# tornadomeet/resnet architecture

> Architecture diagram of the GitHub repository tornadomeet/resnet, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-10-03). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/tornadomeet/resnet
- Repository: https://github.com/tornadomeet/resnet

## Overview

This repository reproduces ResNet-v2 training and provides single-image classification with trained checkpoints. The main workflow is prepared CIFAR-10 or ImageNet record files into training and validation, build the selected network, train/evaluate and save checkpoints, then optionally classify an image. It also supports continued training, memory-saving integration, and plotting accuracy curves. Training, model construction, prediction, and plotting are visible in the sampled sources; dataset preparation and reported model results are documented in the README, not implemented in the shown files.

Main entry points: [`train_resnet.py`](https://github.com/tornadomeet/resnet/blob/master/train_resnet.py) (Training runner), [`predict/predict.py`](https://github.com/tornadomeet/resnet/blob/master/predict/predict.py) (Image classifier), [`log/plot_curve.py`](https://github.com/tornadomeet/resnet/blob/master/log/plot_curve.py) (Curve plotter), [`predict/synset.txt`](https://github.com/tornadomeet/resnet/blob/master/predict/synset.txt) (Class labels)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_model["Network"]
  node_network["ResNet-v2 symbol<br/>[symbol_resnet.py]"]
end

subgraph group_training["Training"]
  node_records["Prepared records"]
  node_trainer["Training runner<br/>[train_resnet.py]"]
  node_record_iterators["Record iterators<br/>[train_resnet.py]"]
  node_checkpoint["Model checkpoints<br/>[train_resnet.py]"]
end

subgraph group_inference["Inference"]
  node_image["Input image"]
  node_synset["Class labels<br/>[synset.txt]"]
  node_predictor["Image classifier<br/>[predict.py]"]
  node_result["Top-1 / top-5"]
end

subgraph group_analysis["Analysis"]
  node_logs["Training logs"]
  node_curve["Curve plotter<br/>[plot_curve.py]"]
end

node_user(("Researcher"))
node_mxnet["MXNet runtime"]

node_user -->|"prepares"| node_records
node_records -->|"provides records"| node_record_iterators
node_trainer -->|"configures"| node_record_iterators
node_trainer -->|"builds"| node_network
node_trainer -->|"trains and evaluates"| node_mxnet
node_trainer -->|"saves / loads"| node_checkpoint
node_user -->|"starts training"| node_trainer
node_user -->|"selects"| node_image
node_user -->|"starts prediction"| node_predictor
node_checkpoint -->|"loads parameters"| node_predictor
node_image -->|"classifies"| node_predictor
node_synset -->|"maps classes"| node_predictor
node_predictor -->|"prints rankings"| node_result
node_trainer -.->|"produces training logs"| node_logs
node_user -->|"requests plot"| node_curve
node_logs -->|"parses"| node_curve

click node_trainer "https://github.com/tornadomeet/resnet/blob/master/train_resnet.py"
click node_record_iterators "https://github.com/tornadomeet/resnet/blob/master/train_resnet.py"
click node_checkpoint "https://github.com/tornadomeet/resnet/blob/master/train_resnet.py"
click node_network "https://github.com/tornadomeet/resnet/blob/master/symbol_resnet.py"
click node_synset "https://github.com/tornadomeet/resnet/blob/master/predict/synset.txt"
click node_predictor "https://github.com/tornadomeet/resnet/blob/master/predict/predict.py"
click node_curve "https://github.com/tornadomeet/resnet/blob/master/log/plot_curve.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_network toneBlue
class node_records,node_trainer,node_record_iterators,node_checkpoint toneAmber
class node_image,node_synset,node_predictor,node_result toneMint
class node_logs,node_curve toneRose
class node_user,node_mxnet toneIndigo
```

## Components

### Network

- [ResNet-v2 symbol](https://github.com/tornadomeet/resnet/blob/master/symbol_resnet.py): `symbol_resnet.py`

### Training

- **Prepared records**
- [Training runner](https://github.com/tornadomeet/resnet/blob/master/train_resnet.py): `train_resnet.py`
- [Record iterators](https://github.com/tornadomeet/resnet/blob/master/train_resnet.py): `train_resnet.py`
- [Model checkpoints](https://github.com/tornadomeet/resnet/blob/master/train_resnet.py): `train_resnet.py`

### Inference

- **Input image**
- [Class labels](https://github.com/tornadomeet/resnet/blob/master/predict/synset.txt): `predict/synset.txt`
- [Image classifier](https://github.com/tornadomeet/resnet/blob/master/predict/predict.py): `predict/predict.py`
- **Top-1 / top-5**

### Analysis

- **Training logs**
- [Curve plotter](https://github.com/tornadomeet/resnet/blob/master/log/plot_curve.py): `log/plot_curve.py`

### Other components

- **Researcher**
- **MXNet runtime**

## Connections

- Researcher → Prepared records: prepares (evidence: `README.md`)
- Prepared records → Record iterators: provides records (evidence: `train_resnet.py`)
- Training runner → Record iterators: configures (evidence: `train_resnet.py`)
- Training runner → ResNet-v2 symbol: builds (evidence: `train_resnet.py`)
- Training runner → MXNet runtime: trains and evaluates (evidence: `train_resnet.py`)
- Training runner → Model checkpoints: saves / loads (evidence: `train_resnet.py`)
- Researcher → Training runner: starts training (evidence: `README.md`)
- Researcher → Input image: selects
- Researcher → Image classifier: starts prediction (evidence: `predict/predict.py`)
- Model checkpoints → Image classifier: loads parameters (evidence: `predict/predict.py`)
- Input image → Image classifier: classifies (evidence: `predict/predict.py`)
- Class labels → Image classifier: maps classes (evidence: `predict/predict.py`)
- Image classifier → Top-1 / top-5: prints rankings (evidence: `predict/predict.py`)
- Training runner → Training logs: produces training logs (evidence: `README.md`)
- Researcher → Curve plotter: requests plot (evidence: `README.md`)
- Training logs → Curve plotter: parses (evidence: `log/plot_curve.py`)
