.. DO NOT EDIT. .. THIS FILE WAS AUTOMATICALLY GENERATED BY SPHINX-GALLERY. .. TO MAKE CHANGES, EDIT THE SOURCE PYTHON FILE: .. "examples/Statistics/plot_StatArray.py" .. LINE NUMBERS ARE GIVEN BELOW. .. only:: html .. note:: :class: sphx-glr-download-link-note :ref:`Go to the end ` to download the full example code. .. rst-class:: sphx-glr-example-title .. _sphx_glr_examples_Statistics_plot_StatArray.py: StatArray Class ---------------- Extends the numpy ndarray class to add extra attributes such as names, and units, and allows us to attach statistical descriptors of the array. The direct extension to numpy maintains speed and functionality of numpy arrays. .. GENERATED FROM PYTHON SOURCE LINES 11-19 .. code-block:: Python import numpy as np import matplotlib.pyplot as plt import h5py from geobipy import DataArray, StatArray, Histogram, Distribution, RectilinearMesh1D # plt.style.use('seaborn-pastel') .. GENERATED FROM PYTHON SOURCE LINES 20-23 Instantiating a new StatArray class +++++++++++++++++++++++++++++++++++ .. GENERATED FROM PYTHON SOURCE LINES 23-69 .. code-block:: Python # Integer test = StatArray(1, name='1') assert isinstance(test, StatArray) and test.size == 1 and test.item() == 0.0, TypeError("da 0") print(test.summary) test = StatArray(10, name='10') assert isinstance(test, StatArray) and test.size == 10 and np.all(test == 0.0), TypeError("da 1") print(test.summary) # tuple/Shape test = StatArray((2, 10), name='(2, 10)') assert isinstance(test, StatArray) and np.all(test.shape == (2, 10)) and np.all(test == 0.0), TypeError("da 2") print(test.summary) test = StatArray([2, 10], name='(2, 10)') assert isinstance(test, StatArray) and np.all(test == [2, 10]), TypeError("da 2") print(test.summary) # float test = StatArray(45.454, name='45.454') assert isinstance(test, StatArray) and test.size == 1 and test.item() == 45.454, TypeError("da 3") print(test.summary) test = StatArray(np.float64(45.454), name='45.454') assert isinstance(test, StatArray) and test.size == 1 and test.item() == 45.454, TypeError("da 4") print(test.summary) # array test = StatArray(np.random.randn(1), name="test", units="$\frac{g}{cc}$") assert isinstance(test, StatArray) and test.size == 1, TypeError("da 5") print(test.summary) test = StatArray(np.arange(10.0), name="test", units="$\frac{g}{cc}$") assert isinstance(test, StatArray) and test.size == 10, TypeError("da 6") print(test.summary) test = DataArray(np.arange(10.0), name="test", units="$\frac{g}{cc}$") test = StatArray(test) assert isinstance(test, StatArray) and test.size == 10, TypeError("da 6") print(test.summary) # The StatArray can take any numpy function that returns an array as an input. # The name and units of the variable can be assigned to the StatArray. .. rst-class:: sphx-glr-script-out .. code-block:: none StatArray Name: 1 Address:['0x15d8a20d0'] Shape: (1,) Values: [0.] Min: 0.0 Max: 0.0 has_posterior: False StatArray Name: 10 Address:['0x15d8a1b50'] Shape: (10,) Values: [0. 0. 0. ... 0. 0. 0.] Min: 0.0 Max: 0.0 has_posterior: False StatArray Name: (2, 10) Address:['0x15d8a2f50'] Shape: (2, 10) Values: [[0. 0. 0. ... 0. 0. 0.] [0. 0. 0. ... 0. 0. 0.]] Min: 0.0 Max: 0.0 has_posterior: False StatArray Name: (2, 10) Address:['0x15d8a1cd0'] Shape: (2,) Values: [ 2 10] Min: 2 Max: 10 has_posterior: False StatArray Name: 45.454 Address:['0x15d8a0a50'] Shape: (1,) Values: [45.454] Min: 45.454 Max: 45.454 has_posterior: False StatArray Name: 45.454 Address:['0x15d8a2950'] Shape: (1,) Values: [45.454] Min: 45.454 Max: 45.454 has_posterior: False StatArray Name: test ($\frac{g}{cc}$) Address:['0x15d8a0fd0'] Shape: (1,) Values: [-1.33270353] Min: -1.3327035333786468 Max: -1.3327035333786468 has_posterior: False StatArray Name: test ($\frac{g}{cc}$) Address:['0x15d8a1e50'] Shape: (10,) Values: [0. 1. 2. ... 7. 8. 9.] Min: 0.0 Max: 9.0 has_posterior: False StatArray Name: test ($\frac{g}{cc}$) Address:['0x15d8a0fd0'] Shape: (10,) Values: [0. 