Initial commit

Author: github-classroom[bot]

.github/workflows/classroom.yml

@@ -0,0 +1 @@
+solution.py

.gitignore

@@ -0,0 +1,31 @@
+
+# Deep Learning In Python - Week 1
+
+<img src="https://github.com/hannesrosenbusch/DLIP_Week1/blob/main/assets/uvalogo.svg.png?raw=true" width="125">    <img src="https://github.com/hannesrosenbusch/DLIP_Week1/blob/main/assets/pythonlogo.jpeg?raw=true" width="380">
+
+
+An introduction to Python can be seen here: <br>
+[Python for R users](https://youtube.com/playlist?list=PLq0cz82QvYapppmpXPYgS76VbHHKRIgbk)
+
+# Steps for Completing the Assignment
+
+## Step 1
+Make a github account if you don't have one already.
+
+## Step 2:
+Clone this github repository onto your laptop via a terminal, or use google colab, or kaggle...anywhere where you can edit python code.
+There are plenty of online resources which demo how to clone a repo e.g. here: https://www.earthdatascience.org/workshops/intro-version-control-git/basic-git-commands. 
+(Notice that you need to install python if you decide to work on your laptop)
+
+## Step 3: 
+Edit the tasks.py file by filling in your solution code.
+Run the file to see whether you have errors.
+You can also run the grading files to preview whether your solution passes, but you are not allowed to change these files.
+
+## Step 4:
+Add this week's kaggle certificate(s) to the repository (course: https://www.kaggle.com/learn/python ; https://www.kaggle.com/learn/pandas). If you received more than one certificate for this week, combine them into one pdf before. Name the file `certificate.pdf` 
+
+
+## Step 5: 
+Push your changes to the repo back to github. The grading will occur automatically but might take a minute.
+Again, the add/commit/push workflow of github is explained in various places online: https://www.earthdatascience.org/workshops/intro-version-control-git/basic-git-commands)

README.md

@@ -0,0 +1,12 @@
+import numpy as np
+from tasks import step
+
+# Test cases for Task 1: step function
+def test_step_positive():
+    assert step(5) == 1, "Failed on positive input"
+
+def test_step_negative():
+    assert step(-3) == -1, "Failed on negative input"
+
+def test_step_zero():
+    assert step(0) == -1, "Failed on zero input"

assets/pythonlogo.jpeg

@@ -0,0 +1,20 @@
+import numpy as np
+from tasks import step, ReLU
+
+
+def test_relu_default_cutoff():
+    array = np.array([-5, 0, 3, -2, 4])
+    expected_output = np.array([0, 0, 3, 0, 4])
+    np.testing.assert_array_equal(ReLu(array.copy()), expected_output, "Failed on default cutoff")
+
+def test_relu_custom_cutoff():
+    array = np.array([-5, 0, 3, -2, 4])
+    expected_output = np.array([2, 2, 3, 2, 4])
+    np.testing.assert_array_equal(ReLu(array.copy(), 2), expected_output, "Failed on custom cutoff")
+
+def test_relu_empty_array():
+    array = np.array([])
+    expected_output = np.array([])
+    np.testing.assert_array_equal(ReLu(array.copy()), expected_output, "Failed on empty array")
+
+

assets/uvalogo.svg.png

@@ -0,0 +1,20 @@
+import numpy as np
+from tasks import step, ReLu, neural_net_layer
+
+def test_neural_net_layer_basic():
+    inputs = np.array([[1, 2], [3, 4]])
+    weights = np.array([1, -1])
+    expected_output = np.array([0, 0])  # (1*1 + 2*(-1)) = -1 -> 0 (after ReLu), (3*1 + 4*(-1)) = -1 -> 0 (after ReLu)
+    np.testing.assert_array_equal(neural_net_layer(inputs, weights), expected_output, "Failed on basic 2D array")
+
+def test_neural_net_layer_custom():
+    inputs = np.array([[2, 3], [1, 1]])
+    weights = np.array([1, 2])
+    expected_output = np.array([8, 3])  # (2*1 + 3*2) = 8, (1*1 + 1*2) = 3
+    np.testing.assert_array_equal(neural_net_layer(inputs, weights), expected_output, "Failed on different inputs")
+
+def test_neural_net_layer_with_negatives():
+    inputs = np.array([[1, -1], [-2, 3]])
+    weights = np.array([-2, 1])
+    expected_output = np.array([0, 7])  # ReLu([-3, 7]) = [0, 7]
+    np.testing.assert_array_equal(neural_net_layer(inputs, weights), expected_output, "Failed on negative matrix values")

grade_task1.py

@@ -0,0 +1,50 @@
+import numpy
+
+# Follow the tasks below to practice basic Python concepts.
+# Write your code in between the dashed lines.
+# Don't import additional packages. Numpy suffices.
+
+# Task 1: 
+# Instructions:
+#Write a function that takes one numeric argument as input. 
+#If the number is larger than zero, the function should return 1, otherwise is should return -1.
+#The name of the function should be step
+
+# Your code here:
+# -----------------------------------------------
+
+def step
+
+
+# -----------------------------------------------
+
+
+# Task 2:
+# Instructions:
+#Write a function that takes in two arguments: a numpy array, and an integer (call argument "cutoff" and set default to 0).
+#The function should return a numpy array of the same length, with all elements smaller than the cutoff being set to cutoff).
+#The name of the function should be ReLu
+
+
+# Your code here:
+# -----------------------------------------------
+def ReLu
+
+
+# -----------------------------------------------
+
+
+# Task 3:
+# Instructions:
+#Write a function that takes in a two-dimensional numpy array of size (n, p) and a one-dimensional numpy array of size p.
+#The function should start by multiplying the two numpy arrays (matrix multiplication).
+#Next, apply the ReLu function from above to the resulting matrix and return the result.
+#Name the function neural_net_layer
+
+# Your code here:
+# -----------------------------------------------
+
+def neural_net_layer
+
+
+# ------------------------------------------