Pit #919
Open
Pit #919
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Collaborator
Author
|
@durgals , @reza-armuei , are you happy to review this PR for PIT? Might be best to start with the tutorial first: @nicholasloveday , please review the tutorials too if you have time |
tennlee
reviewed
Sep 15, 2025
| return value_at_rank | ||
|
|
||
|
|
||
| def rank_histogram( |
Collaborator
There was a problem hiding this comment.
This appears to be a version of the rank histogram which makes the assumption it will be working on ensemble data. Why not make/use a general-purpose rank histogram algorithm? I can see that being generally useful in its own right. Also, is there anything inherently probablistic about rank histograms? Perhaps people would like to use a rank histogram to compare deterministic forecasts as well? If this is only for ensembles and probablities it should not be named just 'rank histogram'
tennlee
reviewed
Sep 15, 2025
| """ | ||
| Methods for the probability integral transform (PIT) classes Pit and Pit_fcst_at_obs. | ||
|
|
||
| The basic approach follows Taggart (2023) and Gneiting and Ranjan (2013); see |
Collaborator
There was a problem hiding this comment.
This wording leads me to wonder whether the approach exactly matches the paper or whether it varies slightly, and what the variation might be. People might want the version from the paper as well as whatever variations may be present here.
tennlee
reviewed
Sep 15, 2025
| Calculates the left-hand and right-hand limits of the PIT distribution :math:`F`, | ||
| accessible via the attributes ``.left`` and ``.right``. | ||
| """ | ||
| if fcst_type not in ["ensemble", "cdf"]: |
Collaborator
There was a problem hiding this comment.
Consider using the match syntax for this, with _ resulting in raising an error, it will be easier to read/debug/follow. I'm happy to supply some suggested code if that would be helpful.
tennlee
reviewed
Sep 15, 2025
| return _variance(self.plotting_points()) | ||
|
|
||
|
|
||
| class Pit_fcst_at_obs: |
Collaborator
There was a problem hiding this comment.
The general naming convention for classes would be PitFcstAtObs
Collaborator
There was a problem hiding this comment.
@rob-taggart Do you think it would make sense to combine these into a single class instead of having two separate ones? The only difference I can see between the current classes is how they calculate the left-hand and right-hand limits of the PIT distribution, while the rest of the logic is duplicated. My idea is to create one class with an additional kwarg that lets the user specify which approach to use (please note that it is just a suggestion).
nicholasloveday
requested changes
Sep 16, 2025
Collaborator
There was a problem hiding this comment.
@rob-taggart I have reviewed both tutorials and they look great! I only have minor suggestions. Also, running black will hopefully get the automated checks passing again
| ], | ||
| "source": [ | ||
| "# generate observations for three locations, all from the standard normal distribution N(0,1)\n", | ||
| "np.random.seed(seed=44)\n", |
Collaborator
There was a problem hiding this comment.
You probably don't need to set the seed a second time.
| "In the first part of this tutorial, we look at some classic types of probabilistic (mis)-calibration\n", | ||
| "of ensemble forecasts and their characteristic rank histograms.\n", | ||
| "\n", | ||
| "We consider a process $Y_t$, where $t=0, 1, \\ldots, m$, which is generated by independent sampling from the normal distribution\n", |
Collaborator
There was a problem hiding this comment.
In this section, can you be more explicit about what the observations are for people who are not as familiar with this kind of synthetic experiment?
| "\n", | ||
| "The `scores` implementation of PIT also handles inputs in three basic forms:\n", | ||
| "\n", | ||
| "1. Inputs are the values $F(y)$ for each of the forecast-observation cases (or possibly the left-hand and right-hand limits of $F$ at $y$ if $F$ has discontinuities), as calculated by the user. This uses the class `scores.probability.Pit_fcst_at_obs`.\n", |
Collaborator
There was a problem hiding this comment.
Rather than Pit_fcst_at_obs, consider naming the class PitFcstAtObs to align the class name with PEP8
| "source": [ | ||
| "### PIT diagrams\n", | ||
| "\n", | ||
| "PIT diagrams (similar to PIT uniform probability plots) provide another way of visualising miscalibration. A PIT diagram is a plot of graph of the empirical CDF of the PIT values for the set of forecast cases. If the PIT values are uniformly distributed, as would be the case for a probabilistically calibrated system, then graph will lie close to diagonal line, which is the graph of the CDF of the standard uniform distribution.\n", |
Collaborator
There was a problem hiding this comment.
Suggested change
| "PIT diagrams (similar to PIT uniform probability plots) provide another way of visualising miscalibration. A PIT diagram is a plot of graph of the empirical CDF of the PIT values for the set of forecast cases. If the PIT values are uniformly distributed, as would be the case for a probabilistically calibrated system, then graph will lie close to diagonal line, which is the graph of the CDF of the standard uniform distribution.\n", | |
| "PIT diagrams (similar to PIT uniform probability plots) provide another way of visualising miscalibration. A PIT diagram is a plot of graph of the empirical CDF of the PIT values for the set of forecast cases. If the PIT values are uniformly distributed, as would be the case for a probabilistically calibrated system, then the graph will lie close to diagonal line, which is the graph of the CDF of the standard uniform distribution.\n", |
| "source": [ | ||
| "The PIT diagrams give the characteristic shapes for different types of miscalibration.\n", | ||
| "\n", | ||
| "Output from the `.plotting_points()` method gives duplicate values in the \"pit_x_value\" dimension. That is because the empirical distribution will has discontinuities (or veritical jumps in the plot). Duplicate values among coordinates can sometimes cause problems. The `plotting_points_parametric()` method provides the same information as `plotting_points` but without duplicate coordinate values.\n", |
Collaborator
There was a problem hiding this comment.
Suggested change
| "Output from the `.plotting_points()` method gives duplicate values in the \"pit_x_value\" dimension. That is because the empirical distribution will has discontinuities (or veritical jumps in the plot). Duplicate values among coordinates can sometimes cause problems. The `plotting_points_parametric()` method provides the same information as `plotting_points` but without duplicate coordinate values.\n", | |
| "Output from the `.plotting_points()` method gives duplicate values in the \"pit_x_value\" dimension. That is because the empirical distribution will have discontinuities (or veritical jumps in the plot). Duplicate values among coordinates can sometimes cause problems. The `plotting_points_parametric()` method provides the same information as `plotting_points` but without duplicate coordinate values.\n", |
| "\n", | ||
| "The classical definition of the PIT for a particular forecast-observation case is $\\mathrm{PIT}(F,y) = F(y)$, where $F$ is the predictive cumulative distribution function (CDF) and $y$ is the value of the observation. If the predictive CDFs are all continuous, and the forecast system is probabililistically calibrated, then the set of all PIT values $F(y)$ across the set of all forecast-observation cases will be approximately uniformly distributed on the unit interval $[0,1]$.\n", | ||
| "\n", | ||
| "This definition of PIT needs adjustment to account for cases when the predictive CDFs are discontinuous. Dicontinuous CDFs commonly occur with forecasts of precipitation, which is a mixture distribution with two components: the probability of dry conditions and the probability of precipitation given that precipitation occurs. The `scores` implementation of PIT handles discontinuous predictive CDFs as well as continuous predictive CDFs. It does this by keeping track of left-hand and right-hand limits of $F(y)$. We'll get onto this later in the tutorial.\n", |
Collaborator
There was a problem hiding this comment.
Suggested change
| "This definition of PIT needs adjustment to account for cases when the predictive CDFs are discontinuous. Dicontinuous CDFs commonly occur with forecasts of precipitation, which is a mixture distribution with two components: the probability of dry conditions and the probability of precipitation given that precipitation occurs. The `scores` implementation of PIT handles discontinuous predictive CDFs as well as continuous predictive CDFs. It does this by keeping track of left-hand and right-hand limits of $F(y)$. We'll get onto this later in the tutorial.\n", | |
| "This definition of PIT needs adjustment to account for cases when the predictive CDFs are discontinuous. Discontinuous CDFs commonly occur with forecasts of precipitation, which is a mixture distribution with two components: the probability of dry conditions and the probability of precipitation given that precipitation occurs. The `scores` implementation of PIT handles discontinuous predictive CDFs as well as continuous predictive CDFs. It does this by keeping track of left-hand and right-hand limits of $F(y)$. We'll get onto this later in the tutorial.\n", |
| "\n", | ||
| "1. Inputs are the values $F(y)$ for each of the forecast-observation cases (or possibly the left-hand and right-hand limits of $F$ at $y$ if $F$ has discontinuities), as calculated by the user. This uses the class `scores.probability.Pit_fcst_at_obs`.\n", | ||
| "2. Inputs are the forecasts and corresponding observations, where the forecasts are in the form of an ensemble. In this case, the predictive CDFs are treated as empirical CDFs derived from the ensemble values. This uses the class `scores.probability.Pit` with the `fcst_type='ensemble'` option.\n", | ||
| "3. Inputs are the forecasts and corresponding observations, where the forecasts are arrays specifing the values (or left- and right-hand limits) of each predictive CDF at a specified set of points $x$. If an observation $y$ is not one of specified points $x$, then the value of $f$ at $y$ is deterrmined using linear interpolation. This uses the class `scores.probability.Pit` with the `fcst_type='cdf'` option.\n", |
Collaborator
There was a problem hiding this comment.
