Ask Ghassem - Recent activity in Python

Which code has best runtime and why?(the one commented or the other one)

Fri, 02 Sep 2022 14:39:49 +0000

# for key, value in dict.items():
            #     if value >= long:
            #         long = value
            #         long_name = key
            #     if value < short:
            #         short = value
            #         short_name = key
        long = max(dict.values())
        long_name = max(dict, key=dict.get)
        short = min(dict.values())
        short_name = min(dict, key=dict.get)

Answered: how to output f1-score instead of accuracy

Sat, 02 Apr 2022 14:32:22 +0000

from sklearn.metrics import f1_score
f1_score(data_test,target_test)

I cannot get this code to work. please help.

Mon, 21 Mar 2022 05:59:53 +0000

from keras.models import Sequential
from keras.layers import Dense
from keras.layers import LSTM
from sklearn.model_selection import train_test_split

model = Sequential()
model.add(LSTM( 10, input_shape=(1, 1)))
model.add(Dense(1, activation="linear"))
model.compile(loss="mse", optimizer="adam")

X, y = get_data()

X_train, X_test, y_train, y_test = train_test_split(x, y, test_size=0.2, random_state=1)
X_train_2, X_val, y_train_2, y_val = train_test_split(X_train, y_train, test_size=0.25, random_state=1)

model.fit(X_train, y_train, epochs=800, validation_data=(X_val, y_val), shuffle=False)

html, body, table, thead, input, textarea, select {color: #bab5ab!important; background: #35393b;} input[type="text"], textarea, select {color: #bab5ab!important; background: #35393b;} [data-darksite-inline-background-image-gradient] {background: linear-gradient(rgba(0, 0, 0, 0.5), rgba(0, 0, 0, 0.5))!important; -webkit-background-size: cover!important; -moz-background-size: cover!important; -o-background-size: cover!important; background-size: cover!important;} [data-darksite-force-inline-background] * {background-color: rgba(0,0,0,0.7)!important;} [data-darksite-inline-background] {background-color: rgba(0,0,0,0.7)!important;} [data-darksite-inline-color] {color: #fff!important;} [data-darksite-inline-background-image] {background-image: linear-gradient(rgba(0,0,0,0.3), rgba(0,0,0,0.3))!important}

How to print confusion matrix if I am using stratifiedkfold method?

Thu, 06 Aug 2020 21:41:19 +0000

I am facing the error after importing torch module in python. How can I solve it? The error link is given below

Fri, 31 Jul 2020 14:56:42 +0000

https://ibb.co/MgXvrkL

Edited: Suggestion For Model

Thu, 30 Jul 2020 22:11:03 +0000

How to split into train and test using PKL file?

Thu, 30 Jul 2020 22:08:47 +0000

Answered: What are some best features of Python3 in comparison to Python2?

Mon, 20 May 2019 16:19:34 +0000

There is a list of great features many do not use yet. Here is the list of some of them:

f-strings (3.6+)
Pathlib (3.4+)
Type hinting (3.5+)
Enumerations (3.4+)
Built-in LRU cache (3.2+)
Extended iterable unpacking (3.0+)
Data classes (3.7+)
Implicit namespace packages (3.3+)

This article shows some examples, and you can take the notebook for your test from this GitHub page.

Answered: What are the best resources for studying Python?

Mon, 20 May 2019 11:57:40 +0000

Hey! There are a ton of good websites that provide both in-depth theoretical knowledge on python, as well as easy to follow practical projects. Here are just a few sites this that:

https://www.learnpython.org/

https://www.codecademy.com/learn/learn-python

Feel free to check out any of these sites!

Hope this helps.

Commented: Who can write a recursive Python program for Fibonacci sequence 0,1,1,2,3,5,8,...?

Fri, 22 Mar 2019 19:39:16 +0000

You should correct your code as follows:

sum = fibonacci(n-1) + fibonacci(n-2)

Answered: How do I create comments in my code?

Fri, 02 Nov 2018 14:27:35 +0000

Just use # to add comments in your code. Below is an example of how you can add the comments in python

import numpy as np
# importing numpy library

Reshown: If s='Sammy' what is the output of s[2:]?

