mlquiz 0.1.0__py3-none-any.whl

mlquiz/cards.csv

@@ -0,0 +1,54 @@
+category,type,question,correctanswer,optiona,optionb,optionc,optiond
+basics,mcq,what does ai stand for?,artificial intelligence,artificial intelligence,automated information,advanced integration,algorithmic input
+basics,fill,what type of ai exists today? (narrow or general),narrow,,,,
+basics,mcq,what is the term for ai that surpasses human intelligence in every domain?,asi,agi,narrow ai,asi,machine learning
+basics,mcq,what is the first step of the ai development pipeline?,collect data,deploy,train model,collect data,validate
+basics,mcq,which year did the deep learning revolution start with alexnet?,2012,2012,2005,2018,1997
+basics,mcq,what is the process of adjusting model parameters to minimize error called?,training,testing,deploying,validating,training
+basics,fill,what does agi stand for?,artificial general intelligence,,,,
+ethics,mcq,what amplifies bias in an ai system?,data,the algorithm,the hardware,data,the user
+ethics,fill,what is the term for when biased ai decisions create biased future data?,feedback loop,,,,
+ethics,mcq,which fairness definition requires equal positive prediction rates across groups?,demographic parity,demographic parity,equalized odds,calibration,individual fairness
+ethics,mcq,is it mathematically possible to satisfy demographic parity and equalized odds at the same time if base rates differ?,no,yes,only with deep learning,only with small data,no
+ethics,mcq,adding calibrated noise to data to protect privacy is called what?,differential privacy,data masking,anonymization,differential privacy,encryption
+ethics,fill,what type of models are least interpretable but often most accurate?,deep neural networks,,,,
+ethics,mcq,what is it called when an ai system makes a wrong prediction that reinforces negative stereotypes?,feedback loop,feedback loop,distribution shift,data leakage,bias mitigation
+data,mcq,what scaling method transforms data to have a mean of 0 and a standard deviation of 1?,standard scaling,min-max scaling,pca scaling,log scaling,standard scaling
+data,mcq,using readmission date to predict hospital readmission is an example of what?,target leakage,feature leakage,target leakage,concept drift,survivorship bias
+data,fill,what type of shift occurs when the relationship between inputs and outputs changes over time?,concept drift,,,,
+data,mcq,a dataset of only successful startups is an example of what?,survivorship bias,survivorship bias,selection bias,non-response bias,convenience sampling
+data,mcq,converting categorical variables like color into binary columns is called what?,one-hot encoding,label encoding,scaling,binning,one-hot encoding
+data,fill,what split is used for tuning hyperparameters?,validation,,,,
+data,mcq,what type of leakage occurs when test data statistics influence training preprocessing?,feature leakage,target leakage,feature leakage,concept drift,label shift
+data,mcq,a spam filter trained on 5% spam deployed when spam is 40% is an example of what?,label shift,covariate shift,label shift,concept drift,feature leakage
+supervised,mcq,what type of supervised learning predicts continuous values?,regression,classification,clustering,regression,generative
+supervised,fill,what function maps any real number to a value between 0 and 1?,sigmoid,,,,
+supervised,mcq,which regularization method tends to make some parameters exactly zero?,l1 regularization,l2 regularization,dropout,l1 regularization,early stopping
