Update Numpy Section

Author: sh-navid

README-PYTHON-HOMEWORKS.md

@@ -117,6 +117,13 @@
 </details>
 
 
+## TP08
+- Date, Time, Format
+- Math
+- Inheritance
+- NumPy
+- ...
+
 
 ## Answers to the questions of students
 - [Get **object** by **id**()](/students/questions/object-get-by-id.py)

lessons/python/modules/numpy/README.md

@@ -1 +1,26 @@
-# NumPy
+# NumPy
+### Installation
+- `pip install numpy`
+### Import
+- `import numpy as np`
+### Concepts
+- [Array](numpy_array.py)
+  - `.array()`
+  - `.dtype, .ndim, .shape`
+- [DataTypes](numpy_datatypes.py)
+  - `i: int, u: unsigned int, c: complex, f: float`
+  - `M: datetime, m: timedelta`
+  - `b: bool`
+  - `O: object`
+  - `S: string, U: unicode string`
+  - `V: byte-like object` ??
+- [Convert Data Type](numpy_convert_data_type.py)
+  - `.astype()`
+- [Access](numpy_access.py)
+  - `.copy(), .view()`
+  - `.base` <sub>check if returned array is copy or view</sub>
+  - `np.nditer(array), np.ndenumerate(array)`
+- [Functions](numpy_functions.py)
+  - `.reshape()`
+  - `np.concatenate()`
+  - `np.stack(), np.hstack(), np.vstack(), np.dstack()`

lessons/python/modules/numpy/numpy_access.py

@@ -0,0 +1,80 @@
+import numpy as np
+
+# ------------------------------------------------------------------
+# View, Copy
+my_array = np.array([1, 2, 3, 4, 5])
+my_array_copy = my_array.copy()
+my_array_view = my_array.view()
+
+
+my_array_copy[0] = 100
+print(my_array, my_array_copy, my_array_copy.base)
+# [1 2 3 4 5] [100   2   3   4   5] None
+
+
+my_array_view[0] = 100
+print(my_array, my_array_view, my_array_view.base)
+# [100   2   3   4   5] [100   2   3   4   5] [100   2   3   4   5]
+
+
+# ------------------------------------------------------------------
+# Iterate
+
+
+my_array = np.array([1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12])
+my_array = my_array.reshape(2, 3, -1)
+
+
+for x in my_array:
+    for y in x:
+        for z in y:
+            print(z)
+"""
+1
+2
+3
+4
+5
+6
+7
+8
+9
+10
+11
+12
+"""
+
+
+for item in np.nditer(my_array):
+    print(item)
+"""
+1
+2
+3
+4
+5
+6
+7
+8
+9
+10
+11
+12
+"""
+
+for index, item in np.ndenumerate(my_array):
+    print(index, item)
+"""
+(0, 0, 0) 1
+(0, 0, 1) 2
+(0, 1, 0) 3
+(0, 1, 1) 4
+(0, 2, 0) 5
+(0, 2, 1) 6
+(1, 0, 0) 7
+(1, 0, 1) 8
+(1, 1, 0) 9
+(1, 1, 1) 10
+(1, 2, 0) 11
+(1, 2, 1) 12
+"""

lessons/python/modules/numpy/numpy_array.py

@@ -0,0 +1,39 @@
+import numpy as np
+
+
+my_array = np.array([1, 2, 3, 4, 5, 6, 7, 8, 9])
+print(my_array, type(my_array), my_array.dtype, "Dimensions", my_array.ndim)
+# [1 2 3 4 5 6 7 8 9] <class 'numpy.ndarray'> int64 Dimensions 1
+
+
+print(my_array.shape)
+# (9,)
+
+
+# Slice 1D
+print(my_array[1:3], "\n\n")
+# [2 3]
+
+
+my_array = np.array([[1, 2, 3, 4, 5, 6, 7, 8, 9], [10, 20, 30, 40, 50, 60, 70, 80, 90]])
+print(my_array, type(my_array), my_array.dtype, "Dimensions", my_array.ndim)
+# [[ 1  2  3  4  5  6  7  8  9]
+#  [10 20 30 40 50 60 70 80 90]] <class 'numpy.ndarray'> int64 Dimensions 2
+
+
+print(my_array.shape)
+# (2, 9)
+
+
+# Slice 2D
+print(my_array[1, 1:3])  # or
+print(my_array[1][1:3], "\n\n")
+# [20 30]
+# [20 30]
+
+
+# Access
+my_array[0][0] = 100
+print(my_array, type(my_array), my_array.dtype, "Dimensions", my_array.ndim)
+# [[100   2   3   4   5   6   7   8   9]
+#  [ 10  20  30  40  50  60  70  80  90]] <class 'numpy.ndarray'> int64 Dimensions 2

lessons/python/modules/numpy/numpy_convert_data_type.py

@@ -0,0 +1,16 @@
+import numpy as np
+
+
+my_array = np.array([2.8, 3.6, 4.1])
+print(my_array, my_array.dtype)
+# [2.8 3.6 4.1] float64
+
+
+my_array = my_array.astype("i1")
+print(my_array, my_array.dtype)
+# [2 3 4] int8
+
+
+my_array = my_array.astype("S1")
+print(my_array, my_array.dtype)
+# [b'2' b'3' b'4'] |S1