1. 2. ... 7. 8. 9.] Min: 0.0 Max: 9.0 has_posterior: False .. GENERATED FROM PYTHON SOURCE LINES 70-89 Attaching Prior and Proposal Distributions to a StatArray +++++++++++++++++++++++++++++++++++++++++++++++++++++++++ The StatArray class has been built so that we may easily attach not only names and units, but statistical distributions too. We won't go into too much detail about the different distribution Two types of distributions can be attached to the StatArray. * Prior Distribution The prior represents how the user believes the variable should behave from a statistical standpoint. The values of the variable can be evaluated against the attached prior, to determine how likely they are to have occured https://en.wikipedia.org/wiki/Prior_probability * Proposal Distribution The proposal describes a probability distribution from which to sample when we wish to perturb the variable https://en.wikipedia.org/wiki/Metropolis%E2%80%93Hastings_algorithm .. GENERATED FROM PYTHON SOURCE LINES 89-101 .. code-block:: Python # Obtain an instantiation of a random number generator. # This is optional, but is an important consideration for parallel programming. from numpy.random import Generator from numpy.random import PCG64DXSM generator = PCG64DXSM(seed=0) prng = Generator(generator) Density = StatArray(10.0, name="test", units="$\frac{g}{cc}$") Density.prior = Distribution('Uniform', -2.0, 2.0, prng=prng) .. GENERATED FROM PYTHON SOURCE LINES 102-103 We can also attach a proposal distribution .. GENERATED FROM PYTHON SOURCE LINES 103-109 .. code-block:: Python Density.proposal = Distribution('Normal', 0.0, 1.0, prng=prng) print(Density.summary) print("Class type of the prior: ",type(Density.prior)) print("Class type of the proposal: ",type(Density.proposal)) .. rst-class:: sphx-glr-script-out .. code-block:: none StatArray Name: test ($\frac{g}{cc}$) Address:['0x10e4625d0' '0x153f358b0' '0x15d91f6f0' ... '0x15da5d970' '0x15d91f7b0' '0x15d91fb70'] Shape: (1,) Values: [10.] Min: 10.0 Max: 10.0 Prior: | Uniform Distribution: | Min: -2.0 | Max: 2.0 Proposal: | Normal | Mean:0.0 | Variance:1.0 has_posterior: False Class type of the prior: Class type of the proposal: .. GENERATED FROM PYTHON SOURCE LINES 110-113 The values in the variable can be evaluated against the prior. In this case, we have 3 elements in the variable, and a univariate Normal for the prior. Therefore each element is evaluated to get 3 probabilities, one for each element. .. GENERATED FROM PYTHON SOURCE LINES 113-115 .. code-block:: Python print(Density.probability(log=False)) .. rst-class:: sphx-glr-script-out .. code-block:: none 0.0 .. GENERATED FROM PYTHON SOURCE LINES 116-117 The univariate proposal distribution can generate random samples from itself. .. GENERATED FROM PYTHON SOURCE LINES 117-119 .. code-block:: Python print(Density.propose()) .. rst-class:: sphx-glr-script-out .. code-block:: none 1.1375024404290368 .. GENERATED FROM PYTHON SOURCE LINES 120-122 From a sampling stand point we can either sample using only the proposal Or we can only generate samples that simultaneously satisfy the prior. .. GENERATED FROM PYTHON SOURCE LINES 122-124 .. code-block:: Python print(Density.propose(relative=True)) .. rst-class:: sphx-glr-script-out .. code-block:: none [10.53816627] .. GENERATED FROM PYTHON SOURCE LINES 125-126 We can perturb the variable by drawing from the attached proposal distribution. .. GENERATED FROM PYTHON SOURCE LINES 126-130 .. code-block:: Python Density.perturb() print(Density.summary) .. rst-class:: sphx-glr-script-out .. code-block:: none StatArray Name: test ($\frac{g}{cc}$) Address:['0x10e4625d0' '0x153f358b0' '0x15d91f6f0' ... '0x15da5d970' '0x15d91f7b0' '0x15d91fb70'] Shape: (1,) Values: [0.38188467] Min: 0.38188466718060166 Max: 0.38188466718060166 Prior: | Uniform Distribution: | Min: -2.0 | Max: 2.0 Proposal: | Normal | Mean:0.0 | Variance:1.0 has_posterior: False .. GENERATED FROM PYTHON SOURCE LINES 131-136 Attaching a Histogram to capture the posterior distribution +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ The StatArray can perturb itself, evaluate its current probability given its priors and a histogram can be attached to capture its posterior distribution. As an example, lets create a Histogram class with bins generated from the prior. .. GENERATED FROM PYTHON SOURCE LINES 136-137 .. code-block:: Python bins = Density.prior.bins() .. GENERATED FROM PYTHON SOURCE LINES 138-139 Attach the histogram .. GENERATED FROM PYTHON SOURCE LINES 139-141 .. code-block:: Python Density.posterior = Histogram(mesh = RectilinearMesh1D(edges=bins)) .. GENERATED FROM PYTHON SOURCE LINES 142-143 In an iterative sense, we can propose and evaluate new values, and update the posterior .. GENERATED FROM PYTHON SOURCE LINES 143-150 .. code-block:: Python for i in range(1000): Density.perturb() p = Density.probability(log=False) if p > 0.0: # This is a simple example! Density.update_posterior() .. GENERATED FROM PYTHON SOURCE LINES 151-154 .. code-block:: Python plt.figure() Density.summaryPlot() .. image-sg:: /examples/Statistics/images/sphx_glr_plot_StatArray_001.png :alt: Prior, Proposal, Posterior :srcset: /examples/Statistics/images/sphx_glr_plot_StatArray_001.png :class: sphx-glr-single-img .. GENERATED FROM PYTHON SOURCE LINES 155-159 Attach a multivariate normal distribution as the prior and proposal +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ Attach the multivariate prior .. GENERATED FROM PYTHON SOURCE LINES 159-165 .. code-block:: Python mean = np.random.randn(Density.size) variance = np.ones(Density.size) Density.prior = Distribution('MvNormal', mean, variance, prng=prng) .. GENERATED FROM PYTHON SOURCE LINES 166-169 Since the prior is multivariate, the appropriate equations are used to evaluate the probability for all elements in the StatArray. This produces a single probability. .. GENERATED FROM PYTHON SOURCE LINES 169-172 .. code-block:: Python print(Density.probability(log=False)) .. rst-class:: sphx-glr-script-out .. code-block:: none 0.19092547942875554 .. GENERATED FROM PYTHON SOURCE LINES 173-174 Attach the multivariate proposal .. GENERATED FROM PYTHON SOURCE LINES 174-180 .. code-block:: Python mean = np.random.randn(Density.size) variance = np.ones(Density.size) Density.proposal = Distribution('MvNormal', mean, variance, prng=prng) .. GENERATED FROM PYTHON SOURCE LINES 181-182 Perturb the variables using the multivariate proposal. .. GENERATED FROM PYTHON SOURCE LINES 182-197 .. code-block:: Python Density.perturb() Density.summary with h5py.File('statarray.h5', 'w') as f: Density.createHdf(f, 'statarray', withPosterior=True, add_axis=3) Density.writeHdf(f, 'statarray', withPosterior=True, index=0) with h5py.File('statarray.h5', 'r') as f: tmp = StatArray.fromHdf(f, 'statarray', index=0, skip_posterior=False) with h5py.File('statarray.h5', 'r') as f: tmp = StatArray.fromHdf(f, 'statarray', skip_posterior=False) .. GENERATED FROM PYTHON SOURCE LINES 198-209 Basic manipulation ++++++++++++++++++ The StatArray contains other functions to perform basic array manipulations These routines essentially wrap around numpy functions, but the result will have the same name and units, and if any prior or proposal are set, those will be carried through too. 1D example __________ .. GENERATED FROM PYTHON SOURCE LINES 209-213 .. code-block:: Python x = StatArray(-np.cumsum(np.arange(10.0))) print(x) .. rst-class:: sphx-glr-script-out .. code-block:: none [ -0. -1. -3. ... -28. -36. -45.] .. GENERATED FROM PYTHON SOURCE LINES 214-219 .. code-block:: Python print(x.insert(i=[0, 9], values=[999.0, 999.0])) .. rst-class:: sphx-glr-script-out .. code-block:: none [999. -0. -1. ... -36. 