pandas is a Python library for manipulating and analyzing tabular data. In a semantic-search pipeline, it turns a loaded dataset into a DataFrame that can be inspected, filtered, and counted. section 4.1
A practical preparation flow uses pandas to:
- Convert the Hugging Face Datasets train split into a DataFrame. section 4.1
- Inspect columns and rows with head(). section 4.1
- Filter records by locality and count hotel names. section 4.1
df = pd.DataFrame(dataset['train'])
Create the pandas DataFrame used for inspection and subsequent filtering. section 4.1
df_paris = df.loc[df.locality == 'Paris']
Select only records whose locality is Paris before continuing semantic-search preparation. section 4.1