lessons/python/modules/numpy/numpy_datatypes.py

@@ -0,0 +1,73 @@
+import numpy as np
+
+
+my_array = np.array(["A", "B", "Hello"], dtype="S")
+print(my_array, type(my_array), my_array.dtype)  # S5
+
+
+my_array = np.array(["A", "B", "Hello"], dtype="U")
+print(my_array, type(my_array), my_array.dtype)  # U5
+
+
+my_array = np.array([3 + 4j, 1 + 2j, 1])
+print(my_array, type(my_array), my_array.dtype)  # complex128
+
+
+my_array = np.array([1.1, 2, 3])
+print(my_array, type(my_array), my_array.dtype)  # float64
+
+
+my_array = np.array(["2020-12-02", "2022-11-03"], dtype="M")
+print(my_array, type(my_array), my_array.dtype)  # datetime64
+
+
+my_array = np.array([100000, 200000], dtype="m")
+print(my_array, type(my_array), my_array.dtype)  # timedelta64
+
+
+my_array = np.array([True, False])
+print(my_array, type(my_array), my_array.dtype)  # bool
+
+
+my_array = np.array([1, 2], "i1")
+print(my_array, type(my_array), my_array.dtype)  # int8 -> 8 bits -> 0 - 255
+
+
+my_array = np.array([1, 2], "i2")
+print(my_array, type(my_array), my_array.dtype)  # int16
+
+
+my_array = np.array([1, 2], "i4")
+print(my_array, type(my_array), my_array.dtype)  # int32
+
+
+my_array = np.array([1, 2], "i8")
+print(my_array, type(my_array), my_array.dtype)  # int64
+
+
+my_array = np.array([1, 2], "b")
+print(my_array, type(my_array), my_array.dtype)  # int8
+
+
+my_array = np.array([1, 2], "f2")
+print(my_array, type(my_array), my_array.dtype)  # float16
+
+
+my_array = np.array([1, 2], "f")  # == f4
+print(my_array, type(my_array), my_array.dtype)  # float32
+
+
+my_array = np.array([1, 2], "f8")
+print(my_array, type(my_array), my_array.dtype)  # float64
+
+
+class MyClass:
+    pass
+
+
+my_array = np.array([MyClass(), MyClass()], "O")
+print(my_array, type(my_array), my_array.dtype)  # object
+
+
+my_array = np.array([b'A', b'B'], "V")
+print(my_array, type(my_array), my_array.dtype)  # V1

lessons/python/modules/numpy/numpy_functions.py

@@ -0,0 +1,86 @@
+import numpy as np
+
+#------------------------------------------------------------------
+# Reshape
+my_array = np.array([1, 2, 3, 4, 5, 6])
+print(my_array, my_array.shape)
+# [1 2 3 4 5 6]     (6,)
+
+
+my_array = my_array.reshape(2, 3)
+print(my_array, my_array.shape)
+# [[1 2 3]
+#  [4 5 6]]         (2, 3)
+
+
+my_array = my_array.reshape(6)
+print(my_array, my_array.shape)
+# [1 2 3 4 5 6]     (6,)
+
+
+my_array = my_array.reshape(3, 2)
+print(my_array, my_array.shape)
+# [[1 2]
+#  [3 4]
+#  [5 6]]           (3, 2)
+
+
+my_array = my_array.reshape(6)
+print(my_array, my_array.shape)
+# [1 2 3 4 5 6]     (6,)
+
+
+my_array = my_array.reshape(2, 3, 1)
+print(my_array, my_array.shape)
+# [[[1]
+#   [2]
+#   [3]]
+#
+#  [[4]
+#   [5]
+#   [6]]]           (2, 3, 1)
+
+
+my_array = my_array.reshape(6)
+print(my_array, my_array.shape)
+# [1 2 3 4 5 6]     (6,)
+
+
+my_array = my_array.reshape(1,-1,2) # When you do not know the correct value for dimention you can pass -1
+print(my_array, my_array.shape)
+# [[[1 2]
+#  [3 4]
+#  [5 6]]] (1, 3, 2)
+
+
+my_array = my_array.reshape(-1) # pass -1 to make a flat array
+print(my_array, my_array.shape)
+# [1 2 3 4 5 6]     (6,)
+
+
+#------------------------------------------------------------------
+# np.concatenate()
+
+x=np.array([4,5,6])
+y=np.array([7,8,9])
+print(np.concatenate([x,y]))
+# [4 5 6 7 8 9]
+
+#------------------------------------------------------------------
+# np.stack()
+
+print(np.stack([x,y]))
+# [[4 5 6]
+#  [7 8 9]]
+
+print(np.hstack([x,y]))
+# [4 5 6 7 8 9]
+
+print(np.vstack([x,y]))
+# [[4 5 6]
+#  [7 8 9]]
+
+print(np.dstack([x,y]))
+# [[[4 7]
+#   [5 8]
+#   [6 9]]]