Spaces:
Sleeping
Sleeping
Download app.py from Sfe61/FieldNames_Standardization: direct link, hf CLI and curl.
- Browser
- Download file 2.78 kB
-
https://huggingface.co/spaces/Sfe61/FieldNames_Standardization/resolve/main/app.py
- Command line
-
hf download hf://spaces/Sfe61/FieldNames_Standardization/app.py
-
curl -L -o app.py https://huggingface.co/spaces/Sfe61/FieldNames_Standardization/resolve/main/app.py
2.78 kB
| import gradio as gr | |
| import re | |
| def sentence_to_odk_fieldname(sentence, answer_type): | |
| # Dictionary for specific abbreviations | |
| abbreviations = { | |
| '%': 'percentage', | |
| 'internally displaced peoples': 'idps', | |
| 'minors under the age of 18': 'minors', | |
| 'minors under 18 years': 'minors', | |
| 'key informant':'ki', | |
| 'households': 'hh', | |
| } | |
| # Process text within parentheses | |
| def include_or_exclude(match): | |
| text = match.group(0) | |
| char_count = len(re.sub(r'[^\w\s]', '', text)) # Count characters excluding punctuation | |
| return text if char_count <= 20 else '' | |
| # Handle '/' character within words | |
| sentence = re.sub(r'\b/\b', '_', sentence) | |
| # Apply the logic for text within parentheses | |
| sentence = re.sub(r'\([^)]*\)', include_or_exclude, sentence) | |
| general_words_to_remove = {'do', 'you', 'to', 'your', 'a', 'the', 'is', 'are', 'of', 'in', 'this'} | |
| measurement_words = {'average', 'total', 'sum', 'count', 'maximum', 'minimum'} | |
| if answer_type in ['select_one', 'select_multiple']: | |
| additional_words_to_remove = {'what', 'which', 'who', 'how', 'why', 'did'} | |
| else: | |
| additional_words_to_remove = set() | |
| words_to_remove = general_words_to_remove.union(additional_words_to_remove) | |
| sentence = sentence.lower() | |
| for key, value in abbreviations.items(): | |
| sentence = sentence.replace(key, value) | |
| words = ''.join(char for char in sentence if char.isalnum() or char.isspace()).split() | |
| filtered_words = [word for word in words if word not in words_to_remove] | |
| if answer_type in ['decimal', 'integer']: | |
| field_name_parts = [word for word in filtered_words if word in measurement_words or word not in measurement_words] | |
| else: | |
| entity, context = None, [] | |
| for word in filtered_words: | |
| if not entity: | |
| entity = word | |
| else: | |
| context.append(word) | |
| field_name_parts = [entity] if entity else [] | |
| if context: | |
| field_name_parts.append('_'.join(context)) | |
| field_name = '_'.join(field_name_parts) | |
| field_name = re.sub('_+', '_', field_name) | |
| field_name = field_name[:128] | |
| return field_name | |
| def main(sentence, answer_type): | |
| field_name = sentence_to_odk_fieldname(sentence, answer_type) | |
| character_count = len(field_name) | |
| return field_name, character_count | |
| answer_types = ["select_one", "select_multiple", "decimal", "integer", "text"] | |
| iface = gr.Interface( | |
| fn=main, | |
| inputs=[ | |
| gr.Textbox(label="Enter Sentence"), | |
| gr.Dropdown(choices=answer_types, label="Answer Type") | |
| ], | |
| outputs=[ | |
| gr.Textbox(label="Field Name"), | |
| gr.Textbox(label="Character Count") | |
| ] | |
| ) | |
| iface.launch() |