# eddaoudi-mohamed/baby_care_chatboot architecture

> Architecture diagram of the GitHub repository eddaoudi-mohamed/baby_care_chatboot, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2025-05-03). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/eddaoudi-mohamed/baby_care_chatboot
- Repository: https://github.com/eddaoudi-mohamed/baby_care_chatboot

## Overview

##### Project type and purpose
- This is an ML fine-tuning toolkit for the ChatGLM-6B language model using the P-Tuning v2 method. Its primary purpose is to run parameter-efficient training, quantization, and evaluation on advertisement-generation (ADGEN) or custom datasets. It embeds into a command-line interface (shell scripts) that drives Python modules for training, evaluation, and optional web demo.

##### File structure
- Top-level files:
- README.md – project overview and usage
- requirements.txt – Python dependencies
- train.sh, evaluate.sh – shell scripts invoking training and evaluation
- ptuning/ directory:
- main.py – entry point parsing args and dispatching tasks
- arguments.py – CLI argument definitions and default configs
- trainer.py, trainer_seq2seq.py – training logic (P-Tuning v2, full fine-tune, seq2seq trainer)
- deepspeed.json – DeepSpeed config for distributed or ZeRO training
- (Other scripts mentioned: ds_train_finetune.sh, train_chat.sh, web_demo.sh are analogous to train.sh/evaluate.sh)

##### README for additional insights
- Describes dataflow: download ADGEN dataset → configure prompt lengths, quantization bits → run train.sh (P-Tuning v2) or ds_train_finetune.sh (full fine-tune) → outputs prefix encoder checkpoint → run evaluate.sh for metrics (Rouge, BLEU)
- Mentions usage of Huggingface Transformers, PyTorch, Deepspeed, quantization, and optional web demo.

a. Main components:
1. Shell Script Layer: train.sh, ds_train_finetune.sh, evaluate.sh, train_chat.sh, web_demo.sh
2. CLI Module: ptuning/main.py & arguments.py
3. Data Pipeline:
– DataLoader reading JSON dataset (ADGEN or custom)
– Tokenizer (from transformers AutoTokenizer)
4. Model Core:
– Base ChatGLM-6B model (pretrained, loaded from Huggingface or local path)
– PrefixEncoder (soft prompts)
5. Training Engine:
– P-Tuning v2 logic in trainer.py (freeze base model, train only PrefixEncoder)
– Seq2Seq or full fine-tune logic in trainer_seq2seq.py / ds_train_finetune.sh + DeepSpeed
– Quantization module (model.quantize(4), .half(), .cuda())
– Gradient accumulation and checkpointing
6. Evaluation Engine:
– Prefix checkpoint + base model loader
– Generation engine (.chat or .generate)
– Metric calculators (ROUGE, BLEU)
7. External Services & Storage:
– Huggingface Hub (model checkpoint repository)
– Cloud storage (Google Drive / Tsinghua) for data
– Local filesystem for output predictions and logs

b. Relationships / Data flow:
- Shell Script → invokes CLI Module → parses arguments → selects Data Pipeline & Training Engine → loads model → runs training → writes checkpoints
- After training, Shell Script → invokes CLI Module in evaluation mode → loads base model + prefix weights → runs generation → pipes outputs to Evaluation Engine → writes metrics and generated_predictions.txt
- Optional web_demo.sh → runs a small HTTP server / interactive loop loading the same model + prefix weights

c. Architectural patterns / design principles
- Modular separation via CLI module vs. core training logic
- Plug-and-play trainer classes (P-Tuning v2 vs. full fine-tune vs. seq2seq)
- Use of external config (deepspeed.json) for scalable multi-GPU / ZeRO
- Shell-script façade abstracts environment variables, hyperparameters

d. Key technologies / frameworks
- Python 3 + PyTorch
- Huggingface Transformers library (AutoTokenizer, AutoModel)
- Deepspeed for distributed / ZeRO training
- Model quantization APIs (INT4, FP16)
- Datasets library for JSON loading
- nltk, jieba, rouge-chinese for evaluation
- Shell scripting for orchestration