999. -45.] .. GENERATED FROM PYTHON SOURCE LINES 220-225 .. code-block:: Python print(x.prepend(999.0)) .. rst-class:: sphx-glr-script-out .. code-block:: none [999. -0. -1. ... -28. -36. -45.] .. GENERATED FROM PYTHON SOURCE LINES 226-231 .. code-block:: Python print(x.prepend([998.0, 999.0])) .. rst-class:: sphx-glr-script-out .. code-block:: none [998. 999. -0. ... -28. -36. -45.] .. GENERATED FROM PYTHON SOURCE LINES 232-237 .. code-block:: Python print(x.append([998.0, 999.0])) .. rst-class:: sphx-glr-script-out .. code-block:: none [ -0. -1. -3. ... -45. 998. 999.] .. GENERATED FROM PYTHON SOURCE LINES 238-243 .. code-block:: Python print(x.resize(14)) .. rst-class:: sphx-glr-script-out .. code-block:: none [-0. -1. -3. ... -1. -3. -6.] .. GENERATED FROM PYTHON SOURCE LINES 244-249 .. code-block:: Python print(x.delete([5,8])) .. rst-class:: sphx-glr-script-out .. code-block:: none [ -0. -1. -3. ... -21. -28. -45.] .. GENERATED FROM PYTHON SOURCE LINES 250-255 .. code-block:: Python print(x.edges()) .. rst-class:: sphx-glr-script-out .. code-block:: none [ 0.5 -0.5 -2. ... -32. -40.5 -49.5] .. GENERATED FROM PYTHON SOURCE LINES 256-261 .. code-block:: Python print(x.internalEdges()) .. rst-class:: sphx-glr-script-out .. code-block:: none [ -0.5 -2. -4.5 ... -24.5 -32. -40.5] .. GENERATED FROM PYTHON SOURCE LINES 262-267 .. code-block:: Python print(x.firstNonZero()) .. rst-class:: sphx-glr-script-out .. code-block:: none 1 .. GENERATED FROM PYTHON SOURCE LINES 268-273 .. code-block:: Python print(x.lastNonZero()) .. rst-class:: sphx-glr-script-out .. code-block:: none 10 .. GENERATED FROM PYTHON SOURCE LINES 274-279 .. code-block:: Python print(x.abs()) .. rst-class:: sphx-glr-script-out .. code-block:: none [ 0. 1. 3. ... 28. 36. 45.] .. GENERATED FROM PYTHON SOURCE LINES 280-282 2D example __________ .. GENERATED FROM PYTHON SOURCE LINES 282-287 .. code-block:: Python x = StatArray(np.asarray([[0, -2, 3],[3, 0, -1],[1, 2, 0]])) print(x) .. rst-class:: sphx-glr-script-out .. code-block:: none [[ 0 -2 3] [ 3 0 -1] [ 1 2 0]] .. GENERATED FROM PYTHON SOURCE LINES 288-293 .. code-block:: Python print(x.insert(i=0, values=4)) .. rst-class:: sphx-glr-script-out .. code-block:: none [[ 4 4 4] [ 0 -2 3] [ 3 0 -1] [ 1 2 0]] .. GENERATED FROM PYTHON SOURCE LINES 294-299 .. code-block:: Python print(x.insert(i=[2, 3], values=5, axis=1)) .. rst-class:: sphx-glr-script-out .. code-block:: none [[ 0 -2 5 3 5] [ 3 0 5 -1 5] [ 1 2 5 0 5]] .. GENERATED FROM PYTHON SOURCE LINES 300-305 .. code-block:: Python print(x.insert(i=2, values=[10, 11, 12], axis=1)) .. rst-class:: sphx-glr-script-out .. code-block:: none [[ 0 -2 10 3] [ 3 0 11 -1] [ 1 2 12 0]] .. GENERATED FROM PYTHON SOURCE LINES 306-311 .. code-block:: Python print(x.prepend(999)) .. rst-class:: sphx-glr-script-out .. code-block:: none [[999 999 999] [ 0 -2 3] [ 3 0 -1] [ 1 2 0]] .. GENERATED FROM PYTHON SOURCE LINES 312-317 .. code-block:: Python print(x.prepend([999, 998, 997], axis=1)) .. rst-class:: sphx-glr-script-out .. code-block:: none [[999 998 997 0 -2 3] [999 998 997 3 0 -1] [999 998 997 1 2 0]] .. GENERATED FROM PYTHON SOURCE LINES 318-323 .. code-block:: Python print(x.append([[999, 998, 997]])) .. rst-class:: sphx-glr-script-out .. code-block:: none [[ 0 -2 3] [ 3 0 -1] [ 1 2 0] [999 998 997]] .. GENERATED FROM PYTHON SOURCE LINES 324-329 .. code-block:: Python print(x.resize([5,5])) .. rst-class:: sphx-glr-script-out .. code-block:: none [[ 0 -2 3 3 0] [-1 1 2 0 0] [-2 3 3 0 -1] [ 1 2 0 0 -2] [ 3 3 0 -1 1]] .. GENERATED FROM PYTHON SOURCE LINES 330-335 .. code-block:: Python print(x.delete(5)) .. rst-class:: sphx-glr-script-out .. code-block:: none [ 0 -2 3 ... 