# biba10/generative-cross-lingual-absa architecture

> Architecture diagram of the GitHub repository biba10/generative-cross-lingual-absa, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-10-05). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/biba10/generative-cross-lingual-absa
- Repository: https://github.com/biba10/generative-cross-lingual-absa

## Overview

This repository implements multilingual aspect-based sentiment analysis (ABSA): reviews are loaded and formatted for a selected task, then analyzed by a generative sequence-to-sequence model or by prompted language models; predictions are evaluated with task-specific F1 metrics. Its principal workflow is training and testing the generative model, with separate open-source LLM and ChatGPT prompting workflows and an instruction-tuning entry point. Relationships for unsampled modules are included only where the README or SOURCE INDEX supports them; the source excerpts do not establish every runtime call.

Main entry points: [`main.py`](https://github.com/biba10/generative-cross-lingual-absa/blob/main/main.py) (Train and test), [`src/data_utils/data_loader.py`](https://github.com/biba10/generative-cross-lingual-absa/blob/main/src/data_utils/data_loader.py) (Seq2seq data loader), [`src/llm_prompting/llm_api_classifier.py`](https://github.com/biba10/generative-cross-lingual-absa/blob/main/src/llm_prompting/llm_api_classifier.py) (ChatGPT classifier), [`main_llm.py`](https://github.com/biba10/generative-cross-lingual-absa/blob/main/main_llm.py) (Open LLM run)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_workflow["Application workflows"]
  node_main_train["Train and test<br/>[main.py]"]
  node_main_llm["Open LLM run<br/>[main_llm.py]"]
  node_main_api["ChatGPT run<br/>[main_prompting.py]"]
  node_instruction["Instruction tuning"]
end

subgraph group_data["Data preparation"]
  node_seq_loader["Seq2seq data loader<br/>[data_loader.py]"]
  node_seq_dataset["Seq2seq dataset<br/>[dataset_seq2seq.py]"]
  node_data_utils["Data preprocessing<br/>[data_utils.py]"]
  node_llm_dataset["LLM dataset<br/>[llm_dataset.py]"]
  node_api_dataset["API dataset<br/>[llm_api_dataset.py]"]
end

subgraph group_models["Models and prompting"]
  node_tokenizer["Model and tokenizer loader"]
  node_generative["Generative ABSA model"]
  node_llm_classifier["Open LLM classifier<br/>[llm_classifier.py]"]
  node_api_classifier["ChatGPT classifier"]
  node_llm_access["LLM API access<br/>[llm_access.py]"]
  node_templates["Prompt templates"]
  node_language_prompts["Language few-shot prompts"]
end

subgraph group_evaluation["Evaluation and metrics"]
  node_evaluation["Seq2seq evaluation<br/>[evaluation.py]"]
  node_seq_f1["Seq2seq F1"]
  node_total_f1["Combined task F1<br/>[total_f1_score.py]"]
  node_llm_evaluation["Prompted-model evaluation<br/>[llm_evaluation.py]"]
  node_f1_general["General F1<br/>[f1_score.py]"]
end

subgraph group_support["Configuration and logging"]
  node_args["CLI arguments<br/>[args_utils.py]"]
  node_config["Task and language config<br/>[config.py]"]
  node_logger["Experiment logging<br/>[logger_utils.py]"]
end

