import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
import numpy as np
datafile = open("all_data.txt", "r").readlines()
all_data = np.zeros((4, len(datafile)))
for i, line in enumerate(datafile):
total = line.split(" ")
all_data[:, i] = total
df = pd.DataFrame(data=all_data)
sns.plot(data=df)
ax.set_xlabel('something', fontsize=16)
ax.set_title('I am title', fontsize=20)
ax.set_xlim([0.,50.])
ax.set_ylim([0, 100])
plt.show()
#<><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>
## Lineplot
# Seaborn
ax = sns.pointplot(x=x, y=y, hue='gender',data=df, marker=['o','x'], linestyle=['-','--'], join=True,
color='#bb3f3f', order=['Dinner','Lunch'], estimator=np.median, capsize=.2)
# Matplotlib
ax = sns.plot(x,y)
ax.set_ylabel('y', fontsize=16)
ax.set_xlabel('x', fontsize=16)
ax.set_title('title', fontsize=20)
ax.set_xlim([0.,50.])
ax.set_ylim([0, 100])
plt.show()
#<><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>
## Heatmap (Correlate)
# Seaborn
mean_corr = bcw_mean[[x for x in bcw_mean.columns if x not in 'id']].corr()
# coding=utf-8
import seaborn as sns
import matplotlib as mt
import pandas as pd
import numpy as np
import sys
if __name__ == "__main__":
# 一维线性回归数据
data1 = pd.read_csv('resources/data1.txt', header=None, names=['x', 'y'])
print(data1)
# 二维线性回归数据
data2 = pd.read_csv('resources/data2.txt',
header=None,
names=['x', 'y', 'z'])
print(data2)
sns.jointplot(x='x', y='y', data=data1)
sns.jointplot(x='x', y='y', data=data2)
sns.plot([1, 2, 3, 4])
sns.plt.show()
def plot(data, x_name='x', y_name='y'):
sns.jointplot(x=x_name, y=y_name, data=data)
sns.plt.show()
def main():
# Initialisieren aller Pygame-Module und
# Fenster erstellen (wir bekommen eine Surface, die den Bildschirm repräsentiert).
pygame.init()
agent = DQNAgent()
counter_games = 0
record = 0
while counter_games < 150:
screen = pygame.display.set_mode((800, 600))
# Titel des Fensters setzen, Mauszeiger nicht verstecken und Tastendrücke wiederholt senden.
pygame.display.set_caption("Pygame-Tutorial: Animation")
pygame.mouse.set_visible(1)
pygame.key.set_repeat(1, 30)
pygame.font.init() # you have to call this at the start,
myfont = pygame.font.SysFont('Comic Sans MS', 30)
# Clock-Objekt erstellen, das wir benötigen, um die Framerate zu begrenzen.
clock = pygame.time.Clock()
# Wir erstellen eine Tilemap.
map = Tilemap.Tilemap()
event = AutoInput.AutoInput()
# Die Schleife, und damit unser Spiel, läuft solange running == True.
running = True
max_steps_reached = False
max_steps = 100
step = 0
max_score = map.player.pos_x
max_score_evolution = []
while running and not max_steps_reached:
agent.epsilon = 80 - counter_games
#get old state
state_old = agent.get_state(map)
map.player.pos_x_old = map.player.pos_x
#perform random actions based on agent.epsilon, or choose the action
if randint(0, 200) < agent.epsilon:
final_move = to_categorical(randint(0, 2), num_classes=3)
else:
# predict action based on the old state
prediction = agent.model.predict(state_old.reshape((1, 7)))
final_move = to_categorical(np.argmax(prediction[0]),
num_classes=3)
# Framerate auf 30 Frames pro Sekunde beschränken.
# Pygame wartet, falls das Programm schneller läuft.
clock.tick(30)
# screen Surface mit Schwarz (RGB = 0, 0, 0) füllen.
screen.fill((198, 209, 255))
map.handle_input(final_move)
#continue jump animation after
if map.player.isjump:
map.player.jump()
# Die Tilemap auf die screen-Surface rendern.
map.render(screen)
textsurface = myfont.render(
"Game " + str(counter_games) + " Step " + str(step) +
" Max Score " + str(max_score), False, (0, 0, 0))
screen.blit(textsurface, (50, 50))
#Print Hindernis onto map and check if there should be a new one
if not map.isThereHindernis:
map.createNewHindernis()
map.isThereHindernis = True
map.hindernis.move()
map.hindernis.render(screen)
map.checkHindernisOnMap()
state_new = agent.get_state(map)
crash = map.collisionDetection()
#set treward for the new state
reward = agent.set_reward(map.player, crash)
#train short memory base on the new action and state
agent.train_short_memory(state_old, final_move, reward, state_new,
running)
# Inhalt von screen anzeigen
pygame.display.flip()
if map.player.pos_x > max_score:
max_score = map.player.pos_x
step += 1
if step >= max_steps:
max_steps_reached = True
max_score_evolution.append(max_score)
agent.remember(state_old, final_move, reward, state_new, running)
#record = get_record(map.player.pos_x, record)
#if display_option:
# #display(player1, food1, game, record)
# pygame.time.wait(speed)
agent.replay_new(agent.memory)
counter_games += 1
agent.model.save_weights('weights.hdf5')
sns.plot(max_score_evolution)
# ##### The segmentation with 4 clusters looks good as it has segmented the high frequent and high monetary customer with less recency time
# In[540]:
retail_RFM_melt = pd.melt(retail_RFM_k4.reset_index(),
id_vars=['CustomerID', 'Cluster'],
value_vars=['Recency', 'Frequency', 'MonetaryValue'],
var_name='Attribute',
value_name='Value')
retail_RFM_melt = pd.DataFrame(retail_RFM_melt)
# In[ ]:
plt.title('Snake Plot for standardized variables')
sns.plot(x="Attribute", y="Value", hue='Cluster', data=retail_RFM_melt)
# In[542]:
cluster_avg = retail_RFM_k4.groupby(['Cluster']).mean()
population_avg = retail_RFM.mean()
relative_imp = cluster_avg / population_avg - 1
# In[543]:
#Relative distance from each parameter
plt.figure(figsize=(10, 2))
sns.heatmap(relative_imp, annot=True, cmap='RdYlGn')
# ### 4. Conclusion
# In[31]:
df.set_index('line').grouby(plot()
# In[30]:
df.plot(x='sample', y='number_of_rhizoids', subplots=True)
# In[25]:
import seaborn as sns
sns.plot(df)
# In[8]:
from numpy.random import choice
# In[9]:
get_ipython().magic('pinfo choice')