Suggested change
| "3. Inputs are the forecasts and corresponding observations, where the forecasts are arrays specifing the values (or left- and right-hand limits) of each predictive CDF at a specified set of points $x$. If an observation $y$ is not one of specified points $x$, then the value of $f$ at $y$ is deterrmined using linear interpolation. This uses the class `scores.probability.Pit` with the `fcst_type='cdf'` option.\n", | |
| "3. Inputs are the forecasts and corresponding observations, where the forecasts are arrays specifying the values (or left- and right-hand limits) of each predictive CDF at a specified set of points $x$. If an observation $y$ is not one of specified points $x$, then the value of $f$ at $y$ is determined using linear interpolation. This uses the class `scores.probability.Pit` with the `fcst_type='cdf'` option.\n", |
| "\n", | ||
| "Both PIT classes then have identical methods for calculating statistics for PIT histograms, PIT diagrams (i.e. PIT uniform probability plots), the variance and expected value of PIT across all forecast cases, and the so-called alpha score. These methods and classes are demonstrated in this tutorial.\n", | ||
| "\n", | ||
| "For ensemble forecasts, `scores.probability.rank_histogram` also provides a method for computing rank histograms, which are closely related to PIT histograms. See the [tutorial on rank histograms](./Rank_Histogram.ipynbipynb) for further information.\n", |
Collaborator
There was a problem hiding this comment.
Suggested change
| "For ensemble forecasts, `scores.probability.rank_histogram` also provides a method for computing rank histograms, which are closely related to PIT histograms. See the [tutorial on rank histograms](./Rank_Histogram.ipynbipynb) for further information.\n", | |
| "For ensemble forecasts, `scores.probability.rank_histogram` also provides a method for computing rank histograms, which are closely related to PIT histograms. See the [tutorial on rank histograms](./Rank_Histogram.ipynb) for further information.\n", |
| "For ensemble forecasts, `scores.probability.rank_histogram` also provides a method for computing rank histograms, which are closely related to PIT histograms. See the [tutorial on rank histograms](./Rank_Histogram.ipynbipynb) for further information.\n", | ||
| "\n", | ||
| "\n", | ||
| "## Case study 1: Predictive CDFs are parametric distributions and continous\n", |
Collaborator
There was a problem hiding this comment.
Suggested change
| "## Case study 1: Predictive CDFs are parametric distributions and continous\n", | |
| "## Case study 1: Predictive CDFs are parametric distributions and continuous\n", |
Comment on lines
+41
to
+43
| "4. **Over-dispersed forecaster**: forceasts are $N(\\mu_t, (1.3)^2)$. That is, forecast distribution is wider than the true distribution. \n", | ||
| "5. **Under-dispersed forecaster**: forceasts are $N(\\mu_t, (1/1.3)^2)$. That is, forecast distribution is narrower than the true distribution. \n", | ||
| "6. **Calibrated (marginal) forecaster**: forceasts are $N(0, 2)$. That is, this forecast distribution well-calibrated but has no knowledge of $\\mu_t$. This is equivalent to issuing the climatological forecast for a meteorological process. Forecasts are reliable over the long-run but not accurate.\n", |
Collaborator
There was a problem hiding this comment.
Suggested change
| "4. **Over-dispersed forecaster**: forceasts are $N(\\mu_t, (1.3)^2)$. That is, forecast distribution is wider than the true distribution. \n", | |
| "5. **Under-dispersed forecaster**: forceasts are $N(\\mu_t, (1/1.3)^2)$. That is, forecast distribution is narrower than the true distribution. \n", | |
| "6. **Calibrated (marginal) forecaster**: forceasts are $N(0, 2)$. That is, this forecast distribution well-calibrated but has no knowledge of $\\mu_t$. This is equivalent to issuing the climatological forecast for a meteorological process. Forecasts are reliable over the long-run but not accurate.\n", | |
| "4. **Over-dispersed forecaster**: foremasts are $N(\\mu_t, (1.3)^2)$. That is, forecast distribution is wider than the true distribution. \n", | |
| "5. **Under-dispersed forecaster**: forecasts are $N(\\mu_t, (1/1.3)^2)$. That is, forecast distribution is narrower than the true distribution. \n", | |
| "6. **Calibrated (marginal) forecaster**: foremasts are $N(0, 2)$. That is, this forecast distribution well-calibrated but has no knowledge of $\\mu_t$. This is equivalent to issuing the climatological forecast for a meteorological process. Forecasts are reliable over the long-run but not accurate.\n", |
Collaborator
There was a problem hiding this comment.
Suggested change
| "4. **Over-dispersed forecaster**: forceasts are $N(\\mu_t, (1.3)^2)$. That is, forecast distribution is wider than the true distribution. \n", | |
| "5. **Under-dispersed forecaster**: forceasts are $N(\\mu_t, (1/1.3)^2)$. That is, forecast distribution is narrower than the true distribution. \n", | |
| "6. **Calibrated (marginal) forecaster**: forceasts are $N(0, 2)$. That is, this forecast distribution well-calibrated but has no knowledge of $\\mu_t$. This is equivalent to issuing the climatological forecast for a meteorological process. Forecasts are reliable over the long-run but not accurate.\n", | |
| "4. **Over-dispersed forecaster**: forecasts are $N(\\mu_t, (1.3)^2)$. That is, forecast distribution is wider than the true distribution. \n", | |
| "5. **Under-dispersed forecaster**: forecasts are $N(\\mu_t, (1/1.3)^2)$. That is, forecast distribution is narrower than the true distribution. \n", | |
| "6. **Calibrated (marginal) forecaster**: forecasts are $N(0, 2)$. That is, this forecast distribution well-calibrated but has no knowledge of $\\mu_t$. This is equivalent to issuing the climatological forecast for a meteorological process. Forecasts are reliable over the long-run but not accurate.\n", |
reza-armuei
reviewed
Dec 8, 2025
Collaborator
reza-armuei
left a comment
reza-armuei
left a comment
There was a problem hiding this comment.
Thanks @rob-taggart - Great work!
Added few comments/suggestions for your consideration.
| "source": [ | ||
| "# The Probability Integral Transform (PIT)\n", | ||
| "\n", | ||
| "The probability intergral transform (PIT) provides a method for assessing to what extent a forecast system is probabililistically calibrated, and may help diagnose particular types of miscalibration. A forecast system is probabilistically calibrated if random draws from its predictive distributions are statistically indistinguishable from the corresponding observations.\n", |
Collaborator
There was a problem hiding this comment.
Suggested change
| "The probability intergral transform (PIT) provides a method for assessing to what extent a forecast system is probabililistically calibrated, and may help diagnose particular types of miscalibration. A forecast system is probabilistically calibrated if random draws from its predictive distributions are statistically indistinguishable from the corresponding observations.\n", | |
| "The probability integral transform (PIT) provides a method for assessing to what extent a forecast system is probabilistically calibrated, and may help diagnose particular types of miscalibration. A forecast system is probabilistically calibrated if random draws from its predictive distributions are statistically indistinguishable from the corresponding observations.\n", |
| "\n", | ||
| "The probability intergral transform (PIT) provides a method for assessing to what extent a forecast system is probabililistically calibrated, and may help diagnose particular types of miscalibration. A forecast system is probabilistically calibrated if random draws from its predictive distributions are statistically indistinguishable from the corresponding observations.\n", | ||
| "\n", | ||
| "The classical definition of the PIT for a particular forecast-observation case is $\\mathrm{PIT}(F,y) = F(y)$, where $F$ is the predictive cumulative distribution function (CDF) and $y$ is the value of the observation. If the predictive CDFs are all continuous, and the forecast system is probabililistically calibrated, then the set of all PIT values $F(y)$ across the set of all forecast-observation cases will be approximately uniformly distributed on the unit interval $[0,1]$.\n", |
Collaborator
There was a problem hiding this comment.
