final logcosh leaderboard

Assignment1/data/plot1/plot.py

+import matplotlib.pyplot as plt
+import csv
+
+x = []
+y = []
+i=1
+
+with open('logcosh.log','r') as csvfile:
+    for row in csvfile:
+        x.append(i)
+        i+=1
+        y.append(float(row[:-1]))
+
+
+
+plt.plot(x,y, label='LOGCOSH')
+plt.xlabel('epoch')
+plt.ylabel('Error')
+
+
+x = []
+y = []
+i=1
+
+
+with open('mae.log','r') as csvfile:
+    for row in csvfile:
+        x.append(i)
+        i+=1
+        y.append(float(row[:-1]))
+
+
+
+plt.plot(x,y, label='MAE')
+
+
+x = []
+y = []
+i=1
+
+
+with open('rmse.log','r') as csvfile:
+    for row in csvfile:
+        x.append(i)
+        i+=1
+        y.append(float(row[:-1]))
+
+
+plt.plot(x,y, label='RMSE')
+plt.plot(x,y, label='')
+
+plt.legend()
+plt.show()
\ No newline at end of file

Assignment1/data/plot1/readme.txt

+python3 main.py --train_file train.csv --test_file test.csv --epoch 8000 --lr 0.4 --loss rmse 
+python3 main.py --train_file train.csv --test_file test.csv --epoch 8000 --lr 0.3 --loss mse 
+python3 main.py --train_file train.csv --test_file test.csv --epoch 8000 --lr 0.4 --loss logcosh

Assignment1/demo.py

@@ -5,15 +5,49 @@ x = []
 y = []
 i=1
 
-with open('error.log','r') as csvfile:
+with open('logcosh.log','r') as csvfile:
     for row in csvfile:
         x.append(i)
         i+=1
         y.append(float(row[:-1]))
 
-plt.plot(x,y, label='Loaded from file!')
-plt.xlabel('x')
-plt.ylabel('y')
-plt.title('Interesting Graph\nCheck it out')
+
+
+plt.plot(x,y, label='LOGCOSH')
+plt.xlabel('epoch')
+plt.ylabel('Error')
+
+
+x = []
+y = []
+i=1
+
+
+with open('mae.log','r') as csvfile:
+    for row in csvfile:
+        x.append(i)
+        i+=1
+        y.append(float(row[:-1]))
+
+
+
+plt.plot(x,y, label='MAE')
+
+
+x = []
+y = []
+i=1
+
+
+with open('rmse.log','r') as csvfile:
+    for row in csvfile:
+        x.append(i)
+        i+=1
+        y.append(float(row[:-1]))
+
+
+plt.plot(x,y, label='RMSE')
+plt.plot(x,y, label='')
+
 plt.legend()
 plt.show()
\ No newline at end of file

Assignment1/main.py

 import numpy as np
 import argparse
 import csv
+import warnings
 # import matplotlib.pyplot as plt
 ''' 
 You are only required to fill the following functions
@@ -24,7 +25,7 @@ def mean_squared_loss(xdata, ydata, weights):
 
 	guess = np.dot(xdata,weights)
 	samples = np.shape(guess)[0]
-	err = (0.5/samples)*np.sum(np.square(ydata-guess))
+	err = (1/samples)*np.sum(np.square(ydata-guess))
 	return err
 	raise NotImplementedError
 
@@ -32,7 +33,7 @@ def mean_squared_gradient(xdata, ydata, weights):
 
 	samples = np.shape(xdata)[0]
 	guess = np.dot(xdata,weights)
-	gradient = (1/samples)*np.dot(xdata.T,(guess-ydata))
+	gradient = (2/samples)*np.dot(xdata.T,(guess-ydata))
 	return gradient
 
 	raise NotImplementedError
@@ -60,7 +61,11 @@ def mean_log_cosh_loss(xdata, ydata, weights):
 
 	guess = np.dot(xdata,weights)
 	samples = np.shape(guess)[0]
-	err = (1/samples)*np.sum(np.square(ydata-guess))
+	warnings.filterwarnings("error")
+	try:
+		err = np.sum(np.log(np.cosh(guess-ydata)))/samples
+	except Exception as e:
+		err = np.sum(np.absolute(guess-ydata)+np.log(2))/samples
 	return err
 	raise NotImplementedError
 
@@ -68,7 +73,7 @@ def mean_log_cosh_gradient(xdata, ydata, weights):
 
 	guess = np.dot(xdata,weights)
 	samples = np.shape(guess)[0]
-	gradient = np.dot(xdata.T,np.tanh(guess-ydata))
+	gradient = np.dot(xdata.T,np.tanh(guess-ydata))/samples
 	return gradient
 
 	raise NotImplementedError
@@ -84,7 +89,8 @@ def root_mean_squared_loss(xdata, ydata, weights):
 def root_mean_squared_gradient(xdata, ydata, weights):
 
 	samples = np.shape(xdata)[0]
-	gradient = -weights.T/np.sqrt(samples)
+	guess = np.dot(xdata,weights)
+	gradient = mean_squared_gradient(xdata, ydata, weights)/(2*root_mean_squared_loss(xdata, ydata, weights))
 	return gradient
 
 	raise NotImplementedError
@@ -277,7 +283,15 @@ def main():
     output = [(i,np.absolute(ytest[i])) for i in range(len(ytest))]
     np.savetxt("output.csv",output,delimiter=',',fmt="%d",header="instance (id),count",comments='')
     np.savetxt("error.log",errlog,delimiter='\n',fmt="%f")
-
+    x,y,z  = np.array([[ 1,  0,  2, -3], [ 1, -1,  0, -3], [-2, -5,  1, -3], [ 0, -5,  3, -3], [ 0, -4,  3, -2]]),np.array( [-2,  1,  1,  2,  0]),np.array( [ 1, 0, -2, -1])
+    print(mean_absolute_loss(x,y,z))
+    print(mean_absolute_gradient(x,y,z))
+    print(mean_squared_loss(x,y,z))
+    print(mean_squared_gradient(x,y,z))
+    print(root_mean_squared_loss(x,y,z))
+    print(root_mean_squared_gradient(x,y,z))
+    print(mean_log_cosh_loss(x,y,z))
+    print(mean_log_cosh_gradient(x,y,z))
 
 if __name__ == '__main__':