node_user(("Experimenter"))
node_chat_service{{"ChatGPT service"}}

node_user -->|"runs training"| node_main_train
node_user -->|"runs prompting"| node_main_llm
node_user -->|"runs prompting"| node_main_api
node_user -->|"runs tuning"| node_instruction
node_main_train -->|"parses arguments"| node_args
node_main_train -->|"loads model"| node_tokenizer
node_main_train -->|"creates loader"| node_seq_loader
node_main_train -->|"fits and tests"| node_generative
node_main_train -->|"configures logging"| node_logger
node_main_llm -->|"builds data loader"| node_seq_loader
node_main_llm -->|"runs classifier"| node_llm_classifier
node_main_llm -->|"loads tokenizer"| node_tokenizer
node_main_api -->|"runs classifier"| node_api_classifier
node_main_api -->|"builds data loader"| node_seq_loader
node_instruction -->|"uses training data"| node_llm_dataset
node_instruction -->|"uses classifier"| node_llm_classifier
node_instruction -->|"parses arguments"| node_args
node_seq_loader -->|"creates datasets"| node_seq_dataset
node_seq_loader -->|"creates datasets"| node_llm_dataset
node_seq_loader -->|"creates datasets"| node_api_dataset
node_llm_dataset -->|"encodes prompts"| node_data_utils
node_llm_dataset -->|"gets instructions"| node_templates
node_templates -->|"selects templates"| node_language_prompts
node_tokenizer -->|"constructs model"| node_generative
node_generative -->|"evaluates predictions"| node_evaluation
node_evaluation -->|"updates metrics"| node_seq_f1
node_evaluation -->|"updates aggregate"| node_total_f1
node_seq_f1 -->|"extends metric"| node_f1_general
node_total_f1 -->|"uses metric"| node_f1_general
node_llm_classifier -->|"scores predictions"| node_seq_f1
node_api_classifier -->|"requests completions"| node_llm_access
node_api_classifier -->|"builds prompts"| node_templates
node_api_classifier -->|"evaluates predictions"| node_llm_evaluation
node_api_classifier -.->|"sends requests"| node_chat_service

