Thanks for posting your question in the Microsoft Q&A forum.
Ensure that your source files contain valid data and remove any empty files and double-check that all data types are correctly set and match your source data structure.
Experiment with different partitioning strategies to see if it improves the output, insert additional sink steps throughout your data flow to identify where the data might be getting lost and try running the pipeline with a small subset of your data to isolate the issue.
Please don't forget to close up the thread here by upvoting and accept it as an answer if it is helpful