# y4h1ko/signal-inference-transformers-normalising-flows architecture

> Architecture diagram of the GitHub repository y4h1ko/signal-inference-transformers-normalising-flows, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-11-22). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/y4h1ko/signal-inference-transformers-normalising-flows
- Repository: https://github.com/y4h1ko/signal-inference-transformers-normalising-flows

## Overview

This research code generates synthetic sinusoidal observations and trains encoder-only Transformers to infer their frequencies. The principal workflow is configured data generation, tensor conversion and splitting, model training, test evaluation, then plots and posterior diagnostics. It covers single-frequency regression and conditional-flow inference, plus linear and nonlinear two-frequency mixtures with permutation ambiguity. The graph also preserves positional encoding, experiment-analysis utilities, and plotting as distinct internal capabilities. Training and several model details are represented from README and call sites; their source excerpts were not provided, so those links are limited to visible calls or documentation.

Main entry points: [`main.py`](https://github.com/y4h1ko/signal-inference-transformers-normalising-flows/blob/main/main.py) (Experiment runner), [`code_files/models.py`](https://github.com/y4h1ko/signal-inference-transformers-normalising-flows/blob/main/code_files/models.py) (Transformer encoders), [`code_files/imports_and_libraries.py`](https://github.com/y4h1ko/signal-inference-transformers-normalising-flows/blob/main/code_files/imports_and_libraries.py) (Shared runtime imports), [`code_files/dataset_creation.py`](https://github.com/y4h1ko/signal-inference-transformers-normalising-flows/blob/main/code_files/dataset_creation.py) (Signal datasets)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_workflow["Inference workflow"]
  node_runner["Experiment runner<br/>[main.py]"]
  node_signal_data["Signal datasets"]
  node_tensor_splits["Tensor conversion and splits"]
  node_training["Training loops<br/>[train_and_test.py]"]
  node_evaluation["Test evaluation<br/>[train_and_test.py]"]
end

subgraph group_modeling["Modeling"]
  node_transformer["Transformer encoders<br/>[models.py]"]
  node_position["Positional encoding"]
  node_regression["Deterministic regression<br/>[models.py]"]
  node_flow_single["Single-frequency flow<br/>[models.py]"]
  node_flow_pair["Two-frequency flow<br/>[models.py]"]
end

subgraph group_analysis["Evaluation and analysis"]
  node_plots["Plots and diagnostics<br/>[visualizations.py]"]
  node_sweeps["Experiment sweeps<br/>[analysis.py]"]
end

subgraph group_support["Configuration"]
  node_config["Experiment settings<br/>[configuration.py]"]
  node_shared_imports["Shared runtime imports"]
end

node_researcher(("Researcher"))

node_researcher -->|"runs experiments"| node_runner
node_runner -->|"generates signals"| node_signal_data
node_runner -->|"converts and splits"| node_tensor_splits
node_runner -->|"trains models"| node_training
node_runner -->|"evaluates models"| node_evaluation
node_runner -->|"plots results"| node_plots
node_runner -->|"selects regression"| node_regression
node_runner -->|"selects flow model"| node_flow_single
node_runner -->|"selects pair model"| node_flow_pair
node_transformer -->|"adds positions"| node_position
node_regression -->|"uses encoder"| node_transformer
node_flow_single -->|"conditions on encoder"| node_transformer
node_flow_pair -->|"conditions on encoder"| node_transformer
node_runner -.->|"analyzes settings"| node_sweeps
node_shared_imports -->|"loads settings"| node_config

