def plot_multimodal_bars(PP):
sns.catplot(data=PP,x='perception',y='logLikelihood',
hue='pragmatics',kind='bar',
order=['pool1','conv42','fc6'],
palette='Paired',
legend=False,
ci=None)
plt.ylabel('log likelihood')
locs, labels = plt.xticks([0,1,2],['early','mid','high'],fontsize=14)
plt.xlabel('visual features')
# plt.ylim([-3000,0])
plt.tight_layout()
plt.savefig('./plots/loglikelihood_models_multimodal.pdf')
def plot_human_bars(PP):
sns.catplot(data=PP,x='pragmatics',y='logLikelihood',
hue='production',kind='bar',
order=['S0','combined'],
hue_order=['nocost','cost'],
palette='Paired',
legend=False,
ci=None)
plt.ylabel('log likelihood')
locs, labels = plt.xticks([0,1],['insensitive','sensitive'],fontsize=14)
plt.xlabel('context')
# plt.ylim([-3000,0])
plt.tight_layout()
plt.savefig('./plots/loglikelihood_models_human.pdf')
## classical hypothesis test
p = scipy.stats.binom_test(p1*n1, n=n1, p=1/32, alternative='two-sided')
print 'Closer proportion diff from chance? p = {}'.format(p)
p = scipy.stats.binom_test(p2*n2, n=n2, p=1/32, alternative='two-sided')
print 'Further proportion diff from chance? p = {}'.format(p)
##### MAKE PLOTS AND SAVE OUT
## plot recognition accuracy by condition
sns.set_context('poster')
fig = plt.figure(figsize=(4,4))
redgld=[(0.8, 0.2, 0.2),(0.9, 0.7, 0.3)]
sns.catplot(y='correct',
x='target_category',
hue='condition',
hue_order=['closer','further'],
order=['bird','car','chair','dog'],
data=X,kind='bar',palette=redgld)
plt.ylim([0,1])
plt.ylabel('proportion correct')
plt.xlabel('category')
h = plt.axhline(1/32,linestyle='dashed',color='black')
plt.savefig('./plots/accuracy_by_category_and_condition.pdf')
plt.close(fig)
## plot recognition accuracy by condition
plt.figure(figsize=(2,4))
# sns.set_context('poster')
redgld=[(0.8, 0.2, 0.2),(0.9, 0.7, 0.3)]
sns.catplot(y='correct',
x='condition',
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
df = pd.read_excel('./residue/无.xlsx')
sns.catplot(x="class", y="Residue", kind="box", data=df)
plt.show()
for metric in metrics:
if 'mask' not in metric:
df[metric + '_mean'] = df[metric].apply(np.mean)
for metric in metrics:
if 'mask' in metric:
continue
# plt.figure()
# ax = sns.swarmplot(data=df, x='site', y=metric + '_mean', hue='fake',
# linewidth=.9, edgecolor="black", size=6, dodge=True)
g = sns.catplot(x='site',
y=metric + '_mean',
hue='fold',
col='fake',
data=df,
kind='swarm',
linewidth=.9,
dodge=False,
sharey=False)
d = sns.catplot(x='fake',
y=metric + '_mean',
hue='real_name',
data=df,
kind='point')
#plt.xticks([sites[x] for x in df['site'].values], rotation=90)
plt.title(metric)
# for metric in metrics:
# if 'mask' in metric:
ax.set_ylabel("Number of Observaitons")
ax.set_xticklabels(df.Year, rotation = 90 ) #Yazılar birbirine giriyorsa daha okunaklı olması için kullanılır
plt.show()
fig.savefig("Year.png")
"""**Seaborn**
> AxesSubplot
> FaceGrid
"""
sns.set_palette("RdBu")
sns.countplot(x="Year", data = df)
plt.show()
sns.catplot(x="Year", aspect=3 , data = df , kind="count")
plt.show()
g = sns.catplot(x="Year", aspect=3 , data = df , kind="count")
g.fig.suptitle("Year Counts", y=1)
plt.xticks(rotation = 90)
plt.show()
sns.scatterplot(x="Year" , y="Selling_Price", data= df, hue="New")
plt.show()
sns.relplot(x="Year" , y="Selling_Price", data= df, hue="New", kind="scatter")
plt.show()
"""**LINE PLOT**"""