+supervised,mcq,what algorithm finds the hyperplane that maximizes the margin between classes?,support vector machine,support vector machine,decision tree,logistic regression,naive bayes
+supervised,fill,what does svm stand for?,support vector machine,,,,
+supervised,mcq,what is the loss function used for linear regression called?,mean squared error,mean absolute error,cross entropy,mean squared error,hinge loss
+supervised,fill,what is the maximum margin classifier also known as? (abbreviation),svm,,,,
+supervised,mcq,which algorithm uses entropy and information gain to split data?,decision tree,linear regression,svm,naive bayes,decision tree
+supervised,fill,what is the name of the algorithm used to train neural networks by calculating gradients backwards?,backpropagation,,,,
+supervised,mcq,what happens when a model is too complex and memorizes training data?,overfitting,overfitting,underfitting,regularization,bias
+supervised,fill,what is the probabilistic classifier based on bayes theorem called?,naive bayes,,,,
+basics,mcq,what subfield of ai focuses on processing images?,computer vision,natural language processing,computer vision,neural networks,robotics
+basics,fill,what type of ai is a chess engine,narrow,,,,
+basics,mcq,what does the term deep in deep learning refer to?,many layers,many layers,complex math,big data,fast gpus
+ethics,fill,what is the trade off called when a model is highly accurate but hard to explain?,interpretability,,,,
+data,mcq,what is the usual train test split ratio?,70-15-15,50-25-25,80-10-10,100-0-0,70-15-15
+data,fill,what typeof shift occurs when the input distribution changes but the relationship stays the same?,covariate shift,,,,
+supervised,mcq,what is the main goal of gradient descent?,minimize the loss function,minimize the loss function,maximize the score,increase variance,reduce features
+supervised,fill,what is the term for a model that is too simple and can't capure the pattern in the data itself?,underfitting,,,,
+supervised,mcq,which algorithm combines multiple desicion trees?,random forest,logistic regression,vector machines,random forest,naive bayes
+supervised,fill,what is the name of the parameter that controls the step size in gradient descent?,learning rate,,,,
+supervised,fill,what is the term for the penalty added to a loss function to prevent overfitting?,regularization,,,,
+supervised,fill,what type of loss function does svm use?,hinge loss,,,,
+evaluation,mcq,what is the mean of precision and recall called?,f1 score,accuracy,f1 score,auc,mean squared error
+evaluation,fill,what metric measures the fraction of actual positives that were correctly identified?,recall,,,,
+evaluation,mcq,what is it  called when a model performs well on training data but not on test data?,overfitting,overfitting,underfitting,bias,variance
+evaluation,mcq,what does high bias indicate about a model?,underfitting,overfitting,underfitting,high variance,data leakage
+evaluation,mcq,if the model is really unbalanced which metric is misleading?,accuracy,recall,presicion,accuracy,f1 score
+kernels,mcq,what allows models to work in high dimensional space without computing the coordinates?,kernel trick,kernel trick,gradient descent,feature scaling,overfitting
+kernels,fill,what kernel is also known as the rbf kernel?,Gaussian,,,,
+kernels,mcq,which property must a kernel matrix satisfy to be valid?,symmetric,diagonal,orthogonal,symmetric,equal