Wed, 31 Oct 2018 17:48:36 +0000

Retagged: Can you share with us how to download ipynb file as pdf through latex?

Wed, 31 Oct 2018 17:43:24 +0000

Retagged: How to force Jupyter Notebook to print all the variables in each cell, not just the last one?

Wed, 31 Oct 2018 17:43:02 +0000

Answered: What are Decorators?

Mon, 29 Oct 2018 23:00:29 +0000

Technically, decorators are just functions (or classes), that take a callable object as an argument, and return ananalogous object with the decoration.

Answered: What is a Magic class methods ?

Mon, 29 Oct 2018 22:58:07 +0000

Magic class methods (not a complete list) Magic methods (which begin and end with double underscore) add functionality to your classes consistent with the broader language.

Answer selected: Hey, anyone knows how to save workspace in jupyter python? I would like to save the variables and functions, etc.

Wed, 24 Oct 2018 21:28:38 +0000

I think you need to use dill for this purpose.
"dill extends python’s pickle module for serializing and de-serializing python objects to the majority of the built-in python types. Serialization is the process of converting an object to a byte stream, and the inverse of which is converting a byte stream back to on python object hierarchy."

conda install -c anaconda dill

To save the Jupyter Notebook session:

import dill
dill.dump_session('notebook_session.db')

To restore the session:

import dill
dill.load_session('notebook_session.db')

Answered: ploting the xticks to be datatime

Sun, 21 Oct 2018 22:18:58 +0000

Please take a look at the following code. I converted the data to time-series, and then showed how to resample it with a different frequency such 30 minutes. For 4 hours, change the first resample parameter to 4H.

You can run the code here.

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import io

data = """2012-01-01 00:15:00	3.807107	22.759602	00:15:00
2012-01-01 00:30:00	5.076142	22.759602	00:30:00
2012-01-01 00:45:00	3.807107	22.759602	00:45:00
2012-01-01 01:00:00	3.807107	22.759602	01:00:00
2012-01-01 01:15:00	5.076142	22.048364	01:15:00"""

df = pd.read_csv(io.StringIO(data), names =["date","time", "MT1", "MT2", "hour"], sep ="\s+")
df.drop(['hour'], axis = 1, inplace = True) #removing redundant hour


#Creating timestamp column by merging date and time
df["timestamp"] = (df["date"].map(str) + " " + df["time"])
df.index = pd.to_datetime(df["timestamp"])
df.drop(['date','timestamp'], axis = 1, inplace = True)
df2 = df.resample('30Min' ,base=0).last()
print("\ndf\n", df)
print("\ndf2\n",df2)

f, (ax1, ax2) = plt.subplots(1, 2,figsize=(20,5) )
ax1.plot(df['time'], df['MT1'],'ro--')
ax2.plot(df2['time'], df2['MT1'],'ro--')
ax1.plot(df['time'], df['MT2'],'bo--')
ax2.plot(df2['time'], df2['MT2'],'bo--')
ax1.grid(True)
ax2.grid(True)
ax1.legend()
ax2.legend()
plt.setp(ax1.xaxis.get_majorticklabels(), rotation=70, ha="right")
plt.setp(ax2.xaxis.get_majorticklabels(), rotation=70, ha="right")

plt.show()
plt.savefig('graph.png')

Answered: How can you tell a Python library is a good one and useful?

Fri, 19 Oct 2018 03:48:16 +0000

One of the features of GitHub is stars developers give to a library and number of forks (copies). If the number of stars is high, it seems you are seeing a great library.

Commented: plotting categorical radio

Wed, 17 Oct 2018 20:59:44 +0000

Thanks for sharing. Two notes:
1) Please use code section on the toolbar when sharing your codes.
2) That would be great to provide executable version on https://repl.it

Answered: In a dataframe of a size (5,10) with random values, how to have exactly 5 NAN values in each row?