top_dados_controle = dados_controle.query('composto in @cod_dados_controle')
cod_dados_droga = dados_droga['composto'].value_counts().index[0:100]
top_dados_droga = dados_droga.query('composto in @cod_dados_droga')
cod_dados_controle
top_dados_controle
cod_dados_droga
top_dados_droga
dados_controle['composto'].unique()
plot1 = sns.catplot(x='composto',data=top_dados_controle,col="dose",kind="count",col_wrap=2 ,height=4, aspect=.7)
plot2 = sns.catplot(x='composto',data=top_dados_droga,col="dose",kind="count",height=7, aspect=.8)
"""#### - achar o composto em 'cacb2b860' dados droga"""
findArray = dados_droga['composto'].unique()
exist = 'cacb2b860' in findArray
exist
"""#### - Posso concluir que o controle so foi realizado no composto 'cacb2b860'
##Separando compostos
"""
overwrite=False)
pipeline = make_pipeline(CSP(n_components=8), LDA())
results = evaluation.process({"csp+lda": pipeline})
# To export the results in CSV within a directory:
if not os.path.exists("./results"):
os.mkdir("./results")
results.to_csv("./results/results_part2-2.csv")
# To load previously obtained results saved in CSV
results = pd.read_csv("./results/results_part2-2.csv")
##############################################################################
# Plotting Results
# ----------------
#
# We plot the results using the seaborn library. Note how easy it
# is to plot the results from the three datasets with just one line.
results["subj"] = [str(resi).zfill(2) for resi in results["subject"]]
g = sns.catplot(
kind="bar",
x="score",
y="subj",
col="dataset",
data=results,
orient="h",
palette="viridis",
)
plt.show()
# plt.show()
# Draw a boxplot with nested grouping by two categorical variables:
ax = sns.boxplot(x="day",
y="total_bill",
hue="smoker",
data=tips,
palette="Set3")
plt.savefig("boxplot with nested grouping by two categorical variables.jpg")
# plt.show()
# Draw a boxplot with nested grouping when some bins are empty:
ax = sns.boxplot(x="day", y="total_bill", hue="time", data=tips, linewidth=2.5)
plt.savefig("boxplot with nested grouping when some bins are empty.jpg")
# plt.show()
# Control box order by passing an explicit order:
ax = sns.boxplot(x="time", y="tip", data=tips, order=["Dinner", "Lunch"])
plt.show()
# Use catplot() to combine a pointplot() and a FacetGrid. This allows grouping within additional categorical variables.
# Using catplot() is safer than using FacetGrid directly, as it ensures synchronization of variable order across facets:
g = sns.catplot(x="sex",
y="total_bill",
hue="smoker",
col="time",
data=tips,
kind="box",
height=4,
aspect=.7)
plt.show()
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
from scipy.stats import ranksums
''' Analise 5 - Hora do dia '''
gorjetas = pd.read_csv('data\\tips_tratados_4.csv',sep=',')
print(gorjetas.head())
print(gorjetas.hora_do_dia.unique())
sns.catplot(x='hora_do_dia',y='valor_da_conta',data=gorjetas)
plt.show()
# Espalha alguns pontos proximos de forma distribuida para uma melhor visualização
sns.catplot(x='hora_do_dia',y='valor_da_conta', kind='swarm',data=gorjetas)
plt.show()
# Demonstra a visualização em formato de violino, onde a maior concentração será apresentada na parte mais gorda
sns.violinplot(x='hora_do_dia',y='valor_da_conta',data=gorjetas)
plt.show()
# Demonstra a estatisticas do metodo describe de uma forma visual
sns.boxplot(x='hora_do_dia',y='valor_da_conta',data=gorjetas)
plt.show()
