# graviraja/seq2seq architecture

> Architecture diagram of the GitHub repository graviraja/seq2seq, 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/graviraja/seq2seq
- Repository: https://github.com/graviraja/seq2seq

## Overview

This repository is a collection of standalone PyTorch sequence-model implementations, not a shared application. Its principal workflow is to train and evaluate a translation model, then use it for translation; the scripts implement alternative recurrent, attention-based, and Transformer approaches. BERT is a separate pretraining example. Training and model components are separated below only to show the calls visible in the excerpts; most behavior lives together in each script. The README documents all six capabilities, but does not establish shared orchestration between them.

Main entry points: [`simple_seq2seq.py`](https://github.com/graviraja/seq2seq/blob/master/simple_seq2seq.py) (Train and Evaluate), [`phrase_seq2seq.py`](https://github.com/graviraja/seq2seq/blob/master/phrase_seq2seq.py) (Train and Evaluate), [`attention_seq2seq.py`](https://github.com/graviraja/seq2seq/blob/master/attention_seq2seq.py) (Train and Evaluate), [`padded_seq2seq.py`](https://github.com/graviraja/seq2seq/blob/master/padded_seq2seq.py) (Train and Evaluate)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_recurrent["Recurrent Translation"]
  node_simple_model["Seq2Seq Model<br/>[simple_seq2seq.py]"]
  node_simple_training["Train and Evaluate<br/>[simple_seq2seq.py]"]
  node_phrase_model["Phrase Seq2Seq<br/>[phrase_seq2seq.py]"]
  node_phrase_training["Train and Evaluate<br/>[phrase_seq2seq.py]"]
end

subgraph group_attention["Attention Translation"]
  node_attention_model["Attention Seq2Seq"]
  node_attention_training["Train and Evaluate"]
  node_padded_model["Padded Attention<br/>[padded_seq2seq.py]"]
  node_padded_training["Train and Evaluate<br/>[padded_seq2seq.py]"]
end

subgraph group_transformers["Transformer Models"]
  node_transformer_model["Transformer Translation<br/>[transformer.py]"]
  node_transformer_training["Train and Evaluate<br/>[transformer.py]"]
  node_bert_model["BERT Pretraining Model<br/>[BERT_basic.py]"]
  node_bert_pretraining["Pretraining Example<br/>[BERT_basic.py]"]
end

node_simple_training -->|"calls model"| node_simple_model
node_phrase_training -->|"calls model"| node_phrase_model
node_attention_training -->|"calls model"| node_attention_model
node_padded_training -->|"calls model"| node_padded_model
node_transformer_training -->|"calls model"| node_transformer_model

click node_simple_model "https://github.com/graviraja/seq2seq/blob/master/simple_seq2seq.py"
click node_simple_training "https://github.com/graviraja/seq2seq/blob/master/simple_seq2seq.py"
click node_phrase_model "https://github.com/graviraja/seq2seq/blob/master/phrase_seq2seq.py"
click node_phrase_training "https://github.com/graviraja/seq2seq/blob/master/phrase_seq2seq.py"
click node_attention_model "https://github.com/graviraja/seq2seq/blob/master/attention_seq2seq.py"
click node_attention_training "https://github.com/graviraja/seq2seq/blob/master/attention_seq2seq.py"
click node_padded_model "https://github.com/graviraja/seq2seq/blob/master/padded_seq2seq.py"
click node_padded_training "https://github.com/graviraja/seq2seq/blob/master/padded_seq2seq.py"
click node_transformer_model "https://github.com/graviraja/seq2seq/blob/master/transformer.py"
click node_transformer_training "https://github.com/graviraja/seq2seq/blob/master/transformer.py"
click node_bert_model "https://github.com/graviraja/seq2seq/blob/master/BERT_basic.py"
click node_bert_pretraining "https://github.com/graviraja/seq2seq/blob/master/BERT_basic.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_simple_model,node_simple_training,node_phrase_model,node_phrase_training toneBlue
class node_attention_model,node_attention_training,node_padded_model,node_padded_training toneAmber
class node_transformer_model,node_transformer_training,node_bert_model,node_bert_pretraining toneMint
```

## Components

### Recurrent Translation

- [Seq2Seq Model](https://github.com/graviraja/seq2seq/blob/master/simple_seq2seq.py): `simple_seq2seq.py`
- [Train and Evaluate](https://github.com/graviraja/seq2seq/blob/master/simple_seq2seq.py): `simple_seq2seq.py`
- [Phrase Seq2Seq](https://github.com/graviraja/seq2seq/blob/master/phrase_seq2seq.py): `phrase_seq2seq.py`
- [Train and Evaluate](https://github.com/graviraja/seq2seq/blob/master/phrase_seq2seq.py): `phrase_seq2seq.py`

### Attention Translation

- [Attention Seq2Seq](https://github.com/graviraja/seq2seq/blob/master/attention_seq2seq.py): `attention_seq2seq.py`
- [Train and Evaluate](https://github.com/graviraja/seq2seq/blob/master/attention_seq2seq.py): `attention_seq2seq.py`
- [Padded Attention](https://github.com/graviraja/seq2seq/blob/master/padded_seq2seq.py): `padded_seq2seq.py`
- [Train and Evaluate](https://github.com/graviraja/seq2seq/blob/master/padded_seq2seq.py): `padded_seq2seq.py`

### Transformer Models

- [Transformer Translation](https://github.com/graviraja/seq2seq/blob/master/transformer.py): `transformer.py`
- [Train and Evaluate](https://github.com/graviraja/seq2seq/blob/master/transformer.py): `transformer.py`
- [BERT Pretraining Model](https://github.com/graviraja/seq2seq/blob/master/BERT_basic.py): `BERT_basic.py`
- [Pretraining Example](https://github.com/graviraja/seq2seq/blob/master/BERT_basic.py): `BERT_basic.py`

## Connections

- Train and Evaluate → Seq2Seq Model: calls model (evidence: `simple_seq2seq.py`)
- Train and Evaluate → Phrase Seq2Seq: calls model (evidence: `phrase_seq2seq.py`)
- Train and Evaluate → Attention Seq2Seq: calls model (evidence: `attention_seq2seq.py`)
- Train and Evaluate → Padded Attention: calls model (evidence: `padded_seq2seq.py`)
- Train and Evaluate → Transformer Translation: calls model (evidence: `transformer.py`)
