import gradio as gr
import pandas as pd
import matplotlib.pyplot as plt
import numpy as np
import csv
from scipy.stats import wilcoxon
import matplotlib.gridspec as gridspec
LIBRARIES = ["7764",
"250833",
#"278964"
"734943",
#"1304532",
#"1714548",
"2009901",
"2670515",
"3828396",
"3836952",
"4456656",
"5261717",
"6206924",
"7071551",
"7128869",
"7603319",
"7776928",
"10152778",
"10171263",
"10171270",
"10171280",
"10174980",
"12450835",
"12584701",
"12894267",
"14052249",
"14976868",
"15502567",
"19921167",
"21149814",
"23285197",
"28419588",
"28644964",
"28847821", "ALOY", "APSTUD", "CLI", "CLOV", "COMPASS", "CONFCLOUD", "CONFSERVER", "DAEMON", "DM", "DNN", "DURACLOUD", "EVG", "FAB",
"MDL", "MESOS" ,"MULE", "NEXUS", "SERVER", "STL", "TIDOC", "TIMOB", "TISTUD", "XD", ]
def grafico(list_output_MbR, list_output_NEOSP, list_output_TFIDF, nome_projeto, pip_choices):
list_results = [["MbR Regressor", np.mean(list_output_MbR)], ["NEOSP-SVR Regressor", np.mean(list_output_NEOSP)], ["TFIDF-SVR Regressor", np.mean(list_output_TFIDF)]]
df = pd.DataFrame(list_results, columns=["Model", "MAE"])
df_list_output_MbR = pd.DataFrame(list_output_MbR, columns=["MAE"])
df_list_output_NEOSP = pd.DataFrame(list_output_NEOSP, columns=["MAE"])
df_list_output_TFIDF = pd.DataFrame(list_output_TFIDF, columns=["MAE"])
fig, ax = plt.subplots(2, 2)
G = gridspec.GridSpec(2, 2)
axes_1 = plt.subplot(G[0, :])
axes_2 = plt.subplot(G[1, 0])
axes_3 = plt.subplot(G[1, 1])
# ax1
axes_1.set_xlabel("Index Execução")
axes_1.set_ylabel("MAE")
axes_1.legend()
axes_1.grid(True, which = "both", axis = "x" )
axes_1.minorticks_on()
if "MbR Regressor" in pip_choices:
axes_1.scatter(range(1,51), df_list_output_MbR["MAE"].loc[1:50], label="MbR Regressor", color="red", alpha=0.5,)
if "NEOSP-SVR Regressor" in pip_choices:
axes_1.scatter(range(1,51), df_list_output_NEOSP["MAE"].loc[1:50], label="NEOSP-SVR Regressor", color = "blue", alpha=0.5)
if "TFIDF-SVR Regressor" in pip_choices:
axes_1.scatter(range(1,51), df_list_output_TFIDF["MAE"].loc[1:50], label="TFIDF-SVR Regressor", color = "green", alpha=0.5)
# ax2
axes_2.set_ylabel("MAE Médio")
axes_2.set_xlabel("Modelos")
if "MbR Regressor" in pip_choices:
graf1 = axes_2.bar(df["Model"].iloc[[0]], df["MAE"].iloc[[0]], color="red", alpha=0.5)
axes_2.bar_label(graf1, fmt="%.03f", size=10, label_type="edge")
if "NEOSP-SVR Regressor" in pip_choices:
graf2 = axes_2.bar(df["Model"].iloc[[1]], df["MAE"].iloc[[1]], color = "blue", alpha=0.5)
axes_2.bar_label(graf2, fmt="%.03f", size=10, label_type="edge")
if "TFIDF-SVR Regressor" in pip_choices:
graf3 = axes_2.bar(df["Model"].iloc[[2]], df["MAE"].iloc[[2]], color = "green", alpha=0.5)
axes_2.bar_label(graf3, fmt="%.03f", size=10, label_type="edge")
# ax3
axes_3.set_xlabel("MAE")
axes_3.set_ylabel("Frequência")
if "MbR Regressor" in pip_choices:
axes_3.hist(df_list_output_MbR["MAE"], color="red", alpha=0.5)
if "NEOSP-SVR Regressor" in pip_choices:
axes_3.hist(df_list_output_NEOSP["MAE"], color="blue", alpha=0.5)
if "TFIDF-SVR Regressor" in pip_choices:
axes_3.hist(df_list_output_TFIDF["MAE"], color="green", alpha=0.5)
# graficos geral
fig.set_figwidth(15)
fig.set_figheight(8)
fig.suptitle("Projeto {}".format(nome_projeto))
# text
resultado, resultado2 = "", ""
#if (pip_choices == ["NEOSP-SVR Regressor", "MbR Regressor"]) or (pip_choices == ["MbR Regressor", "NEOSP-SVR Regressor"]):
res = wilcoxon(list_output_NEOSP, list_output_MbR)
resultado = "NEOSP-SVR vs. MbR => Statistics: {} | valor-p: {}".format(res.statistic, res.pvalue)
#if (pip_choices == ["NEOSP-SVR Regressor", "MbR Regressor"]) or (pip_choices == ["MbR Regressor", "NEOSP-SVR Regressor"]):
res = wilcoxon(list_output_NEOSP, list_output_TFIDF)
resultado2 = "NEOSP-SVR vs. TFIDF-SVR => Statistics: {} | valor-p: {}".format(res.statistic, res.pvalue)
res = wilcoxon(list_output_TFIDF, list_output_MbR)
resultado3 = "TFIDF-SVR vs. MbR => Statistics: {} | valor-p: {}".format(res.statistic, res.pvalue)
return gr.update(value=plt, visible=True), gr.update(value=resultado, visible=True), gr.update(value=resultado2, visible=True), gr.update(value=resultado3, visible=True)
def create_pip_plot(libraries, pip_choices):
nome_projeto = libraries
list_output_MbR = []
with open("metricas/metricas_{}_MbR.csv".format(nome_projeto), "r") as arquivo:
arquivo_csv = csv.reader(arquivo)
for i, linha in enumerate(arquivo_csv):
list_output_MbR.append(float(linha[0]))
list_output_NEOSP_SVR = []
with open("metricas/metricas_{}_NEOSP_SVR.csv".format(nome_projeto), "r") as arquivo:
arquivo_csv = csv.reader(arquivo)
for i, linha in enumerate(arquivo_csv):
list_output_NEOSP_SVR.append(float(linha[0]))
list_output_TFIDF_SVR = []
with open("metricas/metricas_{}_TFIDF.csv".format(nome_projeto), "r") as arquivo:
arquivo_csv = csv.reader(arquivo)
for i, linha in enumerate(arquivo_csv):
list_output_TFIDF_SVR.append(float(linha[0]))
return grafico(list_output_MbR, list_output_NEOSP_SVR, list_output_TFIDF_SVR, nome_projeto, pip_choices)
demo = gr.Blocks()
with demo:
with gr.Row():
with gr.Column():
gr.Markdown("## Conjunto de Dados")
libraries = gr.Dropdown(choices=LIBRARIES, label="Projeto", value="ALOY")
with gr.Column():
gr.Markdown("## Gráficos")
pip = gr.CheckboxGroup(choices=["MbR Regressor", "NEOSP-SVR Regressor", "TFIDF-SVR Regressor"], label="Modelos Preditivos")
# stars = gr.CheckboxGroup(choices=["Stars", "Week over Week"], label="")
# issues = gr.CheckboxGroup(choices=["Issue", "Exclude org members", "week over week"], label="")
with gr.Row():
fetch = gr.Button(value="Fetch")
with gr.Row():
with gr.Column():
star_plot = gr.Text(visible=False, label="Wilcoxon Test")
star_plot2 = gr.Text(visible=False, label="Wilcoxon Test")
star_plot3 = gr.Text(visible=False, label="Wilcoxon Test")
pip_plot = gr.Plot(visible=False)
# issue_plot = gr.Plot(visible=False)
fetch.click(create_pip_plot, inputs=[libraries, pip], outputs=[pip_plot, star_plot, star_plot2, star_plot3])
#fetch.click(create_star_plot, inputs=[libraries, pip], outputs=star_plot)
# fetch.click(create_issue_plot, inputs=[libraries, issues], outputs=issue_plot)
demo.launch()