1 2 0] .. GENERATED FROM PYTHON SOURCE LINES 336-341 .. code-block:: Python print(x.delete(2, axis=0)) .. rst-class:: sphx-glr-script-out .. code-block:: none [[ 0 -2 3] [ 3 0 -1]] .. GENERATED FROM PYTHON SOURCE LINES 342-347 .. code-block:: Python print(x.firstNonZero(axis=0)) .. rst-class:: sphx-glr-script-out .. code-block:: none [1 0 0] .. GENERATED FROM PYTHON SOURCE LINES 348-353 .. code-block:: Python print(x.lastNonZero(axis=0)) .. rst-class:: sphx-glr-script-out .. code-block:: none [3 3 2] .. GENERATED FROM PYTHON SOURCE LINES 354-359 .. code-block:: Python print(x.firstNonZero(axis=1)) .. rst-class:: sphx-glr-script-out .. code-block:: none [1 0 0] .. GENERATED FROM PYTHON SOURCE LINES 360-365 .. code-block:: Python print(x.lastNonZero(axis=1)) .. rst-class:: sphx-glr-script-out .. code-block:: none [3 3 2] .. GENERATED FROM PYTHON SOURCE LINES 366-371 .. code-block:: Python print(x.abs()) .. rst-class:: sphx-glr-script-out .. code-block:: none [[0 2 3] [3 0 1] [1 2 0]] .. GENERATED FROM PYTHON SOURCE LINES 372-377 Plotting ++++++++ We can easily plot the StatArray with its built in plotting functions. All plotting functions can take matplotlib keywords .. GENERATED FROM PYTHON SOURCE LINES 377-385 .. code-block:: Python # The simplest is to just plot the array Density = StatArray(np.random.randn(100),name="Density",units="$\frac{g}{cc}$") Time = StatArray(np.linspace(0, 100, Density.size), name='Time', units='s') Depth = StatArray(np.random.exponential(size=Density.size), name='Depth', units='m') .. GENERATED FROM PYTHON SOURCE LINES 386-391 .. code-block:: Python plt.figure() _ = Density.plot(linewidth=0.5, marker='x', markersize=1.0) .. image-sg:: /examples/Statistics/images/sphx_glr_plot_StatArray_002.png :alt: plot StatArray :srcset: /examples/Statistics/images/sphx_glr_plot_StatArray_002.png :class: sphx-glr-single-img .. GENERATED FROM PYTHON SOURCE LINES 392-393 We can quickly plot a bar graph. .. GENERATED FROM PYTHON SOURCE LINES 393-398 .. code-block:: Python plt.figure() _ = Density.bar() .. image-sg:: /examples/Statistics/images/sphx_glr_plot_StatArray_003.png :alt: plot StatArray :srcset: /examples/Statistics/images/sphx_glr_plot_StatArray_003.png :class: sphx-glr-single-img .. GENERATED FROM PYTHON SOURCE LINES 399-400 We can scatter the contents of the StatArray if it is 1D .. GENERATED FROM PYTHON SOURCE LINES 400-405 .. code-block:: Python plt.figure() _ = Density.scatter(alpha=0.7) .. image-sg:: /examples/Statistics/images/sphx_glr_plot_StatArray_004.png :alt: plot StatArray :srcset: /examples/Statistics/images/sphx_glr_plot_StatArray_004.png :class: sphx-glr-single-img .. GENERATED FROM PYTHON SOURCE LINES 406-413 Histogram Equalization ______________________ A neat trick with colourmaps is histogram equalization. This approach forces all colours in the images to have an equal weight. This distorts the colour bar, but can really highlight the lower and higher ends of whatever you are plotting. Just add the equalize keyword! .. GENERATED FROM PYTHON SOURCE LINES 413-418 .. code-block:: Python plt.figure() _ = Density.scatter(alpha=0.7, equalize=True) .. image-sg:: /examples/Statistics/images/sphx_glr_plot_StatArray_005.png :alt: plot StatArray :srcset: /examples/Statistics/images/sphx_glr_plot_StatArray_005.png :class: sphx-glr-single-img .. GENERATED FROM PYTHON SOURCE LINES 419-422 Take the log base(x) of the data We can also take the data to a log, log10, log2, or a custom number! .. GENERATED FROM PYTHON SOURCE LINES 422-426 .. code-block:: Python plt.figure() _ = Density.scatter(alpha=0.7,edgecolor='k',log='e') # could also use log='e', log=2, log=x) where x is the base you require .. image-sg:: /examples/Statistics/images/sphx_glr_plot_StatArray_006.png :alt: plot StatArray :srcset: /examples/Statistics/images/sphx_glr_plot_StatArray_006.png :class: sphx-glr-single-img .. GENERATED FROM PYTHON SOURCE LINES 427-430 X and Y axes We can specify the x axis of the scatter plot. .. GENERATED FROM PYTHON SOURCE LINES 430-436 .. code-block:: Python plt.figure() _ = Density.scatter(x=Time, alpha=0.7, edgecolor='k') .. image-sg:: /examples/Statistics/images/sphx_glr_plot_StatArray_007.png :alt: plot StatArray :srcset: /examples/Statistics/images/sphx_glr_plot_StatArray_007.png :class: sphx-glr-single-img .. GENERATED FROM PYTHON SOURCE LINES 437-444 Notice that I never specified the y axis, so the y axis defaulted to the values in the StatArray. In this case, any operations applied to the colours, are also applied to the y axis, e.g. log=10. When I take the values of Density to log base 10, because I do not specify the y plotting locations, those locations are similarly affected. I can however force the y co-ordinates by specifying it as input. In the second subplot I explicitly plot distance on the y axis. In the first subplot, the y axis is the same as the colourbar. .. GENERATED FROM PYTHON SOURCE LINES 444-453 .. code-block:: Python plt.figure() ax1 = plt.subplot(211) Density.scatter(x=Time, alpha=0.7, edgecolor='k', log=10) plt.subplot(212, sharex=ax1) _ = Density.scatter(x=Time, y=Depth, alpha=0.7, edgecolor='k', log=10) .. image-sg:: /examples/Statistics/images/sphx_glr_plot_StatArray_008.png :alt: plot StatArray :srcset: /examples/Statistics/images/sphx_glr_plot_StatArray_008.png :class: sphx-glr-single-img .. GENERATED FROM PYTHON SOURCE LINES 454-457 Point sizes Since the plotting functions take matplotlib keywords, I can also specify the size of each points. .. GENERATED FROM PYTHON SOURCE LINES 459-474 .. code-block:: Python s = np.ceil(100*(np.abs(np.random.randn(Density.size)))) plt.figure() plt.tight_layout() ax1 = plt.subplot(211) Density.scatter(x=Time, y=Depth, s=s, alpha=0.7,edgecolor='k', legend_size=2) plt.subplot(212, sharex=ax1) #Density.scatter(x=Time, y=Depth, s=s, alpha=0.7,edgecolor='k', sizeLegend=[1.0, 100, 200, 300]) v = np.abs(Density)+1.0 _ = Density.scatter(x=Time, y=Depth, s=s, alpha=0.7,edgecolor='k', legend_size=[1.0, 100, 200, 300], log=10) .. image-sg:: /examples/Statistics/images/sphx_glr_plot_StatArray_009.png :alt: plot StatArray :srcset: /examples/Statistics/images/sphx_glr_plot_StatArray_009.png :class: sphx-glr-single-img .. GENERATED FROM PYTHON SOURCE LINES 475-476 Of course we can still take the log, or equalize the colour histogram .. GENERATED FROM PYTHON SOURCE LINES 476-481 .. code-block:: Python plt.figure() _ = Density.scatter(x=Time, y=Depth, s=s, alpha=0.7,edgecolor='k',equalize=True,log=10) .. image-sg:: /examples/Statistics/images/sphx_glr_plot_StatArray_010.png :alt: plot StatArray :srcset: /examples/Statistics/images/sphx_glr_plot_StatArray_010.png :class: sphx-glr-single-img .. GENERATED FROM PYTHON SOURCE LINES 482-483 Typically pcolor only works with 2D arrays. The StatArray has a pcolor method that will pcolor a 1D array .. GENERATED FROM PYTHON SOURCE LINES 483-495 .. code-block:: Python plt.figure() plt.subplot(221) Density.pcolor() plt.subplot(222) Density.pcolor(y=Time) plt.subplot(223) Density.pcolor(y=Time, flip=True) plt.subplot(224) _ = Density.pcolor(y=Time, log=10, equalize=True) .. image-sg:: /examples/Statistics/images/sphx_glr_plot_StatArray_011.png :alt: plot StatArray :srcset: /examples/Statistics/images/sphx_glr_plot_StatArray_011.png :class: sphx-glr-single-img .. GENERATED FROM PYTHON SOURCE LINES 496-499 We can add grid lines, and add opacity to each element in the pcolor image This is useful if the colour values need to be scaled by another variable e.g. variance. .. GENERATED FROM PYTHON SOURCE LINES 499-509 .. code-block:: Python plt.figure() plt.subplot(121) Density.pcolor(grid=True, cmap='jet') plt.subplot(122) a = np.linspace(1.0, 0.0, Density.size) _ = Density.pcolor(grid=True, alpha=a, cmap='jet') .. image-sg:: /examples/Statistics/images/sphx_glr_plot_StatArray_012.png :alt: plot StatArray :srcset: /examples/Statistics/images/sphx_glr_plot_StatArray_012.png :class: sphx-glr-single-img .. GENERATED FROM PYTHON SOURCE LINES 510-511 We can plot a histogram of the StatArray .. GENERATED FROM PYTHON SOURCE LINES 511-516 .. code-block:: Python plt.figure() _ = Density.hist(100) .. image-sg:: /examples/Statistics/images/sphx_glr_plot_StatArray_013.png :alt: plot StatArray :srcset: /examples/Statistics/images/sphx_glr_plot_StatArray_013.png :class: sphx-glr-single-img .. GENERATED FROM PYTHON SOURCE LINES 517-518 We can write the StatArray to a HDF5 file. HDF5 files are binary files that can include compression. They allow quick and easy access to parts of the file, and can also be written to and read from in parallel! .. GENERATED FROM PYTHON SOURCE LINES 518-523 .. code-block:: Python with h5py.File('1Dtest.h5','w') as f: Density.toHdf(f,'test') .. GENERATED FROM PYTHON SOURCE LINES 524-526 We can then read the StatArray from the file Here x is a new variable, that is read in from the hdf5 file we just wrote. .. GENERATED FROM PYTHON SOURCE LINES 526-533 .. code-block:: Python x = StatArray.fromHdf('1Dtest.h5', 'test') print('x has the same values as Density? ',np.all(x == Density)) x[2] = 5.0 # Change one of the values in x print('x has its own memory allocated (not a reference/pointer)? ', id(x) != id(Density)) .. rst-class:: sphx-glr-script-out .. code-block:: none x has the same values as Density? True x has its own memory allocated (not a reference/pointer)? True .. GENERATED FROM PYTHON SOURCE LINES 534-535 We can also define a 2D array .. GENERATED FROM PYTHON SOURCE LINES 535-540 .. code-block:: Python Density = StatArray(np.random.randn(50,100),"Density","$\frac{g}{cc}$") Density.summary .. rst-class:: sphx-glr-script-out .. code-block:: none "StatArray\nName: Density ($\\frac{g}{cc}$)\nAddress:['0x15d94b950']\nShape: (50, 100)\nValues: [[ 0.48957079 -0.67702192 0.62111102 ... 1.01637566 1.20209193\n 0.41799596]\n [ 0.44547561 0.48031438 -2.06904734 ... -0.01646932 0.61426381\n -0.54959319]\n [ 0.00538348 -0.78544357 1.95590857 ... -1.59213919 -0.1411701\n -0.9157846 ]\n ...\n [ 0.23210308 2.3900179 0.04757056 ... -0.26968731 -0.99625759\n 0.01622017]\n [-1.23929122 -0.82852075 0.59921237 ... -0.38958349 0.51283965\n 0.58431603]\n [ 0.48587545 0.68364767 1.74001602 ... -0.83376336 1.41033831\n -0.03525473]]\nMin: -3.657253370532356\nMax: 3.7262870836418522\nhas_posterior: False\n" .. GENERATED FROM PYTHON SOURCE LINES 541-544 The StatArray Class's functions work whether it is 1D or 2D We can still do a histogram .. GENERATED FROM PYTHON SOURCE LINES 544-549 .. code-block:: Python plt.figure() _ = Density.hist() .. image-sg:: /examples/Statistics/images/sphx_glr_plot_StatArray_014.png :alt: plot StatArray :srcset: /examples/Statistics/images/sphx_glr_plot_StatArray_014.png :class: sphx-glr-single-img .. GENERATED FROM PYTHON SOURCE LINES 550-551 And we can use pcolor to plot the 2D array .. GENERATED FROM PYTHON SOURCE LINES 551-556 .. code-block:: Python plt.figure() _ = Density.pcolor() .. image-sg:: /examples/Statistics/images/sphx_glr_plot_StatArray_015.png :alt: plot StatArray :srcset: /examples/Statistics/images/sphx_glr_plot_StatArray_015.png :class: sphx-glr-single-img .. GENERATED FROM PYTHON SOURCE LINES 557-560 The StatArray comes with extra plotting options Here we specify the x and y axes for the 2D array using two other 1D StatArrays .. GENERATED FROM PYTHON SOURCE LINES 560-567 .. code-block:: Python plt.figure() x = StatArray(np.arange(101),name='x Axis',units = 'mm') y = StatArray(np.arange(51),name='y Axis',units = 'elephants') _ = Density.pcolor(x=x, y=y) .. image-sg:: /examples/Statistics/images/sphx_glr_plot_StatArray_016.png :alt: plot StatArray :srcset: /examples/Statistics/images/sphx_glr_plot_StatArray_016.png :class: sphx-glr-single-img .. GENERATED FROM PYTHON SOURCE LINES 568-571 We can plot using a log10 scale, in this case, we have values that are less than or equal to 0.0. Plotting with the log option will by default mask any of those values, and will let you know that it has done so! .. GENERATED FROM PYTHON SOURCE LINES 571-576 .. code-block:: Python plt.figure() _ = Density.pcolor(x=x,y=y,log=2) .. image-sg:: /examples/Statistics/images/sphx_glr_plot_StatArray_017.png :alt: plot StatArray :srcset: /examples/Statistics/images/sphx_glr_plot_StatArray_017.png :class: sphx-glr-single-img .. GENERATED FROM PYTHON SOURCE LINES 577-581 A neat trick with colourmaps is histogram equalization. This approach forces all colours in the image to have an equal amount. This distorts the colours, but can really highlight the lower and higher ends of whatever you are plotting .. GENERATED FROM PYTHON SOURCE LINES 581-586 .. code-block:: Python plt.figure() _ = Density.pcolor(x=x, y=y, equalize=True) .. image-sg:: /examples/Statistics/images/sphx_glr_plot_StatArray_018.png :alt: plot StatArray :srcset: /examples/Statistics/images/sphx_glr_plot_StatArray_018.png :class: sphx-glr-single-img .. GENERATED FROM PYTHON SOURCE LINES 587-588 We can equalize the log10 plot too :) .. GENERATED FROM PYTHON SOURCE LINES 588-593 .. code-block:: Python plt.figure() _ = Density.pcolor(x=x,y=y,equalize=True, log=10) .. image-sg:: /examples/Statistics/images/sphx_glr_plot_StatArray_019.png :alt: plot StatArray :srcset: /examples/Statistics/images/sphx_glr_plot_StatArray_019.png :class: sphx-glr-single-img .. GENERATED FROM PYTHON SOURCE LINES 594-595 We can add opacity to each pixel in the image .. GENERATED FROM PYTHON SOURCE LINES 595-599 .. code-block:: Python a = StatArray(np.random.random(Density.shape), 'Opacity from 0.0 to 1.0') .. GENERATED FROM PYTHON SOURCE LINES 600-611 .. code-block:: Python plt.figure() ax1 = plt.subplot(131) ax = Density.pcolor(x=x, y=y, flipY=True, linewidth=0.1, colorbar=False) plt.subplot(132, sharex=ax1, sharey=ax1) ax = Density.pcolor(x=x, y=y, alpha=a, flipY=True, linewidth=0.1, colorbar=False) plt.subplot(133, sharex=ax1, sharey=ax1) _ = a.pcolor(x=x, y=y, flipY=True) .. image-sg:: /examples/Statistics/images/sphx_glr_plot_StatArray_020.png :alt: plot StatArray :srcset: /examples/Statistics/images/sphx_glr_plot_StatArray_020.png :class: sphx-glr-single-img .. GENERATED FROM PYTHON SOURCE LINES 612-613 If the array potentially has a lot of white space around the edges, we can trim the image .. GENERATED FROM PYTHON SOURCE LINES 613-625 .. code-block:: Python Density[:10, :] = 0.0 Density[-10:, :] = 0.0 Density[:, :10] = 0.0 Density[:, -10:] = 0.0 plt.figure() plt.subplot(121) Density.pcolor() plt.subplot(122) _ = Density.pcolor(trim=0.0) .. image-sg:: /examples/Statistics/images/sphx_glr_plot_StatArray_021.png :alt: plot StatArray :srcset: /examples/Statistics/images/sphx_glr_plot_StatArray_021.png :class: sphx-glr-single-img .. GENERATED FROM PYTHON SOURCE LINES 626-627 Create a stacked area plot of a 2D StatArray .. GENERATED FROM PYTHON SOURCE LINES 627-637 .. code-block:: Python A = StatArray(np.abs(np.random.randn(13,100)), name='Variable', units="units") x = StatArray(np.arange(100),name='x Axis',units = 'mm') plt.figure() ax1 = plt.subplot(211) A.stackedAreaPlot(x=x, axis=1) plt.subplot(212, sharex=ax1) _ = A.stackedAreaPlot(x=x, i=np.s_[[1,3,4],:], axis=1, labels=['a','b','c']) plt.show() .. image-sg:: /examples/Statistics/images/sphx_glr_plot_StatArray_022.png :alt: plot StatArray :srcset: /examples/Statistics/images/sphx_glr_plot_StatArray_022.png :class: sphx-glr-single-img .. rst-class:: sphx-glr-timing **Total running time of the script:** (0 minutes 2.756 seconds) .. _sphx_glr_download_examples_Statistics_plot_StatArray.py: .. only:: html .. container:: sphx-glr-footer sphx-glr-footer-example .. container:: sphx-glr-download sphx-glr-download-jupyter :download:`Download Jupyter notebook: plot_StatArray.ipynb ` .. container:: sphx-glr-download sphx-glr-download-python :download:`Download Python source code: plot_StatArray.py ` .. container:: sphx-glr-download sphx-glr-download-zip :download:`Download zipped: plot_StatArray.zip ` .. only:: html .. rst-class:: sphx-glr-signature `Gallery generated by Sphinx-Gallery `_