click node_main_train "https://github.com/biba10/generative-cross-lingual-absa/blob/main/main.py"
click node_main_llm "https://github.com/biba10/generative-cross-lingual-absa/blob/main/main_llm.py"
click node_main_api "https://github.com/biba10/generative-cross-lingual-absa/blob/main/main_prompting.py"
click node_instruction "https://github.com/biba10/generative-cross-lingual-absa/blob/main/instruction_tuning.py"
click node_args "https://github.com/biba10/generative-cross-lingual-absa/blob/main/src/utils/args_utils.py"
click node_config "https://github.com/biba10/generative-cross-lingual-absa/blob/main/src/utils/config.py"
click node_logger "https://github.com/biba10/generative-cross-lingual-absa/blob/main/src/utils/logger_utils.py"
click node_seq_loader "https://github.com/biba10/generative-cross-lingual-absa/blob/main/src/data_utils/data_loader.py"
click node_seq_dataset "https://github.com/biba10/generative-cross-lingual-absa/blob/main/src/data_utils/dataset_seq2seq.py"
click node_data_utils "https://github.com/biba10/generative-cross-lingual-absa/blob/main/src/data_utils/data_utils.py"
click node_llm_dataset "https://github.com/biba10/generative-cross-lingual-absa/blob/main/src/data_utils/llm_dataset.py"
click node_api_dataset "https://github.com/biba10/generative-cross-lingual-absa/blob/main/src/data_utils/llm_api_dataset.py"
click node_tokenizer "https://github.com/biba10/generative-cross-lingual-absa/blob/main/src/model/model_tokenizer_loader.py"
click node_generative "https://github.com/biba10/generative-cross-lingual-absa/blob/main/src/model/model_generative.py"
click node_llm_classifier "https://github.com/biba10/generative-cross-lingual-absa/blob/main/src/llm_prompting/llm_classifier.py"
click node_api_classifier "https://github.com/biba10/generative-cross-lingual-absa/blob/main/src/llm_prompting/llm_api_classifier.py"
click node_llm_access "https://github.com/biba10/generative-cross-lingual-absa/blob/main/src/llm_prompting/llm_access.py"
click node_templates "https://github.com/biba10/generative-cross-lingual-absa/blob/main/src/llm_prompting/templates/prompt_templates.py"
click node_language_prompts "https://github.com/biba10/generative-cross-lingual-absa/tree/main/src/llm_prompting/templates"
click node_evaluation "https://github.com/biba10/generative-cross-lingual-absa/blob/main/src/evaluation/evaluation.py"
click node_seq_f1 "https://github.com/biba10/generative-cross-lingual-absa/blob/main/src/evaluation/f1_score_seq2seq.py"
click node_total_f1 "https://github.com/biba10/generative-cross-lingual-absa/blob/main/src/evaluation/total_f1_score.py"
click node_llm_evaluation "https://github.com/biba10/generative-cross-lingual-absa/blob/main/src/evaluation/llm_evaluation.py"
click node_f1_general "https://github.com/biba10/generative-cross-lingual-absa/blob/main/src/evaluation/f1_score.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_main_train,node_main_llm,node_main_api,node_instruction toneBlue
class node_seq_loader,node_seq_dataset,node_data_utils,node_llm_dataset,node_api_dataset toneAmber
class node_tokenizer,node_generative,node_llm_classifier,node_api_classifier,node_llm_access,node_templates,node_language_prompts toneMint
class node_evaluation,node_seq_f1,node_total_f1,node_llm_evaluation,node_f1_general toneRose
class node_args,node_config,node_logger,node_user,node_chat_service toneIndigo
```

## Components

### Application workflows

- [Train and test](https://github.com/biba10/generative-cross-lingual-absa/blob/main/main.py): `main.py`
- [Open LLM run](https://github.com/biba10/generative-cross-lingual-absa/blob/main/main_llm.py): `main_llm.py`
- [ChatGPT run](https://github.com/biba10/generative-cross-lingual-absa/blob/main/main_prompting.py): `main_prompting.py`
- [Instruction tuning](https://github.com/biba10/generative-cross-lingual-absa/blob/main/instruction_tuning.py): `instruction_tuning.py`

### Data preparation

- [Seq2seq data loader](https://github.com/biba10/generative-cross-lingual-absa/blob/main/src/data_utils/data_loader.py): `src/data_utils/data_loader.py`
- [Seq2seq dataset](https://github.com/biba10/generative-cross-lingual-absa/blob/main/src/data_utils/dataset_seq2seq.py): `src/data_utils/dataset_seq2seq.py`
- [Data preprocessing](https://github.com/biba10/generative-cross-lingual-absa/blob/main/src/data_utils/data_utils.py): `src/data_utils/data_utils.py`
- [LLM dataset](https://github.com/biba10/generative-cross-lingual-absa/blob/main/src/data_utils/llm_dataset.py): `src/data_utils/llm_dataset.py`
- [API dataset](https://github.com/biba10/generative-cross-lingual-absa/blob/main/src/data_utils/llm_api_dataset.py): `src/data_utils/llm_api_dataset.py`

### Models and prompting

- [Model and tokenizer loader](https://github.com/biba10/generative-cross-lingual-absa/blob/main/src/model/model_tokenizer_loader.py): `src/model/model_tokenizer_loader.py`
- [Generative ABSA model](https://github.com/biba10/generative-cross-lingual-absa/blob/main/src/model/model_generative.py): `src/model/model_generative.py`
- [Open LLM classifier](https://github.com/biba10/generative-cross-lingual-absa/blob/main/src/llm_prompting/llm_classifier.py): `src/llm_prompting/llm_classifier.py`