# Histograma = Um gráfico que tem, no eixo X, o valor da variável sendo exibida e no outro eixo, a frequência.
# Histograma do almoço
almoço = gorjetas.query("hora_do_dia == 'Almoço'").valor_da_conta
sns.distplot(almoço)
plt.show()
axis=1,
keys=['Total', 'Percent'])
print(missing_data_1)
#Code ends here
# --------------
#Code starts here
#Setting the figure size
plt.figure(figsize=(10, 10))
#Plotting boxplot between Rating and Category
cat = sns.catplot(x="Category", y="Rating", data=data, kind="box", height=10)
#Rotating the xlabel rotation
cat.set_xticklabels(rotation=90)
#Setting the title of the plot
plt.title('Rating vs Category [BoxPlot]', size=20)
#Code ends here
# --------------
#Importing header files
from sklearn.preprocessing import MinMaxScaler, LabelEncoder
#Importing header files
from sklearn.preprocessing import MinMaxScaler, LabelEncoder
ax = sns.kdeplot(data=train['Age'], shade=True, gridsize = 30)
_ = ax.set(title='Age distribution', ylabel='Distribution', xlabel='Age - months')
# %% [markdown]
# ## We've got similar number of cats and dogs. Most of them are around 1 year old with slightly higher number of female pets.
# %% [code] {"_kg_hide-input":true}
num = 10
mixed_breed_class = 307
plt.figure(figsize=(20,20))
indexes, values = train['Breed1'][(train['Type'] == 'Dog')].value_counts().index[:num], train['Breed1'].value_counts()[:num]
names = [id_to_breed(i) for i in indexes]
s = pd.Series(data={'values': values.values, 'names': names})
ax = sns.catplot(x = 'values', y = 'names' , kind='bar', data = s)
_ = ax.set(title=f'Dog breed classes top {num}', ylabel='Dog breed', xlabel='Count')
indexes, values = train['Breed1'][(train['Type'] == 'Cat')].value_counts().index[:num], train['Breed1'].value_counts()[:num]
names = [id_to_breed(i) for i in indexes]
s = pd.Series(data={'values': values.values, 'names': names})
ax = sns.catplot(x = 'values', y = 'names' , kind='bar', data = s)
_ = ax.set(title=f'Cat breed classes top {num}', ylabel='Cat breed', xlabel='Count')
pure_breeded = train['Breed1'].apply(lambda x: 0 if id_to_breed(x) in ['Mixed Breed', 'Domestic Short Hard', 'Domestic Medium Hair', 'Domestic Long Hair'] else 1)
print(f'Pure breeded pets: {sum(pure_breeded)}\nNot pure breeded pets: {len(pure_breeded)-sum(pure_breeded)}')
# %% [markdown]
# ### As we can see dogs as well as cats breeds are mostly dominated by ~3 classes. We've got 7512 purebreeded pets and 7481 pets that aren't purebreeded.
# %% [code] {"_kg_hide-input":true}
#Histogram Plot of windspeed Column
plt.figure(figsize=(7, 7))
plt.hist(train['windspeed'], bins=10)
plt.xlabel('windspeed')
plt.ylabel('Frequency')
#Histogram Plot of count Column
plt.figure(figsize=(7, 7))
plt.hist(train['count'], bins=10)
plt.xlabel('count')
plt.ylabel('Frequency')
################################################## Bivariate Plots #################################################################################################
for i in cat_cnames:
sns.catplot(x=i, y="count", data=train)
fname = str(i) + '.pdf'
################################################## Density Plots ##################################################################################################
sns.kdeplot(train['season'], shade=True)
sns.kdeplot(train['month'], shade=True)
sns.kdeplot(train['holiday'], shade=True)
sns.kdeplot(train['weekday'], shade=True)
sns.kdeplot(train['workingday'], shade=True)
sns.kdeplot(train['weather'], shade=True)