## Diagram (Mermaid)

```mermaid
flowchart TB
    %% Shell Script Layer
    subgraph "Shell Script Layer"
        direction TB
        TrainSC["train.sh"]:::orchestration
        EvalSC["evaluate.sh"]:::orchestration
        DSTrainSC["ds_train_finetune.sh"]:::orchestration
        ChatTrainSC["train_chat.sh"]:::orchestration
        WebDemoSC["web_demo.sh"]:::orchestration
    end

    %% CLI Module
    subgraph "CLI Module"
        direction TB
        MainPY["main.py"]:::orchestration
        ArgsPY["arguments.py"]:::orchestration
    end

    %% Data Pipeline
    subgraph "Data Pipeline"
        direction TB
        Dataset["JSON Dataset Loader"]:::data
        Tokenizer["AutoTokenizer"]:::data
    end

    %% Model Core
    subgraph "Model Core"
        direction TB
        BaseModel["ChatGLM-6B (Pretrained)"]:::model
        PrefixEnc["PrefixEncoder (Soft Prompts)"]:::model
    end

    %% Training Engine
    subgraph "Training Engine"
        direction TB
        subgraph "P-Tuning v2"
            PTuner["trainer.py"]:::compute
        end
        subgraph "Seq2Seq / Full Fine-Tune"
            SeqTuner["trainer_seq2seq.py"]:::compute
            DSConfig["deepspeed.json"]:::compute
        end
        QuantMod["Quantization Module"]:::compute
        GradAccum["Gradient Accumulation & Checkpointing"]:::compute
        Checkpoint["Checkpoint Storage"]:::storage
    end

    %% Evaluation Engine
    subgraph "Evaluation Engine"
        direction TB
        EvalLoader["Model+Prefix Loader"]:::compute
        Generation["Generation Engine"]:::compute
        Metrics["ROUGE, BLEU Calculator"]:::compute
        PredOutput["generated_predictions.txt"]:::storage
        MetricsOutput["metrics.json"]:::storage
    end

    %% External Services & Storage
    subgraph "External Services & Storage"
        direction TB
        HuggingHub["Huggingface Hub"]:::external
        CloudStore["Cloud Storage (GDrive/Tsinghua)"]:::external
        LocalFS["Local Filesystem"]:::external
    end

    %% Orchestration Flow
    TrainSC --> MainPY
    DSTrainSC --> MainPY
    EvalSC --> MainPY
    ChatTrainSC --> MainPY
    WebDemoSC --> MainPY

    MainPY --> ArgsPY
    MainPY --> Dataset
    MainPY --> Tokenizer
    MainPY --> BaseModel
    MainPY --> PrefixEnc

    BaseModel --"freeze base model"--> PTuner
    PrefixEnc --> PTuner
    PTuner --> QuantMod
    PTuner --> GradAccum
    PTuner --> Checkpoint

    BaseModel --> SeqTuner
    PrefixEnc --> SeqTuner
    SeqTuner --> DSConfig
    SeqTuner --> GradAccum
    SeqTuner --> Checkpoint

    QuantMod --> BaseModel
    QuantMod --> PrefixEnc

    Checkpoint --> EvalLoader
    EvalLoader --> Generation
    Generation --> Metrics
    Generation --> PredOutput
    Metrics --> MetricsOutput

    %% External Data & Models
    Dataset --> CloudStore
    BaseModel --> HuggingHub
    Checkpoint --> LocalFS
    PredOutput --> LocalFS
    MetricsOutput --> LocalFS

    %% Click Events
    click TrainSC "https://github.com/eddaoudi-mohamed/baby_care_chatboot/blob/master/ptuning/train.sh"
    click EvalSC "https://github.com/eddaoudi-mohamed/baby_care_chatboot/blob/master/ptuning/evaluate.sh"
    click MainPY "https://github.com/eddaoudi-mohamed/baby_care_chatboot/blob/master/ptuning/main.py"
    click ArgsPY "https://github.com/eddaoudi-mohamed/baby_care_chatboot/blob/master/ptuning/arguments.py"
    click PTuner "https://github.com/eddaoudi-mohamed/baby_care_chatboot/blob/master/ptuning/trainer.py"
    click SeqTuner "https://github.com/eddaoudi-mohamed/baby_care_chatboot/blob/master/ptuning/trainer_seq2seq.py"
    click DSConfig "https://github.com/eddaoudi-mohamed/baby_care_chatboot/blob/master/ptuning/deepspeed.json"
    click README["README.md"] "README.md"
    click Requirements["requirements.txt"] "requirements.txt"

    %% Styles
    classDef orchestration fill:#ADD8E6,stroke:#000,stroke-width:1px
    classDef data fill:#90EE90,stroke:#000,stroke-width:1px
    classDef model fill:#FFA500,stroke:#000,stroke-width:1px
    classDef compute fill:#9370DB,stroke:#000,stroke-width:1px
    classDef storage fill:#D3D3D3,stroke:#000,stroke-width:1px
    classDef external fill:#A9A9A9,stroke:#000,stroke-width:1px
```
