Update README.md
Browse files
README.md
CHANGED
@@ -35,3 +35,20 @@ The original dataset is sourced from the Hugging Face Hub, specifically the [big
|
|
35 |
Using the flexibility and efficiency of DuckDB, I processed the entire dataset without the need for heavy hardware. DuckDB's ability to handle large datasets efficiently allowed me to concatenate the markdown, code, and output for each notebook path into a single string, simulating the structure of a Jupyter notebook.
|
36 |
|
37 |
The transformation was performed using the following DuckDB query:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
35 |
Using the flexibility and efficiency of DuckDB, I processed the entire dataset without the need for heavy hardware. DuckDB's ability to handle large datasets efficiently allowed me to concatenate the markdown, code, and output for each notebook path into a single string, simulating the structure of a Jupyter notebook.
|
36 |
|
37 |
The transformation was performed using the following DuckDB query:
|
38 |
+
|
39 |
+
```python
|
40 |
+
import duckdb
|
41 |
+
|
42 |
+
Connect to a new DuckDB database
|
43 |
+
new_db = duckdb.connect('merged_notebooks.db')
|
44 |
+
|
45 |
+
Query to concatenate markdown, code, and output
|
46 |
+
query = """
|
47 |
+
SELECT path,
|
48 |
+
STRING_AGG(CONCAT('###Markdown\n', markdown, '\n###Code\n', code, '\n###Output\n', output), '\n') AS concatenated_notebook
|
49 |
+
FROM read_parquet('jupyter-code-text-pairs/data/*.parquet')
|
50 |
+
GROUP BY path
|
51 |
+
"""
|
52 |
+
|
53 |
+
Execute the query and create a new table
|
54 |
+
new_db.execute(f"CREATE TABLE concatenated_notebooks AS {query}")
|