Changes to inplace operations for large dataframes.

This commit is contained in:
2023-01-11 09:53:30 +01:00
parent 41ccc31ac7
commit 2754882147

View File

@@ -194,14 +194,14 @@ if __name__ == "__main__":
# Get the frequency, PDF and CDF for each value in the series
# Frequency
transmission_df.groupby("goodput")["goodput"].agg("count").pipe(pd.DataFrame)
transmission_df.rename(columns={"goodput": 'frequency'}, inplace=True)
transmission_df["gp_frequency"] = transmission_df["goodput"]
transmission_df.groupby("gp_frequency")["gp_frequency"].agg("count").pipe(pd.DataFrame)
# PDF
transmission_df['pdf'] = transmission_df['frequency'] / sum(transmission_df['frequency'])
transmission_df["pdf"] = transmission_df["gp_frequency"] / sum(transmission_df["gp_frequency"])
# CDF
transmission_df['cdf'] = transmission_df['pdf'].cumsum()
transmission_df["cdf"] = transmission_df["pdf"].cumsum()
transmission_df.reset_index(inplace=True)
transmission_df.plot(x="goodput", y=["cdf"], grid=True)