Introduction to Python - Revision history

WikiSysop at 17:31, 16 October 2019

2019-10-16T17:31:08Z

← Older revision		Revision as of 11:31, 16 October 2019
Line 29:		Line 29:
	==== Mac ====		==== Mac ====

	~~Haven't tried this yet, and this is largely a placeholder for a Mac-owning colleague~~ to ~~write more details~~, ~~but [http~~://www.~~thisisthegreenroom~~.com/~~2011/installing-python-numpy-scipy-matplotlib-and-ipython-on-lion~~/~~#python this blog post]~~		Probably best to use Anaconda distribution, which you can download from: https://www.anaconda.com/distribution/
	~~looks very promising for Mac users:~~

	==== Linux ====		==== Linux ====

Awickert: /* How to Write a Model in Python */ minus cancels (strange error fix)

2017-04-05T18:11:13Z

How to Write a Model in Python: minus cancels (strange error fix)

← Older revision		Revision as of 12:11, 5 April 2017
Line 84:		Line 84:
	Combining these two terms together gives us the diffusion equation:		Combining these two terms together gives us the diffusion equation:

	:<math> \frac{\partial z}{\partial t} = - \frac{\partial}{\partial x} \left(-k \frac{\partial z}{\partial x}\right) </math>		:<math> \frac{\partial z}{\partial t} = \frac{\partial}{\partial x} \left(k \frac{\partial z}{\partial x}\right) </math>

	If we assume that the hillslope diffusivity, <math>k</math>, is uniform across the landscape, we obtain the diffusion equation in its more familiar form:		If we assume that the hillslope diffusivity, <math>k</math>, is uniform across the landscape, we obtain the diffusion equation in its more familiar form:

Awickert: /* Theory */ -

2017-04-05T18:09:01Z

Theory: -

← Older revision		Revision as of 12:09, 5 April 2017
Line 84:		Line 84:
	Combining these two terms together gives us the diffusion equation:		Combining these two terms together gives us the diffusion equation:

	:<math> \frac{\partial z}{\partial t} = \frac{\partial}{\partial x} \left(-k \frac{\partial z}{\partial x}\right) </math>		:<math> \frac{\partial z}{\partial t} = - \frac{\partial}{\partial x} \left(-k \frac{\partial z}{\partial x}\right) </math>

	If we assume that the hillslope diffusivity, <math>k</math>, is uniform across the landscape, we obtain the diffusion equation in its more familiar form:		If we assume that the hillslope diffusivity, <math>k</math>, is uniform across the landscape, we obtain the diffusion equation in its more familiar form:

Awickert: /* Theory */ another sign error

2012-09-19T03:04:34Z

Theory: another sign error

← Older revision		Revision as of 21:04, 18 September 2012
Line 87:		Line 87:

	If we assume that the hillslope diffusivity, <math>k</math>, is uniform across the landscape, we obtain the diffusion equation in its more familiar form:		If we assume that the hillslope diffusivity, <math>k</math>, is uniform across the landscape, we obtain the diffusion equation in its more familiar form:
	:<math> \frac{\partial z}{\partial t} = -k \frac{\partial^2 z}{\partial x^2} </math>		:<math> \frac{\partial z}{\partial t} = k \frac{\partial^2 z}{\partial x^2} </math>

	=== Source code ===		=== Source code ===

Awickert: /* Theory */ sign error

2012-09-19T02:49:21Z

Theory: sign error

← Older revision		Revision as of 20:49, 18 September 2012
Line 80:		Line 80:
	Hillslope evolution occurs when we combine these discharges to conserve mass. We use a continuity equation for conservation of volume to state that:		Hillslope evolution occurs when we combine these discharges to conserve mass. We use a continuity equation for conservation of volume to state that:

	:<math> \frac{\partial z}{\partial t} = \frac{\partial Q}{\partial x}</math>		:<math> \frac{\partial z}{\partial t} = - \frac{\partial Q}{\partial x}</math>

	Combining these two terms together gives us the diffusion equation:		Combining these two terms together gives us the diffusion equation:

Awickert: /* Copyright Notice */ CSDMS open source

2012-09-07T17:58:30Z

← Older revision		Revision as of 11:58, 7 September 2012
Line 211:		Line 211:

	The aforementioned code is trivial, but lately (2012) it seems that big, ugly patent wars are being fought over trivial chunks of code. So better to be safe than sorry. Below is text included for version 3 of the [http://www.gnu.org GNU] General Public License (GPL). This is an open-source "copyleft" license, meaning that if this code is included in a project, the project must be open-source with a "copyleft" license as well. This is a very good license for science, which is a community-driven collaborative effort. To give your code the GNU GPL v3, include the following text (though replacing my name with yours in the copyright, unless you are feeling generous).		The aforementioned code is trivial, but lately (2012) it seems that big, ugly patent wars are being fought over trivial chunks of code. So better to be safe than sorry. Below is text included for version 3 of the [http://www.gnu.org GNU] General Public License (GPL). This is an open-source "copyleft" license, meaning that if this code is included in a project, the project must be open-source with a "copyleft" license as well. This is a very good license for science, which is a community-driven collaborative effort. To give your code the GNU GPL v3, include the following text (though replacing my name with yours in the copyright, unless you are feeling generous).

			In order for your code to be included in CSDMS, you need to include some type of open-source license, even if it is not as strong a license as the GPL v3.