sns.kdeplot(train['temperature'], shade=True)
sns.kdeplot(train['atemp'], shade=True)
sns.kdeplot(train['humidity'], shade=True)
sns.kdeplot(train['windspeed'], shade=True)
sns.kdeplot(train['count'], shade=True)
ax[i, -1].set_xlabel('RT (s)')
ax[i, 0].set_ylabel('Reward manipulation = {:.2f}'.format(rewardfactor))
exp1 = pd.read_csv('../data/exp1.csv')
exp2 = pd.read_csv('../data/exp2.csv')
exp1 = exp1[(exp1['dyn'] == 'Dynamic') & (exp1['sub'] != 666) &
(exp1['setsize'] == 12)]
exp2 = exp2[(exp2['dyn'] == 'Dynamic') & (exp2['sub'] != 666) &
(exp2['setsize'] == 12)]
exp1['reward'] = 'None'
newdf = exp1.append(exp1, sort=False)
g = sns.catplot(x='reward',
y='rt',
hue='target',
data=newdf,
kind='point',
order=['Absent', 'None', 'Present'])
g.set_xticklabels(size=18)
g.set_yticklabels(size=18)
g.set_ylabels('RT (s)', size=20)
g.set_xlabels('Reward Condition', size=20)
ax = plt.gca()
ax.plot([0, 1, 2],
sim_mrts[::-1, 1, 1, 1],
label='Sim Present',
c='lightgreen',
lw=2)
ax.plot([0, 1, 2],
sim_mrts[::-1, 1, 0, 0],
plt.style.use('ggplot')
plt.xlabel("Victory Status")
plt.ylabel("Winner")
plt.title("Scatter Plot")
plt.show()
#relplot
sns.set(style="darkgrid")
sns.relplot(x="opening_ply", y="statusEnc", hue="winner", style="rated",ci=None,
dashes=False, markers=True, kind="line", data=games);
plt.ylabel("Victory Status")
plt.show()
black = games.query("winner == 'black'")
white = games.query("winner == 'white'")
draw = games.query("winner == 'draw'")
#jointplot
sns.set(style="white")
sns.jointplot(games.white_rating, games.black_rating, kind="kde", height=7, space=0)
plt.show()
#cat plot
pl = sns.catplot(x='victory_status',y='turns',hue='winner',data=games,
height=6, kind="bar", palette="muted")
pl.despine(left=True)
plt.show()
#Count plot
sns.countplot(x='winner',data=games)
plt.show()
hue='year',
x='race',
)
if SHOW:
show()
else:
savefig('./race_year_count.png', )
del count_figure
cat_figure = figure()
cat = catplot(
col='gender',
data=input_df[[
'gender',
'race',
'year',
]],
hue='year',
kind='count',
x='race',
)
if SHOW:
show()
else:
savefig('./race_year_catplot.png', )
del cat_figure
count_df = DataFrame([(key[0], key[1], key[2], value)
for key, value in dict(
Counter([
tuple(item) for item in input_df[[
#Convert csv file to dataframe
df = pd.read_csv(filepath, encoding = 'ISO-8859-1')
#Seperate only heart rate observations
df_heartrate = df[df['type'].str.contains('HeartRate')]
#Convert creation date column to datetime format
df_heartrate['creation_date'] = pd.to_datetime(df_heartrate['creation_date'],
format='%Y-%m-%d %H:%M:%S', utc=True)
df_heartrate['creation_date'] = df_heartrate['creation_date'].dt.tz_convert(None)
#Find integer value of weekday
df_heartrate['weekday'] = df_heartrate['creation_date'].dt.dayofweek
dayofweek = df_heartrate['weekday'].to_numpy()
#Convert integer value of weekday to string represenation
dayofweek_string = []
for i in range(0, len(dayofweek)):
dayofweek_string.append(calendar.day_name[dayofweek[i]])
df_heartrate['weekday'] = dayofweek_string
#Set index of dataframe as creation date
df_heartrate.set_index('creation_date', inplace=True, drop=True)
#Create and show violin plot
sns.catplot(x='weekday', y='value',
data=df_heartrate, kind='violin',
order=['Sunday', 'Monday', 'Tuesday', 'Wednesday', 'Thursday', 'Friday', 'Saturday'],
inner='quartile')
plt.show()
estimator=sum,
palette=sns.color_palette("muted",
n_colors=len(despesas_sem_vencimentos) + 4),
)
plot.set_xticklabels(plot.get_xticklabels(),
