I am trying to read data from a large parquet file of 30G. My memory do not support default reading with fastparquet in python, so I do not know what I should do to lower the memory usage of the reading process.
Asked
Active
Viewed 566 times
1 Answers
1
You can use pyarrow's iter_batches to read back chunks of rows incrementally.

Micah Kornfield
- 1,325
- 5
- 10