diff --git a/Tasks/day2.py b/Tasks/day2.py
new file mode 100644
index 0000000..9241633
--- /dev/null
+++ b/Tasks/day2.py
@@ -0,0 +1,10 @@
+import numpy as np
+import torch
+
+a=np.random.rand(5,3)
+b=np.random.rand(3,4)
+a2=torch.from_numpy(a)
+b2=torch.from_numpy(b)
+
+c=torch.mm(a2,b2)
+print(c)
diff --git a/Tasks/registration_task/.pytest_cache/v/cache/lastfailed b/Tasks/registration_task/.pytest_cache/v/cache/lastfailed
new file mode 100644
index 0000000..9e26dfe
--- /dev/null
+++ b/Tasks/registration_task/.pytest_cache/v/cache/lastfailed
@@ -0,0 +1 @@
+{}
\ No newline at end of file
diff --git a/Tasks/registration_task/.pytest_cache/v/cache/nodeids b/Tasks/registration_task/.pytest_cache/v/cache/nodeids
new file mode 100644
index 0000000..2984d50
--- /dev/null
+++ b/Tasks/registration_task/.pytest_cache/v/cache/nodeids
@@ -0,0 +1,4 @@
+[
+  "test.py::test_code[ip0-12]",
+  "test.py::test_code[ip1-13]"
+]
\ No newline at end of file
diff --git a/Tasks/registration_task/__pycache__/main.cpython-36.pyc b/Tasks/registration_task/__pycache__/main.cpython-36.pyc
new file mode 100644
index 0000000..cf6e2b1
Binary files /dev/null and b/Tasks/registration_task/__pycache__/main.cpython-36.pyc differ
diff --git a/Tasks/registration_task/__pycache__/test.cpython-36-PYTEST.pyc b/Tasks/registration_task/__pycache__/test.cpython-36-PYTEST.pyc
new file mode 100644
index 0000000..3952102
Binary files /dev/null and b/Tasks/registration_task/__pycache__/test.cpython-36-PYTEST.pyc differ
diff --git a/Tasks/registration_task/main.py b/Tasks/registration_task/main.py
index b79be69..2fbd3f7 100644
--- a/Tasks/registration_task/main.py
+++ b/Tasks/registration_task/main.py
@@ -5,17 +5,40 @@
 y = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]
 """
 
-import numpy
+import numpy as np
 
 class LinearRegression(object):
-  """
-  An implementation of linear regression model
-  """
-  
-  def fit(_input, _output):
-    pass
-    
-  def predict(_input):
-    pass
-    
-    
+    def __init__(self, mx=None, my=None, b_0=None, b_1=None):
+        self.mx=mx
+        self.my=my
+        self.b_0=b_0
+        self.b_1=b_1
+    def fit(self,_input, _output):
+        n=np.size(_output)
+        self.mx=np.mean(_input)
+        self.my=np.mean(_output)
+        sum1,sum2=0,0
+        for i in range(n):
+            sum1+=_input[i][0]*_output[i]
+            sum2+=_input[i][0]*_input[i][0]
+        SS_xy =sum1-n*self.my*self.mx 
+        SS_xx =sum2-n*self.mx*self.mx
+        self.b_1 = SS_xy / SS_xx 
+        self.b_0 = self.my - self.b_1*self.mx
+        print(self.b_0)
+        print(self.b_1)
+        print(self.my)
+        print(self.mx)
+        print(SS_xx)
+        print(SS_xy)
+        print(sum1)
+        print(sum2)
+      
+      
+    def predict(self,_input):
+        n=np.size(_input)
+        _output=[]
+        for i in range(n):
+            a=self.b_0+(self.b_1*_input[i][0])
+            _output.append(a)
+        return _output