle encoding](https://github.com/saashajoshi/piqture/blob/main/piqture/embeddings/angle_encoding.py): `piqture/embeddings/angle_encoding.py`
- [Amplitude encoding](https://github.com/saashajoshi/piqture/blob/main/piqture/embeddings/amplitude_encoder.py): `piqture/embeddings/amplitude_encoder.py`
- [Image encodings](https://github.com/saashajoshi/piqture/tree/main/piqture/embeddings/image_embeddings): `piqture/embeddings/image_embeddings`
- [Image embedding mixin](https://github.com/saashajoshi/piqture/blob/main/piqture/mixin/image_embedding_mixin.py): `piqture/mixin/image_embedding_mixin.py`

### Quantum models

- [QCNN composer](https://github.com/saashajoshi/piqture/blob/main/piqture/neural_networks/qcnn.py): `piqture/neural_networks/qcnn.py`
- [QNN base](https://github.com/saashajoshi/piqture/blob/main/piqture/neural_networks/quantum_neural_network.py): `piqture/neural_networks/quantum_neural_network.py`
- [Layer base](https://github.com/saashajoshi/piqture/blob/main/piqture/neural_networks/layers/base_layer.py): `piqture/neural_networks/layers/base_layer.py`
- [Convolution layer](https://github.com/saashajoshi/piqture/blob/main/piqture/neural_networks/layers/convolutional_layer.py): `piqture/neural_networks/layers/convolutional_layer.py`
- [Pooling layers](https://github.com/saashajoshi/piqture/blob/main/piqture/neural_networks/layers/pooling_layer.py): `piqture/neural_networks/layers/pooling_layer.py`
- [Fully connected layer](https://github.com/saashajoshi/piqture/blob/main/piqture/neural_networks/layers/fully_connected_layer.py): `piqture/neural_networks/layers/fully_connected_layer.py`
- [Quantum autoencoder](https://github.com/saashajoshi/piqture/blob/main/piqture/neural_networks/quantum_autoencoder.py): `piqture/neural_networks/quantum_autoencoder.py`
- [Variational classifiers](https://github.com/saashajoshi/piqture/tree/main/piqture/variational_classifiers): `piqture/variational_classifiers`

### Tensor networks

- [Tensor network base](https://github.com/saashajoshi/piqture/blob/main/piqture/tensor_networks/base_tensor_network.py): `piqture/tensor_networks/base_tensor_network.py`
- [MERA](https://github.com/saashajoshi/piqture/blob/main/piqture/tensor_networks/mera.py): `piqture/tensor_networks/mera.py`
- [MPS](https://github.com/saashajoshi/piqture/blob/main/piqture/tensor_networks/mps.py): `piqture/tensor_networks/mps.py`
- [PEPS](https://github.com/saashajoshi/piqture/blob/main/piqture/tensor_networks/peps.py): `piqture/tensor_networks/peps.py`
- [TTN](https://github.com/saashajoshi/piqture/blob/main/piqture/tensor_networks/ttn.py): `piqture/tensor_networks/ttn.py`

### Circuit foundations

- [Two-qubit unitaries](https://github.com/saashajoshi/piqture/blob/main/piqture/gates/two_qubit_unitary.py): `piqture/gates/two_qubit_unitary.py`
- [Unitary block interface](https://github.com/saashajoshi/piqture/blob/main/piqture/gates/unitary_block.py): `piqture/gates/unitary_block.py`

### Other components

- **User**

## Connections

- User → MNIST loader: loads images (evidence: `README.md`)
- MNIST loader → Image transforms: uses transforms (evidence: `piqture/data_loader/mnist_data_loader.py`)
- User → Image encodings: encodes image (evidence: `README.md`)
- Image encodings → Embedding base: extends base (evidence: `piqture/embeddings/image_embeddings/neqr.py`)
- Image encodings → Image embedding mixin: uses mixin (evidence: `piqture/embeddings/image_embeddings/neqr.py`)
- Angle encoding → Embedding base: uses base (evidence: `piqture/embeddings/angle_encoding.py`)
- User → QCNN composer: composes model (evidence: `README.md`)
- QCNN composer → QNN base: initializes (evidence: `piqture/neural_networks/qcnn.py`)
- QCNN composer → Layer base: validates layers (evidence: `piqture/neural_networks/qcnn.py`)
- QCNN composer → Convolution layer: builds layer
- QCNN composer → Pooling layers: builds layer
- QCNN composer → Fully connected layer: builds layer
- Convolution layer → Layer base: inherits from (evidence: `piqture/neural_networks/layers/convolutional_layer.py`)
- Pooling layers → Layer base: inherits from (evidence: `piqture/neural_networks/layers/pooling_layer.py`)
- Fully connected layer → Layer base: inherits from (evidence: `piqture/neural_networks/layers/fully_connected_layer.py`)
- Convolution layer → MERA: uses MERA (evidence: `piqture/neural_networks/layers/convolutional_layer.py`)
- User → Variational classifiers: selects classifier
- User → Quantum autoencoder: selects model
- MERA → Tensor network base: initializes base (evidence: `piqture/tensor_networks/mera.py`)
- MPS → Tensor network base: uses base (evidence: `piqture/tensor_networks/mps.py`)
- TTN → Tensor network base: initializes base (evidence: `piqture/tensor_networks/ttn.py`)
- MERA → Two-qubit unitaries: uses gates (evidence: `piqture/tensor_networks/mera.py`)
- MPS → Two-qubit unitaries: uses gates (evidence: `piqture/tensor_networks/mps.py`)
- TTN → Two-qubit unitaries: uses gates (evidence: `piqture/tensor_networks/ttn.py`)
- Two-qubit unitaries → Unitary block interface: implements interface (evidence: `piqture/gates/two_qubit_unitary.py`)