rotation=75,
horizontalalignment="right")
plt.show()
# In[48]:
sns.catplot(
x="value",
y="Descricao",
col="month",
data=despesas_com_classificacao,
kind="bar",
height=4,
aspect=0.7,
)
# In[49]:
despesas_com_classificacao[[
"legal_status", "subgroup", "Descricao", "summary", "Codigo"
]].head(100)
# In[50]:
# Auxílio-Alimentação
# Vale refeição/vale alimentação
def main(args):
df, info = basic_compare('./data/toysff.p', './data/amazon.p', 'toysff',
'amazon')
output_as_csv(df, './output/price_differences.csv')
# prepare data for bar plot
df = df.rename(columns={'toysff_name': 'name'})
df_amazon = df.drop(['amazon_name', 'toysff_price'],
1).rename(columns={'amazon_price': 'price'})
df_toysff = df.drop(['amazon_name', 'amazon_price'],
1).rename(columns={'toysff_price': 'price'})
df_amazon['src'] = ['amazon'] * df_amazon['article_nr'].count()
df_toysff['src'] = ['toys for fun'] * df_toysff['article_nr'].count()
result = pd.concat([df_amazon, df_toysff])
print(result)
pickle.dump(result, open('./data/comparison_amazon_toysff', 'wb'))
if args.csv:
output_as_csv(result, './output/comparison_amazon_toysff.csv')
# data for subbrand plot
subbrand_data = result[['subbrand', 'src', 'price']]
subbrand_aggregated = subbrand_data.groupby(
['subbrand', 'src'])['price'].agg(['sum', 'count', 'mean'])
subbrand_aggregated = subbrand_aggregated.reset_index()
# data for price segment plot
segment_data = result[['price_segment', 'src', 'price']]
segment_aggregated = segment_data.groupby(
['price_segment', 'src'])['price'].agg(['sum', 'count', 'mean'])
segment_aggregated = segment_aggregated.reset_index()
# for the correct count- and sum-aggregation we need to drop duplicates
# all values below are the same for 'src=amazon' and 'src=toysff'
result_single = result[[
'article_nr', 'subbrand', 'price_segment', 'diff_abs', 'diff_%'
]].drop_duplicates()
# Output on console and to csv if set true
output_diff_by_category(result_single, 'subbrand', 'diff_abs', args.csv)
output_diff_by_category(result_single, 'subbrand', 'diff_%', args.csv)
output_diff_by_category(result_single, 'price_segment', 'diff_abs',
args.csv)
output_diff_by_category(result_single, 'price_segment', 'diff_%', args.csv)
if args.plot:
# visualisation subbrands
sns.set(style="whitegrid")
sns.catplot(y='subbrand',
x='mean',
orient='h',
height=8,
hue='src',
kind='bar',
data=subbrand_aggregated,
aspect=.8)
plt.title('Subbrands: Average Prices in €')
plt.tight_layout(pad=1.08, h_pad=None, w_pad=None, rect=None)
# visualation price_segments
sns.set(style="whitegrid")
sns.catplot(y='price_segment',
x='mean',
orient='h',
height=8,
hue='src',
kind='bar',
data=segment_aggregated,
aspect=.8)
plt.title('Price Segments: Average Prices €')
plt.tight_layout(pad=1.08, h_pad=None, w_pad=None, rect=None)
plt.show()
print('Main Statistics: ')
print(info)
#plt.leyend('asndlsad')
plt.plot()
plt.barh(resumen(cierre, '19').reset_index()['X ENS'])
#############
sns.set(style="whitegrid")
# Load the example Titanic dataset
titanic = sns.load_dataset(resumen(cierre, '19'))
# Draw a nested barplot to show survival for class and sex
g = sns.catplot(x="PO",
y=["X ENS", 'X PINTAR'],
hue="CUSTOMER",
data=resumen(cierre, '19').reset_index(),
height=6,
kind="bar",
palette="muted")
g.despine(left=True)
g.set_ylabels("survival probability")
titanic = sns.load_dataset("titanic")
sns.catplot(y="CUSTOMER",
hue="CUSTOMER",
kind="count",
palette="pastel",
edgecolor=".6",
data=carga)
def gender_bar_graph():