Suggested change
| "The classical definition of the PIT for a particular forecast-observation case is $\\mathrm{PIT}(F,y) = F(y)$, where $F$ is the predictive cumulative distribution function (CDF) and $y$ is the value of the observation. If the predictive CDFs are all continuous, and the forecast system is probabililistically calibrated, then the set of all PIT values $F(y)$ across the set of all forecast-observation cases will be approximately uniformly distributed on the unit interval $[0,1]$.\n", | |
| "The classical definition of the PIT for a particular forecast-observation case is $\\mathrm{PIT}(F,y) = F(y)$, where $F$ is the predictive cumulative distribution function (CDF) and $y$ is the value of the observation. If the predictive CDFs are all continuous, and the forecast system is probabilistically calibrated, then the set of all PIT values $F(y)$ across the set of all forecast-observation cases will be approximately uniformly distributed on the unit interval $[0,1]$.\n", |
Comment on lines
+41
to
+43
| "4. **Over-dispersed forecaster**: forceasts are $N(\\mu_t, (1.3)^2)$. That is, forecast distribution is wider than the true distribution. \n", | ||
| "5. **Under-dispersed forecaster**: forceasts are $N(\\mu_t, (1/1.3)^2)$. That is, forecast distribution is narrower than the true distribution. \n", | ||
| "6. **Calibrated (marginal) forecaster**: forceasts are $N(0, 2)$. That is, this forecast distribution well-calibrated but has no knowledge of $\\mu_t$. This is equivalent to issuing the climatological forecast for a meteorological process. Forecasts are reliable over the long-run but not accurate.\n", |
Collaborator
There was a problem hiding this comment.
Suggested change
| "4. **Over-dispersed forecaster**: forceasts are $N(\\mu_t, (1.3)^2)$. That is, forecast distribution is wider than the true distribution. \n", | |
| "5. **Under-dispersed forecaster**: forceasts are $N(\\mu_t, (1/1.3)^2)$. That is, forecast distribution is narrower than the true distribution. \n", | |
| "6. **Calibrated (marginal) forecaster**: forceasts are $N(0, 2)$. That is, this forecast distribution well-calibrated but has no knowledge of $\\mu_t$. This is equivalent to issuing the climatological forecast for a meteorological process. Forecasts are reliable over the long-run but not accurate.\n", | |
| "4. **Over-dispersed forecaster**: forecasts are $N(\\mu_t, (1.3)^2)$. That is, forecast distribution is wider than the true distribution. \n", | |
| "5. **Under-dispersed forecaster**: forecasts are $N(\\mu_t, (1/1.3)^2)$. That is, forecast distribution is narrower than the true distribution. \n", | |
| "6. **Calibrated (marginal) forecaster**: forecasts are $N(0, 2)$. That is, this forecast distribution well-calibrated but has no knowledge of $\\mu_t$. This is equivalent to issuing the climatological forecast for a meteorological process. Forecasts are reliable over the long-run but not accurate.\n", |
| "outputs": [ | ||
| { | ||
| "data": { | ||
| "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYYAAAGXCAYAAAC3CvfbAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjkuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8hTgPZAAAACXBIWXMAAA9hAAAPYQGoP6dpAABRDklEQVR4nO3dd1gUV/828HupCypYUBDFhlhBwY6a2HiCDcVKbBhbNIlYUBSNYDSJoFesUWMM+rM8GrvEFh8NdoRYQNHYYl1MBDQqCCognPcPXzZOaLsEmVm8P9e1V9izZ5d7IvCdOTNzjkoIIUBERPT/GckdgIiIlIWFgYiIJFgYiIhIgoWBiIgkWBiIiEiChYGIiCRYGIiISMJE7gBKlJ2djT///BPlypWDSqWSOw4R0b8mhMCzZ89gb28PI6OCjwlYGPLw559/wsHBQe4YRETFLj4+HtWrVy+wDwtDHsqVKwfg9f9AKysrmdMQEf17KSkpcHBw0P59KwgLQx5yho+srKxYGIioVNFleJwnn4mISIKFgYiIJFgYiIhIgoWBiIgkWBiIiEiChYGIiCRYGIiISIKFgYiIJHiDG5U6tQL3yx0hX3dDe8gdgahQPGIgIiIJFgYiIpJgYSAiIgkWBiIikmBhICIiCRYGIiKSYGEgIiIJFgYiIpJgYSAiIgkWBiIikmBhICIiCRYGIiKSYGEgIiIJFgYiIpJgYSAiIgnZC8OKFStQq1YtqNVqtG7dGmfOnCmw//bt29GgQQOo1Wq4uLjgwIEDktdTU1Mxfvx4VK9eHRYWFmjUqBFWrVr1NjeBiKhUkbUwbN26Ff7+/pg9ezZiYmLQtGlTeHp6IikpKc/+p0+fxqBBgzBq1CjExsbC29sb3t7euHz5sraPv78/Dh48iP/+97+4evUqJk2ahPHjx2PPnj0ltVlERAZNJYQQcn3z1q1bo2XLlli+fDkAIDs7Gw4ODvDz80NgYGCu/j4+PkhLS8O+ffu0bW3atIGrq6v2qMDZ2Rk+Pj4ICgrS9mnevDm6deuGr776SqdcKSkpsLa2RnJyMqysrP7NJpIMuIIbUW76/F2TbWnPjIwMnD9/HjNmzNC2GRkZwcPDA1FRUXm+JyoqCv7+/pI2T09PhIeHa5+3bdsWe/bswciRI2Fvb49jx47hxo0bWLx4cb5Z0tPTkZ6ern2ekpJSxK0qHfiHlejdJttQ0qNHj5CVlQVbW1tJu62tLRISEvJ8T0JCQqH9v/32WzRq1AjVq1eHmZkZunbtihUrVuD999/PN0tISAisra21DwcHh3+xZUREhk32k8/F7dtvv0V0dDT27NmD8+fPY+HChfjss8/wyy+/5PueGTNmIDk5WfuIj48vwcRERMoi21CSjY0NjI2NkZiYKGlPTEyEnZ1dnu+xs7MrsP+LFy8wc+ZM7N69Gz16vB5yaNKkCS5cuIBvvvkGHh4eeX6uubk5zM3N/+0mERGVCrIdMZiZmaF58+aIiIjQtmVnZyMiIgLu7u55vsfd3V3SHwAOHz6s7Z+ZmYnMzEwYGUk3y9jYGNnZ2cW8BUREpZNsRwzA60tLhw8fjhYtWqBVq1ZYsmQJ0tLSMGLECACAr68vqlWrhpCQEADAxIkT0aFDByxcuBA9evTAli1bcO7cOaxevRoAYGVlhQ4dOiAgIAAWFhaoWbMmjh8/jg0bNmDRokWybScRkSGRtTD4+Pjg4cOHCA4ORkJCAlxdXXHw4EHtCWaNRiPZ+2/bti02b96MWbNmYebMmXByckJ4eDicnZ21fbZs2YIZM2ZgyJAhePz4MWrWrImvv/4a48aNK/HtIyIyRLLex6BU7/p9DIZ+uaqh5yd6G/T5u1bqrkoiIqJ/h4WBiIgkWBiIiEiChYGIiCRYGIiISIKFgYiIJFgYiIhIgoWBiIgkWBiIiEiChYGIiCRYGIiISIKFgYiIJFgYiIhIgoWBiIgkWBiIiEiChYGIiCRYGIiISIKFgYiIJFgYiIhIgoWBiIgkWBiIiEiChYGIiCRYGIiISIKFgYiIJFgYiIhIgoWBiIgkWBiIiEiChYGIiCRYGIiISIKFgYiIJFgYiIhIgoWBiIgkWBiIiEiChYGIiCRYGIiISIKFgYiIJFgYiIhIgoWBiIgkWBiIiEhC9sKwYsUK1KpVC2q1Gq1bt8aZM2cK7L99+3Y0aNAAarUaLi4uOHDgQK4+V69eRa9evWBtbY0yZcqgZcuW0Gg0b2sTiIhKlSIVhlu3bmHWrFkYNGgQkpKSAAA///wzfvvtN70+Z+vWrfD398fs2bMRExODpk2bwtPTU/uZ/3T69GkMGjQIo0aNQmxsLLy9veHt7Y3Lly9LsrVv3x4NGjTAsWPHEBcXh6CgIKjV6qJsKhHRO0clhBD6vOH48ePo1q0b2rVrhxMnTuDq1auoU6cOQkNDce7cOezYsUPnz2rdujVatmyJ5cuXAwCys7Ph4OAAPz8/BAYG5urv4+ODtLQ07Nu3T9vWpk0buLq6YtWqVQCADz/8EKampti4caM+myWRkpICa2trJCcnw8rKqsifY6hqBe6XO0K+7ob2KLSPoecnehv0+bum9xFDYGAgvvrqKxw+fBhmZmba9s6dOyM6Olrnz8nIyMD58+fh4eHxdxgjI3h4eCAqKirP90RFRUn6A4Cnp6e2f3Z2Nvbv34969erB09MTVapUQevWrREeHl5glvT0dKSkpEgeRETvKr0Lw6VLl9CnT59c7VWqVMGjR490/pxHjx4hKysLtra2knZbW1skJCTk+Z6EhIQC+yclJSE1NRWhoaHo2rUrDh06hD59+qBv3744fvx4vllCQkJgbW2tfTg4OOi8HUREpY3ehaF8+fJ48OBBrvbY2FhUq1atWEIVVXZ2NgCgd+/emDx5MlxdXREYGIiePXtqh5ryMmPGDCQnJ2sf8fHxJRWZiEhx9C4MH374IaZPn46EhASoVCpkZ2cjMjISU6dOha+vr86fY2NjA2NjYyQmJkraExMTYWdnl+d77OzsCuxvY2MDExMTNGrUSNKnYcOGBV6VZG5uDisrK8mDiOhdpXdhmDdvHho0aAAHBwekpqaiUaNGeP/999G2bVvMmjVL588xMzND8+bNERERoW3Lzs5GREQE3N3d83yPu7u7pD8AHD58WNvfzMwMLVu2xPXr1yV9bty4gZo1a+qcjYjoXWaiT2chBBISErBs2TIEBwfj0qVLSE1NhZubG5ycnPT+5v7+/hg+fDhatGiBVq1aYcmSJUhLS8OIESMAAL6+vqhWrRpCQkIAABMnTkSHDh2wcOFC9OjRA1u2bMG5c+ewevVq7WcGBATAx8cH77//Pjp16oSDBw9i7969OHbsmN75iIjeRXoXhrp16+K3336Dk5PTvz5J6+Pjg4cPHyI4OBgJCQlwdXXFwYMHtSeYNRoNjIz+Pqhp27YtNm/ejFmzZmHmzJlwcnJCeHg4nJ2dtX369OmDVatWISQkBBMmTED9+vWxc+dOtG/f/l9lJSJ6V+h9H0Pjxo2xZs0atGnT5m1lkh3vYzDs+wAMPT/R2/BW72MIDQ1FQECA5G5jIiIqPfQaSgJej/s/f/4cTZs2hZmZGSwsLCSvP378uNjCERFRydO7MCxZsuQtxCAiIqXQuzAMHz78beQgIiKF0LswAEBWVhbCw8Nx9epVAK9PSPfq1QvGxsbFGo6IiEqe3oXh5s2b6N69O/744w/Ur18fwOu5hhwcHLB//344OjoWe0giIio5el+VNGHCBDg6OiI+Ph4xMTGIiYmBRqNB7dq1MWHChLeRkYiISpDeRwzHjx9HdHQ0KlasqG2rVKkSQkND0a5du2INR0REJU/vIwZzc3M8e/YsV3tqaqpkfQYiIjJMeheGnj174uOPP8avv/4KIQSEEIiOjsa4cePQq1evt5GRiIhKkN6FYdmyZXB0dIS7uzvUajXUajXatWuHunXrYunSpW8jIxERlSC9zzGUL18eP/30E27evKm9XLVhw4aoW7dusYcjIqKSV6T7GACgbt26LAZERKWQ3kNJ/fr1w/z583O1L1iwAAMGDCiWUEREJB+9C8OJEyfQvXv3XO3dunXDiRMniiUUERHJR+/CkN9lqaampkhJSSmWUEREJB+9C4OLiwu2bt2aq33Lli1o1KhRsYQiIiL56H3yOSgoCH379sWtW7fQuXNnAEBERAR+/PFHbN++vdgDEhFRydK7MHh5eSE8PBzz5s3Djh07YGFhgSZNmuCXX35Bhw4d3kZGIiIqQUW6XLVHjx7o0YNr1xIRlUZ6n2OIj4/H/fv3tc/PnDmDSZMmYfXq1cUajIiI5KF3YRg8eDCOHj0KAEhISICHhwfOnDmDzz//HHPnzi32gEREVLL0LgyXL19Gq1atAADbtm2Di4sLTp8+jU2bNmHdunXFnY+IiEqY3oUhMzMT5ubmAIBffvlFO6NqgwYN8ODBg+JNR0REJU7vwtC4cWOsWrUKJ0+exOHDh9G1a1cAwJ9//olKlSoVe0AiIipZeheG+fPn4/vvv0fHjh0xaNAgNG3aFACwZ88e7RATEREZLr0vV+3YsSMePXqElJQUVKhQQdv+8ccfw9LSsljDERFRySvSfQzGxsaSogAAtWrVKo48REQkM72HkoiIqHRjYSAiIgkWBiIikmBhICIiiSKdfE5LS8Px48eh0WiQkZEheW3ChAnFEoyIiOShd2GIjY1F9+7d8fz5c6SlpaFixYp49OgRLC0tUaVKFRYGIiIDp/dQ0uTJk+Hl5YUnT57AwsIC0dHRuHfvHpo3b45vvvnmbWQkIqISpHdhuHDhAqZMmQIjIyMYGxsjPT0dDg4OWLBgAWbOnPk2MhIRUQnSuzCYmprCyOj126pUqQKNRgMAsLa2Rnx8fPGmIyKiEqf3OQY3NzecPXsWTk5O6NChA4KDg/Ho0SNs3LgRzs7ObyMjERGVIL0Lw7x58/Ds2TMAwNdffw1fX1988skncHJywtq1a4s9ING7qFbgfrkj5OtuKJf1Le30Hkpq0aIFOnXqBOD1UNLBgweRkpKC8+fPa2da1deKFStQq1YtqNVqtG7dGmfOnCmw//bt29GgQQOo1Wq4uLjgwIED+fYdN24cVCoVlixZUqRsRETvGtlvcNu6dSv8/f0xe/ZsxMTEoGnTpvD09ERSUlKe/U+fPo1BgwZh1KhRiI2Nhbe3N7y9vXH58uVcfXfv3o3o6GjY29u/7c0gIio19B5KSkxMxNSpUxEREYGkpCQIISSvZ2Vl6fV5ixYtwpgxYzBixAgAwKpVq7B//36sXbsWgYGBufovXboUXbt2RUBAAADgyy+/xOHDh7F8+XKsWrVK2++PP/6An58f/ve//6FHDx76EpUkDoUZNr0Lw0cffQSNRoOgoCBUrVoVKpWqyN88IyMD58+fx4wZM7RtRkZG8PDwQFRUVJ7viYqKgr+/v6TN09MT4eHh2ufZ2dkYNmwYAgIC0Lhx40JzpKenIz09Xfs8JSVFzy0hIio99C4Mp06dwsmTJ+Hq6vqvv/mjR4+QlZUFW1tbSbutrS2uXbuW53sSEhLy7J+QkKB9Pn/+fJiYmOh8F3ZISAjmzJmjZ3oiotJJ73MMDg4OuYaPlOT8+fNYunQp1q1bp/PRzIwZM5CcnKx98H4MInqX6V0YlixZgsDAQNy9e/dff3MbGxsYGxsjMTFR0p6YmAg7O7s832NnZ1dg/5MnTyIpKQk1atSAiYkJTExMcO/ePUyZMiXfVebMzc1hZWUleRARvat0GkqqUKGCZO87LS0Njo6OsLS0hKmpqaTv48ePdf7mZmZmaN68OSIiIuDt7Q3g9fmBiIgIjB8/Ps/3uLu7IyIiApMmTdK2HT58GO7u7gCAYcOGwcPDQ/IeT09PDBs2THuCm4iI8qdTYXib9wD4+/tj+PDhaNGiBVq1aoUlS5YgLS1N+0fc19cX1apVQ0hICABg4sSJ6NChAxYuXIgePXpgy5YtOHfuHFavXg0AqFSpEipVqiT5HqamprCzs0P9+vXf2nYQEZUWOhWG4cOHv7UAPj4+ePjwIYKDg5GQkABXV1ccPHhQe4JZo9Fo52YCgLZt22Lz5s2YNWsWZs6cCScnJ4SHh3M6DiKiYlKkhXqysrKwe/duXL16FQDQqFEj9O7dGyYmRfo4jB8/Pt+ho2PHjuVqGzBgAAYMGKDz5xfH+RAioneF3n/Jf/vtN/Tq1QsJCQnaoZn58+ejcuXK2Lt3L/fciYgMnN5XJY0ePRqNGzfG/fv3ERMTg5iYGMTHx6NJkyb4+OOP30ZGIiIqQXofMVy4cAHnzp1DhQoVtG0VKlTA119/jZYtWxZrOCIiKnl6HzHUq1cv130EAJCUlIS6desWSygiIpKP3oUhJCQEEyZMwI4dO3D//n3cv38fO3bswKRJkzB//nykpKRoH0REZHj0Hkrq2bMnAGDgwIHam95ypsjw8vLSPlepVHrPtEpERPLTuzAcPXr0beQgIiKF0LswdOjQ4W3kICIihdCpMMTFxen8gU2aNClyGCIikp9OhcHV1RUqlarQ6bZ5XoGIyPDpVBju3LnztnMQEZFC6FQYatas+bZzEBGRQhRt1jsAV65cgUajQUZGhqS9V69e/zoUERHJR+/CcPv2bfTp0weXLl2SnHfIuaeB5xiIiAyb3nc+T5w4EbVr10ZSUhIsLS3x22+/4cSJE2jRokWeU2QTEZFh0fuIISoqCkeOHIGNjQ2MjIxgZGSE9u3ba6fKiI2NfRs5iYiohOh9xJCVlYVy5coBAGxsbPDnn38CeH2C+vr168WbjoiISpzeRwzOzs64ePEiateujdatW2PBggUwMzPD6tWrUadOnbeRkYiISpDehWHWrFlIS0sDAMydOxc9e/bEe++9h0qVKmHr1q3FHpCIiEqW3oXB09NT+3XdunVx7do1PH78GBUqVNBemURERIaryPcxvKlixYrF8TFERKQAep98JiKi0o2FgYiIJFgYiIhIgoWBiIgkWBiIiEiChYGIiCRYGIiISIKFgYiIJFgYiIhIgoWBiIgkWBiIiEiChYGIiCRYGIiISIKFgYiIJFgYiIhIgoWBiIgkWBiIiEiChYGIiCQUURhWrFiBWrVqQa1Wo3Xr1jhz5kyB/bdv344GDRpArVbDxcUFBw4c0L6WmZmJ6dOnw8XFBWXKlIG9vT18fX3x559/vu3NICIqFWQvDFu3boW/vz9mz56NmJgYNG3aFJ6enkhKSsqz/+nTpzFo0CCMGjUKsbGx8Pb2hre3Ny5fvgwAeP78OWJiYhAUFISYmBjs2rUL169fR69evUpys4iIDJbshWHRokUYM2YMRowYgUaNGmHVqlWwtLTE2rVr8+y/dOlSdO3aFQEBAWjYsCG+/PJLNGvWDMuXLwcAWFtb4/Dhwxg4cCDq16+PNm3aYPny5Th//jw0Gk1JbhoRkUGStTBkZGTg/Pnz8PDw0LYZGRnBw8MDUVFReb4nKipK0h8APD098+0PAMnJyVCpVChfvnyer6enpyMlJUXyICJ6V8laGB49eoSsrCzY2tpK2m1tbZGQkJDnexISEvTq//LlS0yfPh2DBg2ClZVVnn1CQkJgbW2tfTg4OBRha4iISgfZh5LepszMTAwcOBBCCHz33Xf59psxYwaSk5O1j/j4+BJMSUSkLCZyfnMbGxsYGxsjMTFR0p6YmAg7O7s832NnZ6dT/5yicO/ePRw5ciTfowUAMDc3h7m5eRG3goiodJH1iMHMzAzNmzdHRESEti07OxsRERFwd3fP8z3u7u6S/gBw+PBhSf+covD777/jl19+QaVKld7OBhARlUKyHjEAgL+/P4YPH44WLVqgVatWWLJkCdLS0jBixAgAgK+vL6pVq4aQkBAAwMSJE9GhQwcsXLgQPXr0wJYtW3Du3DmsXr0awOui0L9/f8TExGDfvn3IysrSnn+oWLEizMzM5NlQIiIDIXth8PHxwcOHDxEcHIyEhAS4urri4MGD2hPMGo0GRkZ/H9i0bdsWmzdvxqxZszBz5kw4OTkhPDwczs7OAIA//vgDe/bsAQC4urpKvtfRo0fRsWPHEtkuIiJDJXthAIDx48dj/Pjxeb527NixXG0DBgzAgAED8uxfq1YtCCGKMx4R0TulVF+VRERE+mNhICIiCRYGIiKSYGEgIiIJFgYiIpJgYSAiIgkWBiIikmBhICIiCRYGIiKSYGEgIiIJFgYiIpJgYSAiIgkWBiIikmBhICIiCRYGIiKSYGEgIiIJFgYiIpJgYSAiIgkWBiIikmBhICIiCRYGIiKSYGEgIiIJFgYiIpJgYSAiIgkWBiIikmBhICIiCRYGIiKSYGEgIiIJFgYiIpJgYSAiIgkWBiIikmBhICIiCRYGIiKSYGEgIiIJE7kDlDa1AvfLHSFfd0N7yB2BiAwAjxiIiEiChYGIiCQ4lERElId3eViYRwxERCTBwkBERBKKKAwrVqxArVq1oFar0bp1a5w5c6bA/tu3b0eDBg2gVqvh4uKCAwcOSF4XQiA4OBhVq1aFhYUFPDw88Pvvv7/NTSAiKjVkP8ewdetW+Pv7Y9WqVWjdujWWLFkCT09PXL9+HVWqVMnV//Tp0xg0aBBCQkLQs2dPbN68Gd7e3oiJiYGzszMAYMGCBVi2bBnWr1+P2rVrIygoCJ6enrhy5QrUarXO2dLS0mBsbJyr3djYWPI5aWlp2q+zM15KO6tUMDI1z//1gvpmvgREfn0BI1N1kfq+ePEC2dnZ+ed4Q3ZmOiDy+2DAyOzvzxWvMiAK+Fx9+qpMzaFSqf5/30yI7CwA0v/XOSwtLbV909PTC/x/rDI1g0r1en9IZGVCZGUVT18TU6iMjAvtm5aWBrVarf25yszMREZGRq5+Odsg/dxXEFmvdMuQnQXxKjP/vsYmUBmbFKlvXv8GOUxNTWFmZqbj5xpDZWz6uq/IhsjM/f+hSH2NjKEyyekrIDLTJa+/md/ExATm5ubavs+fP9e+9s+fI5WREVQmZvm+XtS+/+ZvxPPnzyHy+f1UqVSwtLSU9NWVSuT3qSWkdevWaNmyJZYvXw4AyM7OhoODA/z8/BAYGJirv4+PD9LS0rBv3z5tW5s2beDq6opVq1ZBCAF7e3tMmTIFU6dOBQAkJyfD1tYW69atw4cffpjrM9PT05Ge/vcPT3JyMmrUqJFv5v/85z/YsWOH9rmdnR1evHiRZ1+zag1hO2CO9vn9VaMgXj7Ls69plTqwGxyqff7Hmk+R/exRnn1NKlRH1eGLtM8frPfHqyf38+xrVM4G1UatxOU5ngCAjh07IjY2Ns++FStWRBnfVdrnidtnI+OPq3n2hYkZHMb/V/v0YXgIXt7N+3MBwGHSNu3Xj/YtxIubv+bbt9qnG7SF5K//rcDzq8fz7Xvr1i3Y2NgAAKZMmYKwsLB8+1YduRwmVq93OJ6e3Ihn5/fm29d26Dcws3n9c5ActQ0pv+7It2+VD+fB3K4uACDl3E9IPrUp37779u3De++9BwBYvXo1AgIC8u1bqXcgLGs3AwCk/nYUTw5/l2/fit0no0w9dwBA2o0oPD6wON++Ff7zCco27gQAeH4nBn/9FJp/304jUbZpVwDAy/uX8XDH3Hz7zp07FxMnToTz7P8hPeEmkrbMzLevVev+sHYfCADIeKRB4n+n5tu3XHMvlH9vGADgVUoSHqwdn2/fMk0+QMXOowEAWS+S8ef3Y/LtO2jQIKxa9frnPS0tDfb29vn2tajbGjY9p2ifxy8ZmG9fdS03VPae8Xff5UOBV3kXs6L8jcj5XXZ2dkZ8fHyefevXry8ZfWnevDlu3ryJp0+fwtraOt/sAAAho/T0dGFsbCx2794taff19RW9evXK8z0ODg5i8eLFkrbg4GDRpEkTIYQQt27dEgBEbGyspM/7778vJkyYkOdnzp49W+D1/jYffPDBR6l+xMfHF/q3WdahpEePHiErKwu2traSdltbW1y7di3P9yQkJOTZPyEhQft6Tlt+ff5pxowZ8Pf31z7Pzs7G48ePUalSJe0QhRxSUlLg4OCA+Ph4WFlZyZajqAw9P2D422Do+QFuQ3ERQuDZs2cFHhXlkP0cgxKYm5trxxlzlC9fXp4webCysjLYXwjA8PMDhr8Nhp4f4DYUh0KHkP4/Wa9KsrGxgbGxMRITEyXtiYmJsLOzy/M9dnZ2BfbP+a8+n0lERH+TtTCYmZmhefPmiIiI0LZlZ2cjIiIC7u7ueb7H3d1d0h8ADh8+rO1fu3Zt2NnZSfqkpKTg119/zfcziYjoDYWehXjLtmzZIszNzcW6devElStXxMcffyzKly8vEhIShBBCDBs2TAQGBmr7R0ZGChMTE/HNN9+Iq1evitmzZwtTU1Nx6dIlbZ/Q0FBRvnx58dNPP4m4uDjRu3dvUbt2bfHixYsS375/4+XLl2L27Nni5cuXckcpEkPPL4Thb4Oh5xeC2yAH2QuDEEJ8++23okaNGsLMzEy0atVKREdHa1/r0KGDGD58uKT/tm3bRL169YSZmZlo3Lix2L9/v+T17OxsERQUJGxtbYW5ubno0qWLuH79eklsChGRwZP9PgYiIlIWRUyJQUREysHCQEREEiwMREQkwcJAREQSLAxERCTBKTEU4urVq9iyZQtOnjyJe/fu4fnz56hcuTLc3Nzg6emJfv365Zq2gyhHXFyczn2bNGnyFpPQ06dPsXv37nx/l9u2bSt3xELxclWZxcTEYNq0aTh16hTatWuHVq1awd7eHhYWFnj8+DEuX76MkydPIiUlBdOmTcOkSZMUXSCysrKwbt06REREICkpKde6D0eOHJEpmX6ePn2KM2fO5LkNvr6+MqXKn5GREVQqFYQQhU78mFXAuhJUdH/++SeCg4OxadMm2Nvb5/m7fP78edSsWROzZ8+Gj4+P3JHzxSMGmfXr1w8BAQHYsWNHgRP3RUVFYenSpVi4cCFmzsx/jnu5TZw4EevWrUOPHj3g7Ows6+y0RbV3714MGTIEqampsLKykmyDSqVSZGG4c+eO9uvY2FhMnToVAQEB2mlgoqKisHDhQixYsECuiIVyc3PT+eclJibmLafRn5ubG4YPH47z58+jUaNGefZ58eIFwsPDsWTJEsTHx2vXjFEaHjHILDMzE6ampm+tf0mzsbHBhg0b0L17d7mjFFm9evXQvXt3zJs3T7IClqFo1aoVvvjii1z/BgcOHEBQUBDOnz8vU7KCzZnz92I1L1++xMqVK9GoUSNtcYuOjsZvv/2GTz/9FCEhIXLFzNdff/2FSpUqvbX+JYmFgYqVvb09jh07hnr16skdpcjKlCmDS5cuoU6dOnJHKRILCwvExMSgYcOGkvarV6+iWbNm+a42qCSjR49G1apV8eWXX0raZ8+ejfj4eKxdu1amZO8GFgaZLVu2TOe+EyZMeItJisfChQtx+/ZtLF++3CCHkQCgb9+++PDDDzFwYP5LNypZs2bN4OzsjLCwMO36yxkZGRg9ejQuX76syGGYf7K2tsa5c+fg5OQkaf/999/RokULJCcny5Qsf3v27NG5b69evd5ikn+P5xhktnjxYp36qVQqgygMp06dwtGjR/Hzzz+jcePGuYa9du3aJVMy3fXo0QMBAQG4cuUKXFxccm2D0n+pV61aBS8vL1SvXl17BVJcXBxUKhX27s1/jWslsbCwQGRkZK7CEBkZCbVaLVOqgnl7e+vUT6VSKf4CAB4xULEaMWJEga//3//9XwklKTojo/xv7zGEX2rg9cL2mzZt0i6R27BhQwwePBhlypSROZluQkNDMWfOHIwZMwatWrUCAPz6669Yu3YtgoKCEBgYKHPC0o2FgYgUadu2bVi6dCmuXr0K4HVxmzhxosEO8RkSFgaFuX//Pvbs2QONRoOMjAzJa4sWLZIplf4ePnyI69evAwDq16+PypUry5zo3bJx40Z8//33uH37NqKiolCzZk0sXrwYderUQe/eveWO905IS0vD8ePH8/xdVvywcImvAEH5+uWXX4SlpaVwdnYWJiYmwtXVVZQvX15YW1uLTp06yR1PJ6mpqWLEiBHC2NhYqFQqoVKphImJiRg5cqRIS0uTO57Ojh07Jnr27CkcHR2Fo6Oj8PLyEidOnJA7lk5WrlwpbGxsxFdffSXUarW4deuWEEKI//u//xMdO3aUOZ3unjx5In744QcxY8YM8ddffwkhhDh//ry4f/++zMkKFxMTI+zs7ISVlZUwNjYWlStXFiqVSpQpU0bUrl1b7niFYmFQkJYtW4rg4GAhhBBly5YVt27dEs+ePRO9evUSK1eulDmdbj7++GNRp04dceDAAZGcnCySk5PF/v37haOjoxg3bpzc8XSyceNGYWJiIgYOHCiWLl0qli5dKgYOHChMTU3Fpk2b5I5XqIYNG4rdu3cLIf7+ORJCiEuXLolKlSrJmEx3Fy9eFJUrVxZ169YVJiYm2m34/PPPxbBhw2ROV7gOHTqIMWPGiKysLO2/gUajEe+//77YuXOn3PEKxcKgIGXLlhU3b94UQghRvnx5cfnyZSGEEBcuXBA1a9aUMZnuKlWqJI4ePZqr/ciRI8LGxqbkAxVBgwYNxKJFi3K1L1y4UDRo0ECGRPpRq9Xi7t27QghpYbhx44ZQq9VyRtNZly5dREBAgBBCug2RkZEG8btgbW0trl27pv36ypUrQgghoqOjRf369eWMphPOrqogZcqU0Y5FVq1aFbdu3dK+9ujRI7li6eX58+ewtbXN1V6lShU8f/5chkT6u337Nry8vHK19+rVSzL1hFLVrl0bFy5cyNV+8ODBXDe9KdXZs2cxduzYXO3VqlVDQkKCDIn0Y2pqqr26rUqVKtBoNABe358RHx8vZzSd8D4GBWnTpg1OnTqFhg0bonv37pgyZQouXbqEXbt2oU2bNnLH04m7uztmz56NDRs2aK83f/HiBebMmaOd2kDpHBwcEBERgbp160raf/nlFzg4OMiUSnf+/v747LPP8PLlSwghcObMGfz4448ICQlBWFiY3PF0Ym5ujpSUlFztN27cMIgLGdzc3HD27Fk4OTmhQ4cOCA4OxqNHj7Bx40Y4OzvLHa9wch+y0N9u3bolLl68KIR4fRJ37NixwsXFRfTt21c7NKB0ly5dEvb29qJSpUqic+fOonPnzqJSpUqiWrVq2qExpVu5cqUwMzMT48aNExs2bBAbNmwQY8eOFebm5mLVqlVyx9PJf//7X1G3bl3tBQDVqlUTYWFhcsfS2ahRo4S3t7fIyMgQZcuWFbdv3xb37t0Tbm5uYuLEiXLHK9TZs2fFkSNHhBBCJCYmCk9PT1GuXDnRrFkzceHCBZnTFY6Xq1Kxe/78ea6bq4YMGQILCwuZk+lu9+7dWLhwoeQa+oCAAIO71PP58+dITU1FlSpV5I6il+TkZPTv3x/nzp3Ds2fPYG9vj4SEBLi7u+PAgQMGc6OeoWJhUJinT59ix44duHXrFgICAlCxYkXExMTA1tYW1apVkzseGYAXL15ACKGdGfbevXvYvXs3GjVqhA8++EDmdPqJjIzExYsXkZqaimbNmsHDw0PuSO8EFgYFiYuLg4eHB6ytrXH37l1cv34dderUwaxZs6DRaLBhwwa5I+Zpz5496NatG0xNTQudSEzp8wyVBh988AH69u2LcePG4enTp6hfvz7MzMzw6NEjLFq0CJ988oncEYvk6dOnBa5ZoiSJiYmYOnWqdsGqf/6ZVfq0KiwMCuLh4YFmzZphwYIFKFeuHC5evIg6derg9OnTGDx4MO7evSt3xDwZGRkhISEBVapUMdh5hipWrIgbN27AxsYGFSpUKHBm2MePH5dgMv3Z2Njg+PHjaNy4McLCwvDtt98iNjYWO3fuRHBwsHZ4TMnmz5+PWrVqaVc5GzhwIHbu3Ak7OzscOHAATZs2lTlhwbp16waNRoPx48ejatWquX6elD4kyauSFOTs2bP4/vvvc7Ur/RK9N5e+/OcymIZi8eLFKFeunPZrQ50yHHh9XiFnWw4dOoS+ffvCyMgIbdq0wb1792ROp5tVq1Zh06ZNAIDDhw/j8OHD+Pnnn7Ft2zYEBATg0KFDMics2KlTp3Dy5Em4urrKHaVIWBgUxNAv0QOADRs2wMfHJ9e61BkZGdiyZYsil8UEgOHDh2u//uijj+QLUgzq1q2L8PBw9OnTB//73/8wefJkAEBSUhKsrKxkTqebhIQE7aXB+/btw8CBA/HBBx+gVq1aaN26tczpCufg4JBr+MiQ8AY3BenVqxfmzp2LzMxMAK+HXjQaDaZPn45+/frJnE43I0aMyHMRlWfPnhU6JbdSGBsbIykpKVf7X3/9BWNjYxkS6Sc4OBhTp07V/hHNuX/k0KFDcHNzkzmdbipUqKC9EezgwYPak85CCMUOR75pyZIlCAwMVOzwb2F4xKAgCxcuRP/+/VGlShW8ePECHTp00F6i9/XXX8sdTydCiDyHYe7fvw9ra2sZEukvvz299PR07YpoSta/f3+0b98eDx48kIzFd+nSBX369JExme769u2LwYMHw8nJCX/99Re6desGAIiNjc1146ES+fj44Pnz53B0dISlpWWuxZ6Ufp6KhUFBrK2tcfjwYZw6dQpxcXEGdYmem5sbVCoVVCoVunTpAhOTv3+0srKycOfOHXTt2lXGhIXLWWZVpVIhLCwMZcuW1b6WlZWFEydOoEGDBnLF00lmZiYsLCxw4cKFXEcHOQveGILFixejdu3a0Gg0WLBggfbf4sGDB/j0009lTle4JUuWyB3hX+FVSVQs5syZo/3vlClTJH9UzczMUKtWLfTr10/Re9y1a9cG8Pq6/+rVq0uGjXK2Ye7cuYof465Tpw52796t+Ct38pOZmYmxY8ciKChI+29CJYuFQWY5e6m6UPziHgDWr1+PDz/8MNfJZ0PSqVMn7Nq1CxUqVJA7SpGsWbMGu3btwsaNG1GxYkW54xSJtbU1Lly4YFCFISUlRXtyP6+LSN6k9IsAWBhk9s8f/IcPH+L58+faG3mePn0KS0tLVKlSBbdv35YhoX7Onj2L7OzsXHvVv/76K4yNjdGiRQuZkr073NzccPPmTWRmZqJmzZq5po+IiYmRKZnuhg8fDldXV+0VVYbA2NgYDx480N7Pk9e5tpxzcEo/gc5zDDJ7cxrnzZs3Y+XKlVizZg3q168PALh+/TrGjBmT5xTESvTZZ59h2rRpuQrDH3/8gfnz5+PXX3+VKZnu+vXrh1atWmH69OmS9gULFuDs2bPYvn27TMl04+3tLXeEf83JyQlz585FZGQkmjdvnqu4KfHo+ciRI9ojtKNHj8qc5t/hEYOCODo6YseOHblOGp4/fx79+/c3iLUAypYti7i4ONSpU0fSfufOHTRp0gTPnj2TKZnuKleujCNHjsDFxUXSfunSJXh4eCAxMVGmZO+OgoaQVCqVQRw9GzIeMSjIgwcP8OrVq1ztWVlZBvPHyNzcHImJibkKw4MHDyRXKilZampqnifJTU1NCx07VgpDn4zREHaCChIXF5dnu0qlglqtRo0aNZR9Hq4Ep/imQvTs2VO4ubmJ8+fPa9vOnTsnmjVrJry8vGRMprsPP/xQdOjQQTx9+lTb9uTJE9GhQwcxYMAAGZPprmXLlmLOnDm52mfPni2aNWsmQyL9GPp6yW9KT08X165dE5mZmXJH0YtKpRJGRkb5PszNzYWvr6948eKF3FHzxMKgIElJSaJbt25CpVIJMzMzYWZmJoyMjES3bt1EYmKi3PF0cv/+fVGnTh1hbW0tOnbsKDp27CjKly8v6tevLzQajdzxdLJnzx5hYmIifH19xbp168S6devEsGHDhImJidi9e7fc8Qpl6OslCyFEWlqaGDlypDA2NhbGxsbabRg/frwICQmROV3hwsPDRf369UVYWJiIi4sTcXFxIiwsTDRs2FBs2bJF/Pe//xXVq1cXU6ZMkTtqnlgYFOj69evip59+Ej/99JO4fv263HH0lpqaKr7//nvx6aefiilTpoj169eLjIwMuWPpZd++faJt27bC0tJSVKpUSXTq1EkcO3ZM7lg6sbKyEjdv3hRCSAvD3bt3hbm5uZzRdDZhwgTRvHlzcfLkSVGmTBntNoSHhwtXV1eZ0xWuZcuW4uDBg7naDx48KFq2bCmEEGL37t2iTp06JR1NJ4Yx6PuOqVevHurVqyd3jCIrU6YMPv74Y7lj/Cs9evRAjx495I5RJKVhMsbw8HBs3boVbdq0kVz22bhxY9y6dUvGZLq5dOkSatasmau9Zs2auHTpEgDA1dUVDx48KOloOmFhkJm/vz++/PJLlClTBv7+/gX2XbRoUQml0g8X6lGWnMkYt23bBsAwJ2N8+PBhnsuRpqWlGcSU6A0aNEBoaChWr16tvZAhMzMToaGh2mlV/vjjD9ja2soZM18sDDKLjY3VzqYaGxubbz8l/zJ4e3trF+op6Bp6Jd/YU5oW6ikNkzG2aNEC+/fvh5+fH4C/f/7DwsK0s8Uq2YoVK9CrVy9Ur14dTZo0AfD6KCIrKwv79u0DANy+fVux8z7xPgYiSKfyWL9+fYF931y7QckMcTLGHKdOnUK3bt0wdOhQrFu3DmPHjsWVK1dw+vRpHD9+HM2bN5c7YqGePXuGTZs24caNGwCA+vXrY/DgwdpFlJSMhYGIFOnWrVsIDQ3FxYsXtcVt+vTpuW48VJrMzEw0aNAA+/btQ8OGDeWOUyQsDDIbN24cZs2aherVqxfad+vWrXj16hWGDBlSAsl0VxomAtTnxjWlT4AGABEREVi8eLF2feeGDRti0qRJBnXUYMiqVauGX375hYWBiiYoKAjLli1Du3bt4OXlhRYtWsDe3h5qtRpPnjzBlStXcOrUKWzZsgX29vZYvXq1dsxSKUrDRID5TXqWF6WeJ8mxcuVKTJw4Ef3799eOx0dHR2PHjh1YvHgxPvvsM5kT6iYrKwu7d+/WFrdGjRqhd+/eBnEH/bx583Djxg2EhYUZRN5/YmFQgMTERISFhWHLli24cuWK5LVy5crBw8MDo0ePVvxCN0DhEwEq7Wgnx/Hjx7Vf3717F4GBgfjoo4+0f1ijoqKwfv16hISEKP4cQ/Xq1REYGIjx48dL2lesWIF58+bhjz/+kCmZ7n777Tf06tULCQkJ2p+jnMtt9+7dC2dnZ5kTFqxPnz6IiIhA2bJl4eLikmsSwF27dsmUTDcsDArz5MkTaDQavHjxAjY2NnB0dFT0FUn/VBomAuzSpQtGjx6NQYMGSdo3b96M1atX49ixY/IE01HZsmVx4cKFXEtg/v7773Bzc0NqaqpMyXTn7u6OypUrY/369dp1MZ48eYKPPvoIDx8+xOnTp2VOWLDC1jf/v//7vxJKUjQsDFSsLC0tcfz4cbRs2VLSfubMGXTs2BHPnz+XKZnuLC0tcfHiRTg5OUnab9y4AVdXV8Vvw+DBg+Hm5oaAgABJ+zfffINz585hy5YtMiXTnYWFBc6dO4fGjRtL2i9fvoyWLVvixYsXMiV7Nxje4BcpWpcuXTB27FiEhYWhWbNmAF4fLXzyyScGc+LTwcEBP/zwAxYsWCBpDwsLg4ODg0ypdNeoUSN8/fXXOHbsmOQcQ2RkJKZMmSK5WECpFwPUq1cPiYmJuQpDUlJSriMhKn48YqBi9fDhQwwfPhwHDx6EqakpAODVq1fw9PTEunXr8rybVWkOHDiAfv36oW7dutoFh86cOYPff/8dO3fuRPfu3WVOWDBdl8NU8roGBw4cwLRp0/DFF1+gTZs2AF4Xt7lz5yI0NBTt27fX9lXqVWI7duzAtm3boNFokJGRIXlN6avosTDQW3Hjxg1cu3YNwOvpAQxt7qf4+Hh899132m1o2LAhxo0bZxBHDKWBkZGR9uucc2w5f6refK7Uu+mXLVuGzz//HB999BFWr16NESNG4NatWzh79iw+++wzxd+BzsJAb0VGRgbu3LkDR0dHg7xcrzTJysrSTuqWcyJX6d68SqwwHTp0eItJiqZBgwaYPXs2Bg0ahHLlyuHixYuoU6cOgoOD8fjxYyxfvlzuiAUryalcqfQz9Hn0c5w4cUIMGTJEuLu7i/v37wshhNiwYYM4efKkzMkKN3HiRBEWFiaEEOLVq1eibdu2QqVSiTJlyoijR4/KG+4dYWFhIe7evSuEEKJy5criwoULQgghbty4ISpWrChnNJ0YFV466G1yc3NDs2bNdHoYghkzZuDixYs4duwY1Gq1tt3DwwNbt26VMZnudu7cCU9PT1hYWCAmJgbp6ekAgOTkZMybN0/mdIXbsWMHmjZtCgDYu3cv7t69i2vXrmHy5Mn4/PPPZU6nm4MHD+LUqVPa5ytWrICrqysGDx6MJ0+eyJhMN3Z2dtrJFmvUqIHo6GgAr5csFQYwSMPCIDNvb2/07t0bvXv3hqenJ27dugVzc3N07NgRHTt2hFqtxq1bt+Dp6Sl3VJ2Eh4dj+fLlaN++vUHOow8AX331FVatWoUffvhBewIdANq1a6f4k4YA8OjRI9jZ2QF4fRJ3wIABqFevHkaOHKldC0DpAgICtNOUXLp0Cf7+/ujevTvu3LlT6PT0StC5c2ftFPQjRozA5MmT8Z///Ac+Pj7o06ePzOkKx8Ffmc2ePVv79ejRozFhwgR8+eWXufrEx8eXdLQiMfR59IHXd2q///77udqtra3x9OnTkg+kJ1tbW1y5cgVVq1bFwYMH8d133wEAnj9/DmNjY5nT6ebOnTto1KgRgNdHcF5eXpg3bx5iYmIUf1UYAKxevRrZ2dkAgM8++wyVKlXC6dOn0atXL4wdO1bmdIXjEYOCbN++Hb6+vrnahw4dip07d8qQSH858+jnMLR59IHXwwA3b97M1X7q1CnUqVNHhkT6GTFiBAYOHAhnZ2eoVCrt/SO//vqrdpEYpTMzM9PeSPjLL7/ggw8+APB63Qx9JjyUi5GRkeSiiw8//BDLli2Dn5+fduEeJeMRg4JYWFggMjIy1x23kZGRkvF6JZs3bx66deuGK1eu4NWrV1i6dKlkHn1DMGbMGEycOBFr166FSqXCn3/+iaioKEydOhVBQUFyxyvUF198AWdnZ8THx2PAgAEwNzcHABgbGyMwMFDmdLpp3749/P390a5dO5w5c0Z7furGjRs6zUSsBC9fvkRcXBySkpK0Rw85FL+Sodxnv+lvISEhQq1WCz8/P7Fx40axceNGMX78eGFpaWlQV/TcunVLjB49WrRs2VI0bNhQDBkyRMTFxckdS2fZ2dniq6++EmXKlBEqlUqoVCqhVqvFrFmz5I72zrh3757o0aOHaNKkifYKKyGEmDRpkvDz85MxmW5+/vlnUblyZe3Pz5sPIyMjueMVivcxKMy2bduwdOlSyTz6EydOxMCBA2VOVrjMzEyMHTsWQUFBOt99qzRZWVmIjIxEkyZNYGlpiZs3byI1NRWNGjVC2bJl5Y6Xr2XLluHjjz+GWq0udH0MpU6DUZo4OTnhgw8+QHBwsGLXdS4ICwMVK2tra1y4cMFgCwMAqNVqXL161aC2oXbt2jh37hwqVapUYG4lT4ORkpKind6isPMISp0GI4eVlRViY2Ph6Ogod5Qi4TkGhXn69Cl27Ni |
