USPTO Dataset for: Fast Chemical Reaction Condition Suggestion via Rule-Based Classification and Similarity Search

<p>USPTO database that is analyzed with Rxn-INSIGHT (<a href="https://github.com/mrodobbe/Rxn-INSIGHT">https://github.com/mrodobbe/Rxn-INSIGHT</a>).</p><p>This gzip file contains a very large Pandas DataFrame that can be loaded via pd.read_parquet('uspto_rxn_insight.gzip'). Because of the large size of the data, PyArrow version 13.0 must be used.&nbsp;</p><p>To use parquet in Pandas, install PyArrow and fastparquet using pip:</p><p>pip install pyarrow==13.0<br>pip install fastparquet</p>