Merge pull request #24 from praksharma/development

Major improvements in geometry and sampling

DeepINN/__init__.py

@@ -3,10 +3,9 @@
 
 from . import backend
 from . import config
-# from . import callbacks
-# from . import data
-from . import geometry
-# from . import gradients as grad
-# from . import BC
-# from . import nn
-#from . import utils
+
+from .geometry import spaces
+from .geometry import domains
+from .geometry import samplers
+
+

DeepINN/geometry/__init__.py

@@ -1,5 +1,5 @@
 # Make the subdirectory callable
 from . import domains
-from . import conditions
+#from . import conditions
 from . import samplers
 from . import spaces

DeepINN/geometry/domains/domainoperations/sampler_helper.py

@@ -4,7 +4,8 @@
 import torch
 import warnings
 
-from torchphysics.problem.spaces.points import Points
+from ....geometry.spaces.points import Points # some relative path fixing
+#from torchphysics.problem.spaces.points import Points
 
 
 def _inside_random_with_n(main_domain, domain_a, domain_b, n, params, invert, device):

DeepINN/utils/plotting/scatter_points.py

@@ -5,7 +5,7 @@
 import matplotlib.pyplot as plt
 
 
-def scatter(subspace, *samplers):
+def scatter(subspace, *samplers, dpi=False, save=False):
     """Shows (one batch) of used points in the training. If the sampler is
     static, the shown points will be the points for the training. If not
     the points may vary, depending of the sampler. 
@@ -19,22 +19,32 @@ def scatter(subspace, *samplers):
         The diffrent samplers for which the points should be plotted.
         The plot for each sampler will be created in the order there were
         passed in.
+    dpi : dpi of the saved figure.
+    save : A flag to save figure.
+
+    In future, I might use a dictionary to add more functions.
 
     Returns
     -------
     fig : matplotlib.pyplot.figure
-        The figure handle of the plot.
+        The figure handle of the plot. The Figure is created on the STDOUT, so actual return isn't required.
     """
     assert subspace.dim <= 3, "Can only scatter points in dimensions <= 3."
 
-    fig, ax, scatter_fn = _choose_scatter_function(subspace.dim)
-    ax.grid()
+    #scatter_fn = _choose_scatter_function(subspace.dim)
+
     for sampler in samplers:
         points = sampler.sample_points()[:, list(subspace.keys())]
         numpy_points = points.as_tensor.detach().cpu().numpy()
         labels = _create_labels(subspace)
-        scatter_fn(ax, numpy_points, labels)
-    return fig
+        #scatter_fn(numpy_points, labels)
+
+    if subspace.dim == 1:
+        _scatter_1D(numpy_points, labels, dpi, save)
+    elif subspace.dim == 2:
+        _scatter_2D(numpy_points, labels, dpi, save)
+    else:
+        _scatter_3D(numpy_points, labels, dpi, save)
 
 def _create_labels(subspace):
     labels = []
@@ -47,31 +57,43 @@ def _create_labels(subspace):
     return labels
 
 def _choose_scatter_function(space_dim):
-    fig = plt.figure()
+
     if space_dim == 1:
-        ax = fig.add_subplot()
-        return fig, ax, _scatter_1D
+        return _scatter_1D
     elif space_dim == 2:
-        ax = fig.add_subplot()
-        return fig, ax, _scatter_2D    
+        return _scatter_2D    
     else:
-        ax = fig.add_subplot(projection='3d')
-        return fig, ax, _scatter_3D  
+        return _scatter_3D  
 
 
-def _scatter_1D(ax, points, labels):
+def _scatter_1D(points, labels, dpi, save):
+    fig = plt.figure()
+    ax = fig.add_subplot()
+    ax.grid()
     ax.scatter(points, np.zeros_like(points))
     ax.set_xlabel(labels[0])
+    if save:
+        plt.savefig('geom.jpg', dpi = dpi,bbox_inches='tight',transparent=True)
 
 
-def _scatter_2D(ax, points, labels):
+def _scatter_2D(points, labels, dpi, save):
+    fig = plt.figure()
+    ax = fig.add_subplot()
+    ax.grid()
     ax.scatter(points[:, 0], points[:, 1])
     ax.set_xlabel(labels[0])
     ax.set_ylabel(labels[1])
+    if save:
+        plt.savefig('geom.jpg', dpi = dpi,bbox_inches='tight',transparent=True)
 
 
-def _scatter_3D(ax, points, labels):
+def _scatter_3D(points, labels, dpi, save):
+    fig = plt.figure()
+    ax = fig.add_subplot(projection='3d')
+    ax.grid()
     ax.scatter(points[:, 0], points[:, 1], points[:, 2])
     ax.set_xlabel(labels[0])
     ax.set_ylabel(labels[1])
-    ax.set_zlabel(labels[2])
+    ax.set_zlabel(labels[2])
+    if save:
+        plt.savefig('geom.jpg', dpi = dpi,bbox_inches='tight',transparent=True)

Tutorials/Basic_domains.ipynb

@@ -0,0 +1,65 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# This is only valid when the package is not installed\n",
+    "import sys\n",
+    "sys.path.append('/home/hell/Desktop/repos/DeepINN/')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "Using default backend: PyTorch\n",
+      "Using Pytorch:  2.0.1+cu117\n"
+     ]
+    }
+   ],
+   "source": [
+    "import DeepINN as dp"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "T = dp.spaces.R1('t') # we need a one dimensional space\n",
+    "I = dp.domains.Interval(T,0,5 )"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": ".venv",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.8.10"
+  },
+  "orig_nbformat": 4
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}