Collaborator
There was a problem hiding this comment.
Suggested change
| "image/png": "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 | |
| "# and forecast type (the default is \"ensemble\")\n", |
| "outputs": [ | ||
| { | ||
| "data": { | ||
| "image/png": "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 |
Collaborator
There was a problem hiding this comment.
Suggested change
| "image/png": "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 | |
| "The combination of these two facts concentrates more PIT values in the first bin. If we instead call `pit.hist_values(bins, right=False)`, we'll transfer the problem from the fisrt bin to the last bin. Ultimately, the problem stems from the fact that the empirical CDF is not perfectly calibrated. For this reason, users may prefer to use `scores.probability.rank_histogram` to generate similar plots where the ensemble is not being interpreted as an empirical CDF. Strengths and weaknesses of the two approaches are further discuused in the [rank histogram tutorial](./Rank_Histogram.ipynb).\n", |
| y_pos = param_plotting_points["y_plotting_position"] | ||
| # gradient of chord AB where A(x_pos[i-1], y_pos[i-1]), B(x_pos[i], y_pos[i]) | ||
| gradient = y_pos.diff("plotting_point") / x_pos.diff("plotting_point") | ||
| # solution if there is a desired point of intersection, optained by solving |
Collaborator
There was a problem hiding this comment.
Suggested change
| # solution if there is a desired point of intersection, optained by solving | |
| # solution if there is a desired point of intersection, obtained by solving |
| Args: | ||
| fcst: xarray object of ensemble forecasts, including dimension `ens_member_dim` | ||
| obs: xarray object of observations | ||
| ens_member_dim: name of the ensemble memeber dimension in `fcst` |
Collaborator
There was a problem hiding this comment.
Suggested change
| ens_member_dim: name of the ensemble memeber dimension in `fcst` | |
| ens_member_dim: name of the ensemble member dimension in `fcst` |
| ensures that the expected rank histogram of a probabilistically calibrated ensemble is flat. | ||
|
|
||
| In the case when there is a ``NaN`` in one of the ensemble members, the entire ensemble is | ||
| treates as ``NaN`` for this particular forecast case. |
Collaborator
There was a problem hiding this comment.
Suggested change
| treates as ``NaN`` for this particular forecast case. | |
| treats as ``NaN`` for this particular forecast case. |
| """ | ||
| # x values are sorted, with some duplicates | ||
| x_vals = [0, 0, 1, 1, 2, 2] | ||
| # y values are assending, possibly with jumps at common x values |
Collaborator
There was a problem hiding this comment.
Suggested change
| # y values are assending, possibly with jumps at common x values | |
| # y values are ascending, possibly with jumps at common x values |
| def test__pit_values_final_processing(pit_values, weights, expected_left, expected_right): | ||
| """ | ||
| Tests that `_pit_values_final_processing` returns as expected. | ||
| This test specifically tests that the weighted means are resaled correctly for left |
Collaborator
There was a problem hiding this comment.
Suggested change
| This test specifically tests that the weighted means are resaled correctly for left | |
| This test specifically tests that the weighted means are rescaled correctly for left |
Collaborator
Author
|
Thanks @reza-armuei , really appreciate the review. @durgals , I have two options going forward:
Do you have a preference? I know that some things raised by the other reviewers will need to be addressed irrespective of your review (e.g. python naming conventions of classes) |
Contributor
@rob-taggart , sorry for the delay in reviewing- I've been tied up with Modsim and then a project deliverable that's due at the end of this month. I don't want to hold you up, so please feel free to go ahead and respond to the review without waiting for my input. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
5 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Please work through the following checklists. Delete anything that isn't relevant.
Development for new xarray-based metrics
reduce_dims,preserve_dims, andweightsargs.xr.DataArraysandxr.Datasetsif possibleDocstrings
Testing of new xarray-based metrics
xr.DataArraysandxr.DatasetsTutorial notebook
Documentation