- [ChatGPT classifier](https://github.com/biba10/generative-cross-lingual-absa/blob/main/src/llm_prompting/llm_api_classifier.py): `src/llm_prompting/llm_api_classifier.py`
- [LLM API access](https://github.com/biba10/generative-cross-lingual-absa/blob/main/src/llm_prompting/llm_access.py): `src/llm_prompting/llm_access.py`
- [Prompt templates](https://github.com/biba10/generative-cross-lingual-absa/blob/main/src/llm_prompting/templates/prompt_templates.py): `src/llm_prompting/templates/prompt_templates.py`
- [Language few-shot prompts](https://github.com/biba10/generative-cross-lingual-absa/tree/main/src/llm_prompting/templates): `src/llm_prompting/templates`

### Evaluation and metrics

- [Seq2seq evaluation](https://github.com/biba10/generative-cross-lingual-absa/blob/main/src/evaluation/evaluation.py): `src/evaluation/evaluation.py`
- [Seq2seq F1](https://github.com/biba10/generative-cross-lingual-absa/blob/main/src/evaluation/f1_score_seq2seq.py): `src/evaluation/f1_score_seq2seq.py`
- [Combined task F1](https://github.com/biba10/generative-cross-lingual-absa/blob/main/src/evaluation/total_f1_score.py): `src/evaluation/total_f1_score.py`
- [Prompted-model evaluation](https://github.com/biba10/generative-cross-lingual-absa/blob/main/src/evaluation/llm_evaluation.py): `src/evaluation/llm_evaluation.py`
- [General F1](https://github.com/biba10/generative-cross-lingual-absa/blob/main/src/evaluation/f1_score.py): `src/evaluation/f1_score.py`

### Configuration and logging

- [CLI arguments](https://github.com/biba10/generative-cross-lingual-absa/blob/main/src/utils/args_utils.py): `src/utils/args_utils.py`
- [Task and language config](https://github.com/biba10/generative-cross-lingual-absa/blob/main/src/utils/config.py): `src/utils/config.py`
- [Experiment logging](https://github.com/biba10/generative-cross-lingual-absa/blob/main/src/utils/logger_utils.py): `src/utils/logger_utils.py`

### Other components

- **Experimenter**
- **ChatGPT service**

## Connections

- Experimenter → Train and test: runs training (evidence: `README.md`)
- Experimenter → Open LLM run: runs prompting (evidence: `README.md`)
- Experimenter → ChatGPT run: runs prompting (evidence: `README.md`)
- Experimenter → Instruction tuning: runs tuning (evidence: `README.md`)
- Train and test → CLI arguments: parses arguments (evidence: `main.py`)
- Train and test → Model and tokenizer loader: loads model (evidence: `main.py`)
- Train and test → Seq2seq data loader: creates loader (evidence: `main.py`)
- Train and test → Generative ABSA model: fits and tests (evidence: `main.py`)
- Train and test → Experiment logging: configures logging (evidence: `main.py`)
- Open LLM run → Seq2seq data loader: builds data loader (evidence: `main_llm.py`)
- Open LLM run → Open LLM classifier: runs classifier (evidence: `main_llm.py`)
- Open LLM run → Model and tokenizer loader: loads tokenizer (evidence: `main_llm.py`)
- ChatGPT run → ChatGPT classifier: runs classifier (evidence: `main_prompting.py`)
- ChatGPT run → Seq2seq data loader: builds data loader (evidence: `main_prompting.py`)
- Instruction tuning → LLM dataset: uses training data (evidence: `instruction_tuning.py`)
- Instruction tuning → Open LLM classifier: uses classifier (evidence: `instruction_tuning.py`)
- Instruction tuning → CLI arguments: parses arguments (evidence: `instruction_tuning.py`)
- Seq2seq data loader → Seq2seq dataset: creates datasets (evidence: `src/data_utils/data_loader.py`)
- Seq2seq data loader → LLM dataset: creates datasets (evidence: `src/data_utils/data_loader.py`)
- Seq2seq data loader → API dataset: creates datasets (evidence: `src/data_utils/data_loader.py`)
- LLM dataset → Data preprocessing: encodes prompts (evidence: `src/data_utils/llm_dataset.py`)
- LLM dataset → Prompt templates: gets instructions (evidence: `src/data_utils/llm_dataset.py`)
- Prompt templates → Language few-shot prompts: selects templates (evidence: `src/llm_prompting/templates/prompt_templates.py`)
- Model and tokenizer loader → Generative ABSA model: constructs model (evidence: `src/model/model_tokenizer_loader.py`)
- Generative ABSA model → Seq2seq evaluation: evaluates predictions (evidence: `src/model/model_generative.py`)
- Seq2seq evaluation → Seq2seq F1: updates metrics (evidence: `src/evaluation/evaluation.py`)
- Seq2seq evaluation → Combined task F1: updates aggregate (evidence: `src/evaluation/evaluation.py`)
- Seq2seq F1 → General F1: extends metric (evidence: `src/evaluation/f1_score_seq2seq.py`)
- Combined task F1 → General F1: uses metric (evidence: `src/evaluation/total_f1_score.py`)
- Open LLM classifier → Seq2seq F1: scores predictions (evidence: `src/llm_prompting/llm_classifier.py`)
- ChatGPT classifier → LLM API access: requests completions (evidence: `src/llm_prompting/llm_api_classifier.py`)
- ChatGPT classifier → Prompt templates: builds prompts (evidence: `src/llm_prompting/llm_api_classifier.py`)
- ChatGPT classifier → Prompted-model evaluation: evaluates predictions (evidence: `src/llm_prompting/llm_api_classifier.py`)
- ChatGPT classifier → ChatGPT service: sends requests
