README.md

---
    language: vi
    license: apache-2.0
    tags:
    - vietnamese
    - poem-analysis
    - instruction-tuned
    - flan-t5
    datasets:
    - kienhoang123/Vietnamese_Poem_Analysis_VN
    ---

    # Instruction-Tuned T5 Model for Vietnamese Poem Analysis

    This model was fine-tuned on kienhoang123/Vietnamese_Poem_Analysis_VN to analyze Vietnamese poetry using an instruction-based approach.

    ## Model Details

    - **Base Model**: google/flan-t5-small
    - **Training Data**: Vietnamese poem analysis dataset
    - **Tasks**: Extract emotion, metaphor, setting, motion, and prompt from Vietnamese poems

    ## Usage

    ```python
    from transformers import AutoTokenizer, AutoModelForSeq2SeqLM

    tokenizer = AutoTokenizer.from_pretrained("kienhoang123/Poem_Analysis_Instruct_VN")
    model = AutoModelForSeq2SeqLM.from_pretrained("kienhoang123/Poem_Analysis_Instruct_VN")

    # Create an instruction-based input
    instruction = '''
    Below is an instruction that describes a task.
    ### Instruction:
    Generate emotion, metaphor, setting, motion and prompt in Vietnamese for the following content.
    ### Input:
    Your Vietnamese poem here
    ### Output:
    '''

    inputs = tokenizer(instruction, return_tensors="pt")
    outputs = model.generate(**inputs, max_length=150)
    result = tokenizer.decode(outputs[0], skip_special_tokens=True)
    print(result)
    ```

    The output is formatted as: "emotion ||| metaphor ||| setting ||| motion ||| prompt"