app.py

import gradio as gr
from hazm import Normalizer, word_tokenize, Lemmatizer, POSTagger, Chunker, DependencyParser

# Initialize Hazm components
normalizer = Normalizer()
lemmatizer = Lemmatizer()
tagger = POSTagger(model='resources/postagger.model')
chunker = Chunker(model='resources/chunker.model')
parser = DependencyParser(tagger=tagger, lemmatizer=lemmatizer)

def process_text(text, operations):
    result = {}
    if 'normalize' in operations:
        text = normalizer.normalize(text)
        result['Normalized Text'] = text
    if 'tokenize' in operations:
        tokens = word_tokenize(text)
        result['Tokens'] = tokens
    if 'lemmatize' in operations:
        lemmas = [lemmatizer.lemmatize(token) for token in word_tokenize(text)]
        result['Lemmas'] = lemmas
    if 'pos_tag' in operations:
        pos_tags = tagger.tag(word_tokenize(text))
        result['POS Tags'] = pos_tags
    if 'chunk' in operations:
        pos_tags = tagger.tag(word_tokenize(text))
        chunks = chunker.parse(pos_tags)
        result['Chunks'] = str(chunks)
    if 'dependency_parse' in operations:
        parse_tree = parser.parse(word_tokenize(text))
        result['Dependency Parse'] = str(parse_tree)
    return result

# Define Gradio interface
operations = ['normalize', 'tokenize', 'lemmatize', 'pos_tag', 'chunk', 'dependency_parse']
iface = gr.Interface(
    fn=process_text,
    inputs=[
        gr.inputs.Textbox(lines=10, label="Input Text"),
        gr.inputs.CheckboxGroup(operations, label="Operations")
    ],
    outputs="json",
    title="Persian Text Processor with Hazm",
    description="Select operations to perform on the input text using Hazm."
)

if __name__ == "__main__":
    iface.launch()