click node_runner "https://github.com/y4h1ko/signal-inference-transformers-normalising-flows/blob/main/main.py"
click node_signal_data "https://github.com/y4h1ko/signal-inference-transformers-normalising-flows/blob/main/code_files/dataset_creation.py"
click node_tensor_splits "https://github.com/y4h1ko/signal-inference-transformers-normalising-flows/blob/main/code_files/dataset_creation.py"
click node_training "https://github.com/y4h1ko/signal-inference-transformers-normalising-flows/blob/main/code_files/train_and_test.py"
click node_evaluation "https://github.com/y4h1ko/signal-inference-transformers-normalising-flows/blob/main/code_files/train_and_test.py"
click node_transformer "https://github.com/y4h1ko/signal-inference-transformers-normalising-flows/blob/main/code_files/models.py"
click node_position "https://github.com/y4h1ko/signal-inference-transformers-normalising-flows/blob/main/code_files/positional_encodings.py"
click node_regression "https://github.com/y4h1ko/signal-inference-transformers-normalising-flows/blob/main/code_files/models.py"
click node_flow_single "https://github.com/y4h1ko/signal-inference-transformers-normalising-flows/blob/main/code_files/models.py"
click node_flow_pair "https://github.com/y4h1ko/signal-inference-transformers-normalising-flows/blob/main/code_files/models.py"
click node_plots "https://github.com/y4h1ko/signal-inference-transformers-normalising-flows/blob/main/code_files/visualizations.py"
click node_sweeps "https://github.com/y4h1ko/signal-inference-transformers-normalising-flows/blob/main/archive/analysis.py"
click node_config "https://github.com/y4h1ko/signal-inference-transformers-normalising-flows/blob/main/code_files/configuration.py"
click node_shared_imports "https://github.com/y4h1ko/signal-inference-transformers-normalising-flows/blob/main/code_files/imports_and_libraries.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_runner,node_signal_data,node_tensor_splits,node_training,node_evaluation toneBlue
class node_transformer,node_position,node_regression,node_flow_single,node_flow_pair toneAmber
class node_plots,node_sweeps toneMint
class node_config,node_shared_imports toneRose
class node_researcher toneIndigo
```

## Components

### Inference workflow

- [Experiment runner](https://github.com/y4h1ko/signal-inference-transformers-normalising-flows/blob/main/main.py): `main.py`
- [Signal datasets](https://github.com/y4h1ko/signal-inference-transformers-normalising-flows/blob/main/code_files/dataset_creation.py): `code_files/dataset_creation.py`
- [Tensor conversion and splits](https://github.com/y4h1ko/signal-inference-transformers-normalising-flows/blob/main/code_files/dataset_creation.py): `code_files/dataset_creation.py`
- [Training loops](https://github.com/y4h1ko/signal-inference-transformers-normalising-flows/blob/main/code_files/train_and_test.py): `code_files/train_and_test.py`
- [Test evaluation](https://github.com/y4h1ko/signal-inference-transformers-normalising-flows/blob/main/code_files/train_and_test.py): `code_files/train_and_test.py`

### Modeling

- [Transformer encoders](https://github.com/y4h1ko/signal-inference-transformers-normalising-flows/blob/main/code_files/models.py): `code_files/models.py`
- [Positional encoding](https://github.com/y4h1ko/signal-inference-transformers-normalising-flows/blob/main/code_files/positional_encodings.py): `code_files/positional_encodings.py`
- [Deterministic regression](https://github.com/y4h1ko/signal-inference-transformers-normalising-flows/blob/main/code_files/models.py): `code_files/models.py`
- [Single-frequency flow](https://github.com/y4h1ko/signal-inference-transformers-normalising-flows/blob/main/code_files/models.py): `code_files/models.py`
- [Two-frequency flow](https://github.com/y4h1ko/signal-inference-transformers-normalising-flows/blob/main/code_files/models.py): `code_files/models.py`

### Evaluation and analysis

- [Plots and diagnostics](https://github.com/y4h1ko/signal-inference-transformers-normalising-flows/blob/main/code_files/visualizations.py): `code_files/visualizations.py`
- [Experiment sweeps](https://github.com/y4h1ko/signal-inference-transformers-normalising-flows/blob/main/archive/analysis.py): `archive/analysis.py`

### Configuration

- [Experiment settings](https://github.com/y4h1ko/signal-inference-transformers-normalising-flows/blob/main/code_files/configuration.py): `code_files/configuration.py`
- [Shared runtime imports](https://github.com/y4h1ko/signal-inference-transformers-normalising-flows/blob/main/code_files/imports_and_libraries.py): `code_files/imports_and_libraries.py`

### Other components

- **Researcher**

## Connections

- Researcher → Experiment runner: runs experiments (evidence: `README.md`)
- Experiment runner → Signal datasets: generates signals (evidence: `main.py`)
- Experiment runner → Tensor conversion and splits: converts and splits (evidence: `main.py`)
- Experiment runner → Training loops: trains models (evidence: `main.py`)
- Experiment runner → Test evaluation: evaluates models (evidence: `main.py`)
- Experiment runner → Plots and diagnostics: plots results (evidence: `main.py`)
- Experiment runner → Deterministic regression: selects regression (evidence: `main.py`)
- Experiment runner → Single-frequency flow: selects flow model (evidence: `main.py`)
- Experiment runner → Two-frequency flow: selects pair model (evidence: `main.py`)
- Transformer encoders → Positional encoding: adds positions (evidence: `code_files/models.py`)
- Deterministic regression → Transformer encoders: uses encoder (evidence: `README.md`)
- Single-frequency flow → Transformer encoders: conditions on encoder (evidence: `README.md`)
- Two-frequency flow → Transformer encoders: conditions on encoder (evidence: `README.md`)
- Experiment runner → Experiment sweeps: analyzes settings
- Shared runtime imports → Experiment settings: loads settings (evidence: `code_files/imports_and_libraries.py`)