	<syntaxhighlight lang=python>		<syntaxhighlight lang=python>

Awickert: /* Copyright Notice */ description

2012-09-07T17:57:22Z

← Older revision		Revision as of 11:57, 7 September 2012
Line 210:		Line 210:
	==== Copyright Notice ====		==== Copyright Notice ====

			The aforementioned code is trivial, but lately (2012) it seems that big, ugly patent wars are being fought over trivial chunks of code. So better to be safe than sorry. Below is text included for version 3 of the [http://www.gnu.org GNU] General Public License (GPL). This is an open-source "copyleft" license, meaning that if this code is included in a project, the project must be open-source with a "copyleft" license as well. This is a very good license for science, which is a community-driven collaborative effort. To give your code the GNU GPL v3, include the following text (though replacing my name with yours in the copyright, unless you are feeling generous).

	<syntaxhighlight lang=python>		<syntaxhighlight lang=python>
Line 230:		Line 231:
	</syntaxhighlight>		</syntaxhighlight>

			On a technical note, the triple quotes at the beginning and end of the block are used in Python to create blocks of text as strings. These can be used later inside functions, blocks of code designed for a single purpose, to provide accessible documentation. For now, since we are not assigning this string anywhere, it simply functions as a comment.

	==== Introductory Comments and Authorship ====		==== Introductory Comments and Authorship ====

Awickert: /* Source code */ GNU GPL v3

2012-09-07T05:40:36Z

Source code: GNU GPL v3

@@ Line 99: / Line 99: @@
 import numpy as np
 from matplotlib import pyplot as plt
 ################################################################################
@@ Line 190: / Line 207: @@
 Namespaces are important, because it means that you cannot easily accidentally overwrite the name of a function by giving a variable that name. For example, typing "np.mean = 1.5" seems a less-likely accident than writing "mean = 1.5".
 ==== Introductory Comments and Authorship ====

Awickert: /* Source code */ Split code into relevant sections

2012-09-07T05:15:55Z

Source code: Split code into relevant sections

← Older revision		Revision as of 23:15, 6 September 2012
Line 190:		Line 190:

	Namespaces are important, because it means that you cannot easily accidentally overwrite the name of a function by giving a variable that name. For example, typing "np.mean = 1.5" seems a less-likely accident than writing "mean = 1.5".		Namespaces are important, because it means that you cannot easily accidentally overwrite the name of a function by giving a variable that name. For example, typing "np.mean = 1.5" seems a less-likely accident than writing "mean = 1.5".

			==== Introductory Comments and Authorship ====

			This section is huge overkill for this short of a model, but I put it in here to be complete.

			<syntaxhighlight lang=python>
			################################################################################
			# hill.py
			#
			# HILLSLOPE PROFILE
			#
			# Center of hill is at 0, two incising channels at steady rate at edges
			# no channel-hillslope feedback and assuming entire hill is made of mobile
			# regolith
			#
			# This is a very simple example code in which variables are defined inline,
			# and the equations are solved by forward difference methods without
			# consideration for stability
			#
			# Written by Andrew D. Wickert; August, 2013
			################################################################################
			</syntaxhighlight>

			==== Initialization ====

			<syntaxhighlight lang=python>
			# We will have the simple boundary condition of rivers on either end of the
			# hillslope that are incising at the same, steady rate.
			zdot_channel = 2E-4 # [m], 0.5 mm/yr boundary incision

			# time
			dt = 100. # [years], the time step (from guessing, not from stability analysis)
			t_final = 4E5 # [years], the final time for hillslope evolution
			timesteps = np.arange(0,t_final + dt/10.,dt) # [years]

			# Set up domain
			dx = 10 # [m]
			xmax = 500 # [m]
			x = np.arange(-xmax, xmax + dx/10., dx) # [m], +dx/10. to make sure that edges are included
			z = t_final * zdot_channel * np.ones(x.shape) # [m], elevation in meters - set such that the edges are 0 at t_final

			# We're using a very simplified assumption that the rate of hillslope
			# material transport is linearly proportional to local slope, and that this
			# constant of proportionality is constant across the hill
			k = 5E-3 # Slope -- soil flux scaling
			</syntaxhighlight>

			==== Main Loop ====

			<syntaxhighlight lang=python>
			# Loop through time and evolve elevation
			for t in timesteps:
			# The channels set the boundaries
			z[0] -= zdot_channel * dt
			z[-1] -= zdot_channel * dt
			# for np.diff, out[n] = a[n+1] - a[n]
			# We are calculating the slopes between each of the cells, then using
			# these to solve for discharge of material between interior cells.
			S = np.diff(z) # [-] slope
			Q = -kS # [m*2/yr] discharge, goes downslope, hence the negative sign # POSITIVE - WHY? CANCELLED OUT AGAIN!
			# The change in internal elevation, due to conservation of mass, is equal
			# once again to the x-defivative
			dzdt_interior = np.diff(Q)
			z[1:-1] -= dzdt_interior * dt
			# (It would have been quicker to take both derivatives at once, but would be
			# somewhat less intuitive for descriptive purposes)
			</syntaxhighlight>

			==== Plotting ====

			<syntaxhighlight lang=python>
			# Plot
			plt.plot(x,z,'k-',linewidth=3)
			plt.title('Final hillslope profile', fontsize=20, weight='bold')
			plt.xlabel('Distance across hillside [m]', fontsize=16)
			plt.ylabel('Elevation [m]', fontsize=16)
			plt.xlim((x[0],x[-1]))
			plt.show()
			</syntaxhighlight>

Awickert: /* Downloading and Installing */ Windows / generic w/ enthought

2012-09-04T02:17:32Z

Downloading and Installing: Windows / generic w/ enthought

← Older revision		Revision as of 20:17, 3 September 2012
Line 25:		Line 25:

	==== Windows ====		==== Windows ====
			This is a [http://www.southampton.ac.uk/~fangohr/computing/pythonsoft.html cross-platform guide] to installing the enthought Python distribution. Haven't tested it yet.

	==== Mac ====		==== Mac ====