sns.catplot(x="sex", kind="count", palette="magma", data=data, height=6)
plt.title("Gender of students : F - female,M - male")
plt.show()
# stats
pValsDF, bhDF = statsFeat(FeatMatGrouped, 'normal', 0.1, control)
bhDF = bhDF.reset_index()
bhDF['drug'] = [list(i) for i in bhDF.metadata.str if i.dtype == object][0]
bhDF['date'] = bhDF.metadata.str[0]
bhDF['worm_number'] = bhDF.metadata.str[1]
bhDF['window'] = bhDF.metadata.str[-1]
bhDF = bhDF.drop(columns='metadata')
#plot the total number of sig feats by worm number and drug (with window as hue) for each date
for date in metadata_dict['date']:
sns.catplot(x = 'worm_number', \
y = 'sumSig',\
data = bhDF[bhDF['date']==date], \
hue = 'window', \
col = 'drug',\
kind ='bar', \
palette = 'colorblind'
)
plt.savefig(os.path.join(save_dir, 'T_test_number_sig_feats5mins_{}.tif'.format(date)), bbox_inches='tight', \
pad_inches=0.03)
#and for all data combined
sns.catplot(x = 'worm_number', \
y = 'sumSig',\
data = bhDF, \
hue = 'window', \
col = 'drug',\
kind ='bar', \
palette = 'colorblind',
ci= 'sd'
)
df2.groupby(['biological_sample_group', 'sex'])[['data_point']].mean() #how to pass this along to the plot functions? #seems like a whisker plot can be generated directly from the group object #scatter plot of values in the 4 groups - put those into 4 variables???? # need color or shape to show M vs F and KO vs WT import matplotlib.pyplot as plt x = df2['date_of_experiment'] y = df2['data_point'] plt.scatter(x,y) plt.show() #how to do a scatterplot and color and shape the 4 groups ?? #here is column plot of KO vs WT grouped by sex import seaborn as sns x = df2['date_of_experiment'] y = df2['data_point'] sns.catplot(x = 'sex', y = 'data_point', col = 'biological_sample_group', kind = 'bar', data=df2) plt.show() # does not have mut vs wt # any reason to show weight? #sns.violinplot(x = 'sex', y = 'weight', col = 'biological_sample_group', kind = 'violin', data=df2) #plt.show()
df_sub['p_corrected'] = ps_corrected[1]
temp.append(df_sub)
anova_results = pd.concat(temp)
anova_results['stars'] = anova_results['p_corrected'].apply(utils.stars)
anova_results = anova_results.sort_values(['roi', 'condition', 'model'])
anova_results.to_csv('../../../../results/{}/RP/{}/one way ANOVA.csv'.format(
experiment, 'encoding 15 stats'),
index=False)
g = sns.catplot(
x='roi_name',
y='mean_variance',
hue='model_name',
hue_order=[
'VGG19', 'DenseNet1211', 'MobileNetV2', 'Fast Text', 'GloVe',
'Word2Vec'
],
row='condition',
data=df,
kind='bar',
aspect=6,
sharey=False,
)
g._legend.set_title('Encoding Models')
(g.set_axis_labels(
"ROIs",
"Mean Variance Explained").set_titles("{row_name}").set(ylim=(0, 0.06)))
g.axes[0][0].set(title='Shallow Process')
g.axes[1][0].set(title='Deep Process')
k = {'I2V': -0.25, 'W2V': 0.175}
j = 0.15
l = 0.0005
import seaborn as sns
import matplotlib.pyplot as plt
df = sns.load_dataset("tips")
sns.set()
sns.catplot(x="sex",
y="total_bill",
data=df,
kind="bar",
col="day",
col_wrap=2)
plt.show()
plt.figure(figsize=(5, 5)) plt.tight_layout() sns.boxplot(x='Type', y='Rating', data=df) # <b><i style="font-size:14pt;"><u>Installs</u> par <u>Price</u> :</i></b> # In[277]: # Paid Vs free et le nombre d'Installs installs_greater_1000 = df[df["Installs"] > 1000] installs_greater_1000 = installs_greater_1000.sort_values(['Price']) # In[278]: plt.figure(figsize=(20, 20)) sns.catplot(x="Installs", y="Price", data=installs_greater_1000) plt.xticks(rotation=90) plt.show() # <b><i style="font-size:14pt;"><u>Category</u> par <u>Size</u> :</i></b> # In[279]: plt.figure(figsize=(5, 15)) sns.barplot(x='Size', y='Category', data=df) # > <b><i style="font-size:14pt;">Analyses de la variable <u>Installs</u> : dans cette partie on va se consacrer de la variable Installs avec les autres variables</i></b> # # <b><i style="font-size:14pt;">Groupement des nombres d'Installs en 4 groupes : A, B, C, Highest.</i></b>
