|
import pandas as pd |
|
import gradio as gr |
|
import matplotlib.pyplot as plt |
|
import seaborn as sns |
|
from seaborn import FacetGrid |
|
import plotly.express as px |
|
|
|
|
|
HEIGHT = 600 |
|
WIDTH = 1000 |
|
|
|
|
|
def plot_daily_invalid_trades_plotly(invalid_trades: pd.DataFrame): |
|
fig = px.histogram(invalid_trades, x="creation_date") |
|
return gr.Plot(value=fig) |
|
|
|
|
|
def plot_daily_dist_invalid_trades(invalid_trades: pd.DataFrame): |
|
"""Function to paint the distribution of daily invalid trades, no matter which market""" |
|
sns.set_theme(palette="viridis") |
|
plt.figure(figsize=(25, 10)) |
|
plot2 = sns.histplot(data=invalid_trades, x="creation_date", kde=True) |
|
plt.xlabel("Creation date") |
|
plt.ylabel("Daily number of invalid trades") |
|
plt.xticks(rotation=45, ha="right") |
|
daily_trades_fig = plot2.get_figure() |
|
return gr.Plot(value=daily_trades_fig) |
|
|
|
|
|
def plot_daily_nr_invalid_markets(invalid_trades: pd.DataFrame): |
|
"""Function to paint the number of invalid markets over time""" |
|
daily_invalid_markets = ( |
|
invalid_trades.groupby("creation_date") |
|
.agg(trades_count=("title", "count"), nr_markets=("title", "nunique")) |
|
.reset_index() |
|
) |
|
daily_invalid_markets["creation_date"] = daily_invalid_markets[ |
|
"creation_date" |
|
].astype(str) |
|
daily_invalid_markets.columns = daily_invalid_markets.columns.astype(str) |
|
|
|
return gr.LinePlot( |
|
value=daily_invalid_markets, |
|
x="creation_date", |
|
y="nr_markets", |
|
y_title="nr_markets", |
|
interactive=True, |
|
show_actions_button=True, |
|
tooltip=["creation_date", "nr_markets", "trades_count"], |
|
height=HEIGHT, |
|
width=WIDTH, |
|
) |
|
|
|
|
|
def plot_ratio_invalid_trades_per_market(invalid_trades: pd.DataFrame): |
|
"""Function to paint the number of invalid trades that the same market accummulates""" |
|
cat = invalid_trades["title"] |
|
codes, uniques = pd.factorize(cat) |
|
|
|
|
|
invalid_trades["title_id"] = codes |
|
plot: FacetGrid = sns.displot(invalid_trades, x="title_id") |
|
plt.xlabel("market id") |
|
plt.ylabel("Total number of invalid trades by market") |
|
plt.title("Distribution of invalid trades per market") |
|
return gr.Plot(value=plot.figure) |
|
|
|
|
|
def plot_top_invalid_markets(invalid_trades: pd.DataFrame): |
|
"""Function to paint the top markets with the highest number of invalid trades""" |
|
top_invalid_markets: pd.DataFrame = ( |
|
invalid_trades.title.value_counts().reset_index() |
|
) |
|
print(top_invalid_markets.head(5)) |
|
top_invalid_markets = top_invalid_markets.head(5) |
|
top_invalid_markets.rename(columns={"count": "nr_invalid_trades"}, inplace=True) |
|
return gr.DataFrame(top_invalid_markets) |
|
|