mlquiz/main.py

@@ -0,0 +1,108 @@
+import pandas as pd
+import os
+
+def clear():
+    os.system('cls' if os.name =='nt' else 'clear')
+
+def run_quiz(df):
+    clear()
+    #df = pd.read_csv(filename)
+    score = 0
+    wrong = []
+
+    for index, row in df.iterrows():
+        print(f"question {index+1} {row['question']}")
+        if row['type'] == 'mcq':
+            print(f"a) {row['optiona']}")
+            print(f"b) {row['optionb']}")
+            print(f"c) {row['optionc']}")
+            print(f"d) {row['optiond']}")
+            
+        answer = input("\nyour answer ")
+        answer = answer.strip().lower()
+
+        if row['type'] == 'mcq':
+            if answer == 'a':
+                answer = str(row['optiona']).strip().lower()
+            elif answer == 'b':
+                answer = str(row['optionb']).strip().lower()
+            elif answer == 'c':
+                answer = str(row['optionc']).strip().lower()
+            elif answer == 'd':
+                answer = str(row['optiond']).strip().lower()
+                
+        if answer == str(row['correctanswer']).strip().lower():
+            print("correct")
+            score += 1
+        else:
+            print(f"wrong answer, correct answer is {row['correctanswer']}")
+            wrong.append(row)
+
+        input("\npress enter for next question")
+        clear()
+
+    print(f"quiz is over: {score}/{len(df)}")
+
+    if len(wrong) > 0:
+        wrong_df = pd.DataFrame(wrong)
+        wrong_df.to_csv('mistakes.csv', index=False)
+        print(f"you got {len(wrong)} questions wrong, saved as mistakes.csv")
+    else:
+        if os.path.exists('mistakes.csv'):
+            os.remove('mistakes.csv')
+        print("perfect score! no mistakes")
+
+
+def main():
+    current_dir=os.path.dirname(os.path.abspath(__file__))
+    csv_path=os.path.join(current_dir,'cards.csv')
+
+    currentdir = os.path.dirname(os.path.abspath(__file__))
+    csv_path= os.path.join(currentdir, 'cards.csv')
+    questions=pd.read_csv(csv_path)
+    while True:
+        clear()
+        print("welcome to mlquiz!\n")
+        print("1. take full quiz")
+        print("2. choose a category")
+        
+        if os.path.exists('mistakes.csv'):
+            print("3. study mistakes")
+            
+        print("4. quit")
+
+        choice = input("\nselect an option: ")
+
+        if choice == '1':
+            run_quiz(questions)
+            input("\npress enter to return to menu")
+        elif choice =='2':
+            categories=questions['category'].unique()
+            print("\ncategories:")
+            for i in range(len(categories)):
+                print(f"{i+1}. {categories[i]}")
+            choice=input("\npick a number ")
+
+            try:
+                index=int(choice)-1
+                if 0 <= index < len(categories):
+                    selected=categories[index]
+                    filtered = questions[questions['category']==selected]
+                    run_quiz(filtered)
+                    input("\npress enter to return to menu")
+                else:
+                    print("invalid number")
+                    input("press enter to reenter number")
+            except ValueError:
+                print("please type a number")
+                input("press enter to try again")
+        
+        elif choice == '3' and os.path.exists('mistakes.csv'):
+            mistakes_df=pd.read_csv('mistakes.csv')
+            run_quiz(mistakes_df)
+            input("\npress enter to return to menu")
+        elif choice=='4':
+            break
+
+if __name__ == "__main__":
+    main()

mlquiz-0.1.0.dist-info/METADATA

@@ -0,0 +1,6 @@
+Metadata-Version: 2.4
+Name: mlquiz
+Version: 0.1.0
+Summary: A CLI quiz for machine learning
+Requires-Python: >=3.8
+Requires-Dist: pandas

mlquiz-0.1.0.dist-info/RECORD

@@ -0,0 +1,8 @@
+mlquiz/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+mlquiz/cards.csv,sha256=s_q2qaifQTpDzG4PKiGp7yx6NYYlXpx7yai0TjFuKUE,6869
+mlquiz/main.py,sha256=a9RI78rcu1mj0cScGT6NNk_zLvMIQ9AHdwt3cgZDqjg,3612
+mlquiz-0.1.0.dist-info/METADATA,sha256=KUo1tkiTV8BzMNsi3hPrU12nOmq7mj1DPvphhTKF5Kg,142
+mlquiz-0.1.0.dist-info/WHEEL,sha256=mOdXR9C3wMVov7cSY8PY-XCzo3SmmWCHvGQr5csJoao,91
+mlquiz-0.1.0.dist-info/entry_points.txt,sha256=Ck39Yr8YwRzJITXq0Mvt4J-E07O_t8WFF0PP_zQw4y4,44
+mlquiz-0.1.0.dist-info/top_level.txt,sha256=dFNPDfMSifjhBGxvOccaeq_zow4KpjKM3YG0oHFFCAc,7
+mlquiz-0.1.0.dist-info/RECORD,,

mlquiz-0.1.0.dist-info/WHEEL

@@ -0,0 +1,5 @@
+Wheel-Version: 1.0
+Generator: setuptools (84.2.1)
+Root-Is-Purelib: true
+Tag: py3-none-any
+

mlquiz-0.1.0.dist-info/entry_points.txt

@@ -0,0 +1,2 @@
+[console_scripts]
+mlquiz = mlquiz.main:main

mlquiz-0.1.0.dist-info/top_level.txt

@@ -0,0 +1 @@
+mlquiz