Merge branch 'master' of https://github.com/csc-training/hpc-python

Author: jussienko

multiprocessing/simple-calculation/README.md

@@ -0,0 +1,4 @@
+## Simple calculation
+
+Calculate the square of all numbers 1...10 using a separate `Process` to
+calculate the square of each number and print out the result.

multiprocessing/work-distribution/README.md

@@ -0,0 +1,22 @@
+## Work distribution
+
+Read the atom coordinates of the Zika virus [5ire.pdb](5ire.pdb) using the
+simple parser provided in [pdb.py](pdb.py) (see comments in the file for
+examples on how to use it). In [center-of-coords.py](center-of-coords.py) is a
+skeleton code you may also use as a starting point.
+
+Calculate the center of coordinates in parallel by dividing the coordinates
+in smaller chunks (e.g. 100 coordinates) to separate tasks. After completing
+the tasks in worker processes, combine the results in the master process.
+
+If needed, figure out also how to deal with any coordinates left out of the
+neat uniform-sized chunks.
+
+a) Use a **Pool of Workers** to distribute the tasks to worker processes.
+   *Hint: if you return also the weight of each chunk, you can process also
+   non-uniform sized chunks.*
+
+OR
+
+b) Use a **Queue** to distribute the tasks to worker processes. Store the
+   results for each chunk in a separate queue.

numpy/array-creation/README.md

@@ -0,0 +1,11 @@
+## Array creation
+
+Start from a Python list containing both integers and floating point values,
+and construct then a NumPy array from the list.
+
+Generate a 1D NumPy array containing all numbers from -2.0 to 2.0 with a
+spacing of 0.2. Use optional start and step arguments of the `np.arange()`
+function.
+
+Generate another 1D NumPy array containing 11 equally spaced values between
+0.5 and 1.5. Extract every second element of the array.

numpy/array-slicing/README.md

@@ -0,0 +1,17 @@
+## Array slicing
+
+First, create a 4x4 array with arbitrary values, then
+
+1. Extract every element from the second row.
+2. Extract every element from the third column.
+3. Assign a value of 0.21 to upper left 2x2 subarray.
+
+Next, create a 8x8 array with checkerboard pattern, i.e. alternating zeros
+and ones:
+
+```
+1 0 1 ...
+0 1 0 ...
+1 0 1 ...
+ ...
+```

numpy/split-combine/README.md

@@ -0,0 +1,8 @@
+## Split and combine arrays
+
+Continue with the previous 4x4 array
+
+1. Use `np.split()` function for splitting the array into two new 2x4 arrays.
+   Reconstruct the original 4x4 array by using `np.concatenate()`.
+2. Repeat the above exercise but create now 4x2 subarrays and then combine
+   them.

numpy/subdiagonal-matrix/README.md

@@ -0,0 +1,13 @@
+## Subdiagonal matrix
+
+Create a 6x6 matrix with 1’s above and below the diagonal and zeros
+otherwise:
+```
+0 1 0 0 0 0
+1 0 1 0 0 0
+0 1 0 1 0 0
+0 0 1 0 1 0
+0 0 0 1 0 1
+0 0 0 0 1 0
+```
+Use the `numpy.eye()` function.