'awareness'] + ', ' + df_plot['confidence']
#temp = []
#for (target,subject),df_sub in df_plot.groupby(['success','sub']):
# df_sub['prob'] = df_sub['count'] / df_sub['count'].sum()
# temp.append(df_sub)
#df_plot = pd.concat(temp)
df_plot.to_csv(os.path.join(saving_dir, 'pos_for_plot.csv'))
df_plot = df_plot.sort_values(['awareness'])
g = sns.catplot(
x='awareness',
y='prob',
hue='confidence',
col='correctness',
row='success',
row_order=['high pos', 'low pos'],
data=df_plot,
kind='bar',
aspect=2,
)
(g.set_axis_labels('Awareness',
'Probability').set_titles("{row_name} | {col_name}").set(
ylim=(0., 0.85)).despine(left=True))
for ii, (target, df_sub) in enumerate(df_plot.groupby('success')):
# formula = 'prob ~ C(correctness)*C(awareness)*C(confidence)'
# model = ols(formula, df_sub).fit()
# aov_table = anova_lm(model, typ=2)
# s = f"{target}, F({model.df_model: .0f},{model.df_resid: .0f}) = {model.fvalue: .3f}, p = {model.f_pvalue: .4f}"
# print(s)
# g.axes[ii][0].annotate(s,xy=(-0.45,.8))
ci=None,
)
plt.xlabel("Season")
plt.ylabel("Hourly number of bikes rented")
plt.title("Number of bikes rented per hour by weather condition and season")
plt.xticks(ticks=(0, 1, 2, 3, 3.5))
plt.grid(which="major", axis="y")
# In[15]:
sns.catplot(
x="mnth",
y="cnt",
kind="point",
hue="workingday",
data=hour.compute(),
ci=None,
palette="Set1",
aspect=2.3,
legend=False,
)
plt.legend(("Weekend", "Workday"),
loc="upper right",
bbox_to_anchor=(1.2, 0.5))
plt.xlabel("Month")
plt.ylabel("Hourly number of bikes rented")
plt.title("Number of bikes rented per hour by type of day")
plt.axhline(hour.cnt.mean().compute(), ls="--", color="#a5a5a5")
plt.text(0.5, hour.cnt.mean().compute() - 10, "Average", color="#a5a5a5")
# In[16]:
train_data["Duration_hours"] = duration_hours
train_data["Duration_mins"] = duration_mins
train_data.drop(["Duration"], axis = 1, inplace = True)
train_data.head()
"""---"""
train_data["Airline"].value_counts()
# From graph we can see that Jet Airways Business have the highest Price.
# Apart from the first Airline almost all are having similar median
# Airline vs Price
sns.catplot(y = "Price", x = "Airline", data = train_data.sort_values("Price", ascending = False), kind="boxen", height = 6, aspect = 3)
plt.show()
# As Airline is Nominal Categorical data we will perform OneHotEncoding
Airline = train_data[["Airline"]]
Airline = pd.get_dummies(Airline, drop_first= True)
Airline.head()
train_data["Source"].value_counts()
# Source vs Price
sns.catplot(y = "Price", x = "Source", data = train_data.sort_values("Price", ascending = False), kind="boxen", height = 4, aspect = 3)
"""
Plotting a three-way ANOVA
==========================
_thumb: .42, .5
"""
import seaborn as sns
sns.set(style="whitegrid")
# Load the example exercise dataset
df = sns.load_dataset("exercise")
# Draw a pointplot to show pulse as a function of three categorical factors
g = sns.catplot(x="time", y="pulse", hue="kind", col="diet",
capsize=.2, palette="YlGnBu_d", height=6, aspect=.75,
kind="point", data=df)
g.despine(left=True)