# In[4]:
### Using an index, which is what I would do.
ax.set_xlabel('something', fontsize=16)
ax.set_title('I am title', fontsize=20)
ax.set_xlim([0.,50.])
ax.set_ylim([0, 100])
plt.show()
#<><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>
## Lineplot
# Seaborn
ax = sns.pointplot(x=x, y=y, hue='gender',data=df, marker=['o','x'], linestyle=['-','--'], join=True,
color='#bb3f3f', order=['Dinner','Lunch'], estimator=np.median, capsize=.2)
# Matplotlib
ax = sns.plot(x,y)
ax.set_ylabel('y', fontsize=16)
ax.set_xlabel('x', fontsize=16)
ax.set_title('title', fontsize=20)
ax.set_xlim([0.,50.])
ax.set_ylim([0, 100])
plt.show()
#<><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>
## Heatmap (Correlate)
# Seaborn
corr = df.corr
ax = sns.heatmap(corr, vmin=0, vmin=1, annot=False, square=True, linewidths=.5, cmap="YlGnBu", cbar=True)
# In[3]:
path = '../../resources/SimpleRegressionDataSet/simple_regression.csv'
# In[4]:
data_1 = pd.read_csv(path, header=None)
# In[5]:
data_1.head(n=10)
# In[12]:
data_1.columns = ['population', 'profit']
data_1.head(n=2)
# In[ ]:
import seaborn as sns
sns.set(color_codes=True)
tips = sns.load_dataset("tips")
plot = sns.lmplot(x="population", y="profit", data=data_1)
# In[ ]:
plot = sns.plot(x="population", y="profit", data=data_1)
# In[ ]:
from sklearn.model_selection import train_test_split
from sklearn.neighbors import KNeighborsClassifier
from sklearn.metrics import classification_report, confusion_matrix
'''
PROJECT TO DETERMINE BEST VALUE FOR KNN
-given artificial dataset
-code can be used for parameter tuning
'''
df = pd.read_csv('KNN_Project_Data')
df.head()
sns.pairplot(df, hue='TARGET CLASS', palette='coolwarm')
sns.plot()
scaler = StandardScaler()
scaler.fit(df.drop('TARGET CLASS', axis=1))
scaled_features = scaler.transform(df.drop('TARGET CLASS', axis=1))
df_feat = pd.DataFrame(scaled_features, columns=df.columns[:-1])
df_feat.head()
X_train, X_test, y_train, y_test = train_test_split(scaled_features,
df['TARGET CLASS'],
test_size=0.30)
knn = KNeighborsClassifier(n_neighbors=1)
knn.fit(X_train, y_train)
orient='v')
OARGBvCHM.set(xlabel='Feature groups', ylabel='UA (%)')
loc, labels = plt.xticks()
sns.despine()
OARGBvCHMFig = OARGBvCHM.get_figure()
OARGBvCHMFig.savefig(
r"C:\Users\Study\OneDrive - University of the Sunshine Coast\Documents\Thesis\Manuscript\Figures\Table5_ModelPerf\UA-Pine-RGBvCHM-Delta.pdf",
bbox_inches='tight',
dpi=300)
# %%
plt.figure(figsize=(13, 6))
sns.set_theme(style="ticks")
OARGBvCHM = sns.plot(accValues['UA-Lan round'],
palette='colorblind',
order=['RGB', 'TEX', 'RGB-TEX'],
capsize=0.025,
orient='v')
OARGBvCHM.set(xlabel='Feature groups', ylabel='UA (%)')
loc, labels = plt.xticks()
sns.despine()
OARGBvCHMFig = OARGBvCHM.get_figure()
OARGBvCHMFig.savefig(
r"C:\Users\Study\OneDrive - University of the Sunshine Coast\Documents\Thesis\Manuscript\Figures\Table5_ModelPerf\UA-Pine-RGBvTEX-Delta.pdf",
bbox_inches='tight',
dpi=300)
# %%
g = sns.FacetGrid(accValues,
col=['Lantana', 'Slash pine'],
col_wrap=2,
data_df = df[['GeoID', 'Date'] + sir_cols]
us_data = data_df[data_df["GeoID"] == "United States__nan"]
# In[ ]:
# In[2]:
sns.plot(us_data["Date"], us_data["Infected"])
# In[ ]:
odeint()
# In[10]:
# routine to forecast the next X days given an SIR model and params
def forecast_future(S, I, R, beta, gamma, days):
y_i = S, I, R
#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
Created on Wed Jan 24 10:26:22 2018
@author: peternagy96
"""
import pickle
from matplotlib import pyplot as plt
import seaborn as sb
results = pickle.load(open('final_results_0.332.p', 'rb'))
plt.figure()
sb.set_style('darkgrid')
pred_hist = sb.load_dataset(results['pred_hist'])
sb.plot(pred_hist)
plt.show()