Wed, 17 Oct 2018 20:48:31 +0000

You can create a random DataFrame, then generate random column numbers and replace those columns for each row with NaN. You can run it here.

import pandas as pd
import numpy as np
import random

rows = 5
cols = 10 
nan_n = 5

df = pd.DataFrame(np.random.randint(0,100,size=(rows, cols)))

print(df)
for row in range(0,rows):
    nan_cols = np.asarray(random.sample(range(0, 10), nan_n))
    df.at[row, nan_cols] = np.nan

print(df)

My own transformation for dropping feature doesn't work

Wed, 17 Oct 2018 00:52:29 +0000

I made my own transformation to drop some of the features not required in the dataframe but I can't seem to get it working. Thoughts ?

https://repl.it/@Neo_sauga/HarmfulNearApplicationstack

Answered: How to fetch the data from a url using Python?

Mon, 15 Oct 2018 20:33:57 +0000

Use the following code, and do not forget the space in URLs should be converted to %20. You can run the code here.

import os
import urllib.request

def fetch_credit_card_data(url, directory, file_name):
    if not os.path.isdir(directory):
        os.makedirs(directory)
    urllib.request.urlretrieve(url, directory+'/'+file_name)


url = "https://archive.ics.uci.edu/ml/machine-learning-databases/00350/default%20of%20credit%20card%20clients.xls"
directory = 'directory'
file_name = 'credit_card.xls'

fetch_credit_card_data(url, directory, file_name)

Answered: How to update "conda" version itself?

Wed, 10 Oct 2018 16:58:20 +0000

Please use the following command in your Terminal (Linux/Mac) or Command (Windows):

conda update -n base conda -c anaconda

Answered: Access denied while updating Anaconda?

Sun, 30 Sep 2018 04:54:44 +0000

Do you have administrative privileges to update Anaconda? You can try Run as administrator this time when you run cmd (figure below: Right click on Command Prompt -> Run as administrator) to see if it gives you error again.

Answered: What is the difference between single quotation (' ') and double ("") quotation in Python?

Fri, 28 Sep 2018 14:37:44 +0000

A string as a sequence of characters not intended to have numeric value. In Python, such sequence of characters is included inside single or double quotes. As far as language syntax is concerned, there is no difference in single or double quoted string. Both representations can be used interchangeably. However, if either single or double quote is a part of the string itself, then the string must be placed in double or single quotes respectively.

For example a text Hello “Python” is to be stored as a string, then one should use:

str1='Hello "Python"'

On the other hand, if string should resemble Hello ‘Python’ then it should be declared as follows:

Str2="Hello 'Python'"

source

Answer selected: What do L1-norm and L2-norm mean as it applies to Norm?

Wed, 26 Sep 2018 15:41:08 +0000

This article describes both of them in more details, but in short, we define $|| x ||_p$ as p-norm.

p-norm: Given $x$, a vector with $i$ components, a p-norm is defined as:

$|| x ||_p = \left(\sum_i |x_i|^p\right)^{1/p}$

The simplest norm conceptually is Euclidean distance. This is what we typically think of as distance between two points in space:

2-norm (Euclidean distance ro L2-norm)

$|| x ||_2 = \sqrt{\left(\sum_i x_i^2\right)} = \sqrt{x_1^2 + x_2^2 + \ldots + x_i^2}$

1-norm (Manhattan distance or L1-norm)

$|| x ||_1 = \sum_i |x_i| = |x_1| + |x_2| + \ldots + |x_i|$

Python implementation

If you want to calculate them in Python, numpy.linalg.norm provides functions for it. In addition, you can use Distance computations (scipy.spatial.distance) library. For example, for the Euclidean distance between two vectors, $a$ and $b$, you can use the following code:

from scipy.spatial import distance
distance.euclidean(a, b)

and if you want to calculate the Manhattan distance, you can use the following code:

from scipy.spatial import distance
distance.cityblock(a, b)

How did I know this time it is called cityblock? I read the documentation and looked up for Manhattan distance.

Answered: Can someone please post the video tutorial for Pandas, Numpy and Matplotlib ?

Wed, 19 Sep 2018 18:51:30 +0000

There are many good tutorials out there. For Lynda.com (here you can see you to use it for free), I recommend the following tutorials.

NumPy Data Science Essential Training (4 hours)
Pandas for Data Science (2 hours)
Matplotlib (20 minutes)

You can also take all of them all to together in a shorter series in Data Science Foundations: Python Scientific Stack (3.5 hours) course.