# 지역별로 subplot 그리기
# col = 어떤 기준으로 subplot 그릴건지?
sns.relplot(data=df_last,
x="연도",
y="평당분양가격",
hue="지역명",
kind="line",
col="지역명",
col_wrap=4,
ci=None)
# 연도별 평당 가격을 지역별로 subplot bar chart로 표현
sns.catplot(data=df_last,
x="연도",
y="평당분양가격",
kind="bar",
col="지역명",
col_wrap=4)
# box plot
sns.boxplot(data=df_last, x="연도", y="평당분양가격")
# hue 사용해서 전용면적별로
plt.figure(figsize=(12, 3))
sns.boxplot(data=df_last, x="연도", y="평당분양가격", hue="전용면적")
# violin plot (box plot에 밀도추정 값을 같이 볼수 있다)
sns.violinplot(data=df_last, x="연도", y="평당분양가격")
# ### lmplot & swarmplot
# 연도별 평당분양가격을 lmplot으로
# One can find many ways to handle categorical data. Some of them categorical data are,
# 1. <span style="color: blue;">**Nominal data**</span> --> data are not in any order --> <span style="color: green;">**OneHotEncoder**</span> is used in this case
# 2. <span style="color: blue;">**Ordinal data**</span> --> data are in order --> <span style="color: green;">**LabelEncoder**</span> is used in this case
# In[24]:
df_train["Airline"].value_counts()
# In[25]:
# Airline vs Price
sns.catplot(x = "Airline", y = "Price", data = df_train.sort_values("Price", ascending = False), kind = "boxen", height = 6, aspect = 2)
plt.show()
# In[26]:
# From the above graph it is clear that jet airways has the maximun price
# And also apart from the jet airways almost all the other airlines has the same median
# In[27]:
# As Airline is a Nominal Category we will perform one hot encoding
from sklearn.preprocessing import MinMaxScaler, LabelEncoder
#Loading the data
data = pd.read_csv(path)
data['Rating'].plot(kind='hist')
plt.show()
#Code starts here
data = data[data['Rating'] <= 5]
data['Rating'].plot(kind='hist')
plt.show()
nulls = data.isnull().sum()
data.dropna(inplace=True)
plt.figure(figsize=(10, 10))
cat = sns.catplot(x='Category', y='Rating', kind='box', data=data, height=8)
cat.set_xticklabels(rotation=90)
plt.title("Rating vs Category boxplot", size=20)
data['Genres'] = data['Genres'].str.split(";", expand=True)[0]
mean_rating = data.groupby('Genres')['Rating'].mean()
max_rating = data.groupby('Genres')['Rating'].max()
min_rating = data.groupby('Genres')['Rating'].min()
rating_data = {
'mean_rating': mean_rating,
'max_rating': max_rating,
'min_rating': min_rating
}
rating_data_df = pd.DataFrame(
rating_data, columns=['mean_rating', 'max_rating', 'min_rating'])
# -*- coding: utf-8 -*-
"""
Created on Sun Jan 13 14:06:15 2019
@author: Eldrich
"""
import os
import pandas as pd
import seaborn as sns
from matplotlib import pyplot as plt
os.chdir('E:\\University\\data\\Class Work\\Thesis - KTH\\Extra Work\\Papers\\Journal Paper\\V2')
df=pd.read_csv('Delay_Table.csv')
sns.set(style='whitegrid')
#sns.set(rc={'figure.figsize':(11.7,8.27)})
#g=sns.barplot(x=df['Run number'], y=df.Delay, hue=df.Pulse, data=df, color='gray')
#g.despine(left=True)
g=sns.catplot('Run number', 'Delay', 'Pulse', data=df, kind='bar', color='gray', legend=False, aspect=2, size=3)
g.set_ylabels('Delay (s)')
g.set_xlabels(label='')
g.set(ylim=(0,0.5),xlim=g.ax.axes.get_xbound())
sns.despine()
plt.plot([-0.54, 2.5400000000000005], [0.3435263157894737,0.3435263157894737], linewidth=1, linestyle='--', color='k')
#sns.set_context("paper")
plt.annotate('Average',xy=(1,0), xytext=(2,0.35))
plt.tight_layout()
# Graphing the differences of infrastructure reports:
# merge the two datasets and assign column to distinguish them
infra_vergleich = pd.concat([
vor_profilsumme_infra.assign(Monat='Vormonat'),
neu_profilsumme_infra.assign(Monat='aktueller Monat')
])
infra_vergleich.reset_index(inplace=True)
fig, ax = plt.subplots(figsize=(15, 15))
ax = sns.catplot(x="Profil",
y="Preis 1 (€)",
hue="Monat",
data=infra_vergleich,
height=8,
kind="bar",
palette="muted",
legend=False)
ax.set_title = 'Infrastructure Vergleich zu Vormonat'
ax.set_ylabels("Umsatz in EURO")
ax.set_xlabels("Profile")
plt.legend(loc="upper left")
fig.tight_layout()
plt.savefig("diff_infra.png")
plt.clf()
# Graphing the differences of extern reports:
