improve

32.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4aee59c935e5203ff354f7a34a9e1ea99a4ee7f43a2bcd028d84335bba312f2f
+size 6713611

64-guardado.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d6b832725a9ac119612e7e578c9c32d93c9c90edad65ec1c7a88f0da44f90207
+size 10202443

64.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d6b832725a9ac119612e7e578c9c32d93c9c90edad65ec1c7a88f0da44f90207
-size 10202443
+oid sha256:37f658c9e5db8fc3340a133ef1be94c3b1127d2eb74a6d7f94ab812e989c2dff
+size 6713611

app.ipynb

@@ -2,7 +2,7 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 11,
+   "execution_count": 6,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -20,14 +20,13 @@
     "import os\n",
     "from resamplear import *\n",
     "import open3d as o3d\n",
-    "from meshSubplot import sTree, plotTree\n",
-    "import meshplot as mp\n",
-    "from vedo import *"
+    "from vedo import *\n",
+    "from meshSubplot import plotTree"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 12,
+   "execution_count": 7,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -139,7 +138,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 13,
+   "execution_count": 8,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -204,23 +203,20 @@
     "\n",
     "def prepararAristas( grafo ):\n",
     "    for arista in grafo.edges():\n",
-    "        nx.set_edge_attributes( grafo, {arista : {'procesada':False}})\n"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 14,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "t_list = os.listdir(\"./nuevostrees/\" )\n",
-    "dataset = tDataset(t_list, \"./nuevostrees/\")\n",
-    "data_loader = DataLoader(dataset, batch_size = batch_size, shuffle=True, collate_fn=my_collate)"
+    "        nx.set_edge_attributes( grafo, {arista : {'procesada':False}})\n",
+    "\n",
+    "def traversefeatures(root, features):\n",
+    "       \n",
+    "    if root is not None:\n",
+    "        traversefeatures(root.left, features)\n",
+    "        features.append(root.radius.tolist()[0][3])\n",
+    "        traversefeatures(root.right, features)\n",
+    "        return features\n"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 15,
+   "execution_count": 9,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -232,371 +228,194 @@
     "        \n",
     "        z = torch.randn(1, latent_size)\n",
     "        generated_images = decode_testing(z, 100, Grassdecoder, mult, 1)\n",
-    "        print(\"generated images\", generated_images)\n",
     "        count = []\n",
     "        numerar_nodos(generated_images, count)\n",
+    "        plotTree(generated_images, True)\n",
     "        #tr(generated_images)\n",
     "        generated_images.toGraph( G, 0, True, flag = 0)\n",
-    "\n",
+    "        r_list = []\n",
+    "        r_list = traversefeatures(generated_images, r_list)\n",
+    "        max_radius = max(r_list)\n",
+    "        min_radius = min(r_list)\n",
+    "        \n",
+    "        '''\n",
+    "        #generated_images = iter(data_loader).next()[0]\n",
+    "        generated_images = next(iter(data_loader))[0]\n",
+    "        filename = generated_images[1]\n",
+    "        print(\"filename\", filename)\n",
+    "        bifurcation_nodes = []\n",
+    "        leaf_nodes = []\n",
+    "       \n",
+    "        generated_images[0].toGraph( G, 0, False, 0)\n",
+    "       \n",
+    "        \n",
+    "        generated_images = generated_images[0]\n",
+    "        original_mesh = load(\"mallas/\"+filename.split(\"_\")[0]+\".obj\")\n",
+    "        write(original_mesh, \"original.obj\")\n",
+    "        #generated_images.toGraph( graph, 0, False, flag = 0)\n",
+    "        #tr2(generated_images)\n",
+    "        '''\n",
     "        \n",
     "        #nx.write_gpickle(graph, \"grafo0.gpickle\" )\n",
     "        for arista in G.edges():\n",
     "            nx.set_edge_attributes( G, {arista : {'procesada':False}})\n",
-    "        #graph_resampled = resamplear(G, puntosPorUnidad=1)\n",
+    "        #graph_resampled = resamplear(G, puntosPorUnidad=5)\n",
     "        graph_resampled = G\n",
     "        prepararAristas(graph_resampled)\n",
     "        graphOfCenters = GrafoCentros(graph_resampled)\n",
     "        \n",
-    "\n",
-    "        try:\n",
-    "            graphOfCenters.tile()\n",
-    "            mesh = tm.Trimesh( graphOfCenters.getVertices(), graphOfCenters.getCaras() )\n",
-    "            mesh_o3d = mesh.as_open3d\n",
-    "            m2 = mesh_o3d \n",
-    "            o3d.io.write_triangle_mesh(\"sinsub.obj\", m2)\n",
-    "            mesh_o3d = mesh_o3d.subdivide_loop(3)\n",
+    "        if abs(max_radius/min_radius)<10:\n",
+    "            try:\n",
+    "                graphOfCenters.tile()\n",
+    "                mesh = tm.Trimesh( graphOfCenters.getVertices(), graphOfCenters.getCaras() )\n",
+    "                mesh_o3d = mesh.as_open3d\n",
+    "                m2 = mesh_o3d \n",
+    "                o3d.io.write_triangle_mesh(\"sinsub.obj\", m2)\n",
+    "                mesh_o3d = mesh_o3d.subdivide_loop(3)\n",
     "            \n",
-    "            v_matrix = np.asarray(mesh_o3d.vertices )\n",
-    "            f_matrix = np.asarray(mesh_o3d.triangles )\n",
-    "            v_matrix_2 = np.asarray(m2.vertices )\n",
-    "            f_matrix_2 = np.asarray(m2.triangles )\n",
-    "            success = True\n",
+    "                v_matrix = np.asarray(mesh_o3d.vertices )\n",
+    "                f_matrix = np.asarray(mesh_o3d.triangles )\n",
+    "                v_matrix_2 = np.asarray(m2.vertices )\n",
+    "                f_matrix_2 = np.asarray(m2.triangles )\n",
+    "                success = True\n",
     "            \n",
-    "        except Exception as e:\n",
-    "            print(\"No se pudo generar\")\n",
-    "            print(str(e))\n",
+    "            except Exception as e:\n",
+    "                print(\"No se pudo generar\")\n",
+    "                pass\n",
+    "        else:\n",
+    "            print(\"ratio\", max_radius/min_radius)\n",
     "            pass\n",
-    "        \n",
-    "    #print(\"x\", v_matrix[:,0])\n",
-    "    #print(\"y\", v_matrix[:,1])\n",
-    "    #print(\"z\", v_matrix[:,2])\n",
+    "    '''\n",
+    "    M = max((max(v_matrix[:,0]) - min(v_matrix[:,0])), (max(v_matrix[:,1]) - min(v_matrix[:,1])), (max(v_matrix[:,2]) - min(v_matrix[:,2])))\n",
+    "    x = (v_matrix[:,0] - min(v_matrix[:,0]))/M\n",
+    "    y = (v_matrix[:,1] - min(v_matrix[:,1]))/ M\n",
+    "    z = (v_matrix[:,2] - min(v_matrix[:,2]))/M\n",
+    "    '''\n",
+    "\n",
+    "    M = max((max(v_matrix[:,0]) - min(v_matrix[:,0])), (max(v_matrix[:,1]) - min(v_matrix[:,1])), (max(v_matrix[:,2]) - min(v_matrix[:,2])))\n",
+    "    x = (v_matrix[:,0] - np.mean(v_matrix[:,0]))/M\n",
+    "    y = (v_matrix[:,1] - np.mean(v_matrix[:,1]))/ M\n",
+    "    z = (v_matrix[:,2] - np.mean(v_matrix[:,2]))/M\n",
+    "\n",
+    "    minimo = min(min(x), min(y), min(z))\n",
+    "    maximo = max(max(x), max(y), max(z))\n",
+    "  \n",
     "    fig = go.Figure(go.Mesh3d(\n",
-    "            x=v_matrix[:,0], y=v_matrix[:,1], z=v_matrix[:,2], \n",
+    "            x=x, y=y, z=z, \n",
     "            i=f_matrix[:,0], j=f_matrix[:,1], k=f_matrix[:,2], \n",
     "            color='red'))\n",
+    "\n",
+    "    m = go.Mesh3d(\n",
+    "            x=v_matrix[:,0], y=v_matrix[:,1], z=v_matrix[:,2], \n",
+    "            i=f_matrix[:,0], j=f_matrix[:,1], k=f_matrix[:,2], \n",
+    "            color='red')\n",
     "    \n",
     "    fig.update_layout(\n",
-    "    scene = dict(\n",
-    "        xaxis = dict(visible=False),\n",
-    "        yaxis = dict(visible=False),\n",
-    "        zaxis =dict(visible=False)\n",
-    "        )\n",
-    "    )\n",
-    "    \n",
+    "        scene = dict(\n",
+    "            xaxis_range=[minimo, maximo],\n",
+    "            yaxis_range=[minimo, maximo],\n",
+    "            zaxis_range=[minimo, maximo],\n",
+    "            aspectratio=dict(x = 1, y = 1, z = 1),\n",
+    "            xaxis = dict(visible=False),\n",
+    "            yaxis = dict(visible=False),\n",
+    "            zaxis =dict(visible=False),\n",
+    "        ))\n",
     "    fig2 = go.Figure(go.Mesh3d(\n",
     "            x=v_matrix_2[:,0], y=v_matrix_2[:,1], z=v_matrix_2[:,2], \n",
     "            i=f_matrix_2[:,0], j=f_matrix_2[:,1], k=f_matrix_2[:,2], \n",
     "            color='red'))\n",
     "    \n",
-    "    fig2.update_layout(\n",
-    "    scene = dict(\n",
-    "        xaxis = dict(visible=False),\n",
-    "        yaxis = dict(visible=False),\n",
-    "        zaxis =dict(visible=False)\n",
-    "        )\n",
-    "    )\n",
-    "    o3d.io.write_triangle_mesh(\"output.obj\", mesh_o3d)\n",
-    "    plotTree(generated_images, True)\n",
     "\n",
+    "    o3d.io.write_triangle_mesh(\"output.obj\", mesh_o3d)\n",
     "    #o3d.io.write_triangle_mesh(\"generadas/\"+filename.split(\"_\")[0]+\".obj\", mesh_o3d)\n",
-    "    return fig"
+    "    return fig\n"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 17,
+   "execution_count": 10,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Running on local URL:  http://127.0.0.1:7860\n",
-      "generated images <modelovae.Node object at 0x000001F3A9D89FC0>\n"
-     ]
-    },
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "a7c9d42f5242485e92437771b60ab88c",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "Renderer(camera=PerspectiveCamera(children=(DirectionalLight(color='white', intensity=0.6, position=(-0.003762…"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "generated images <modelovae.Node object at 0x000001F3B5E47130>\n"
-     ]
-    },
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "0beb7306167340709957143fe3d8412d",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "Renderer(camera=PerspectiveCamera(children=(DirectionalLight(color='white', intensity=0.6, position=(0.4200476…"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "generated images <modelovae.Node object at 0x000001F3B5E47820>\n"
-     ]
-    },
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "15699e29642b4c969757709a93569c09",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "Renderer(camera=PerspectiveCamera(children=(DirectionalLight(color='white', intensity=0.6, position=(-0.002811…"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "generated images <modelovae.Node object at 0x000001F3B8AA2F50>\n"
-     ]
-    },
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "e8f7321033e74f2cb61d0830f179ce4a",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "Renderer(camera=PerspectiveCamera(children=(DirectionalLight(color='white', intensity=0.6, position=(0.4228083…"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "generated images <modelovae.Node object at 0x000001F3B8AF30D0>\n"
-     ]
-    },
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "e45f9cfb42a742928111bd96c0bea6c8",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "Renderer(camera=PerspectiveCamera(children=(DirectionalLight(color='white', intensity=0.6, position=(0.2544228…"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "generated images <modelovae.Node object at 0x000001F3B8AA35B0>\n"
-     ]
-    },
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "e8b93804c83f437ba86dfb28fc839eda",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "Renderer(camera=PerspectiveCamera(children=(DirectionalLight(color='white', intensity=0.6, position=(0.2635734…"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "generated images <modelovae.Node object at 0x000001F3B8AF09A0>\n"
-     ]
-    },
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "7a670daef41b4e1a8aad5cab83ac220c",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "Renderer(camera=PerspectiveCamera(children=(DirectionalLight(color='white', intensity=0.6, position=(0.5811820…"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "generated images <modelovae.Node object at 0x000001F3B8AF07F0>\n"
-     ]
-    },
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "4a0fa599af6846d299e6868130c17a2b",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "Renderer(camera=PerspectiveCamera(children=(DirectionalLight(color='white', intensity=0.6, position=(0.3441835…"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "generated images <modelovae.Node object at 0x000001F3B8AA2F50>\n"
+      "Running on local URL:  http://127.0.0.1:7861\n",
+      "\n",
+      "To create a public link, set `share=True` in `launch()`.\n"
      ]
     },
     {
      "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "1921954654de45e7835c0ecd7bb0c55a",
-       "version_major": 2,
-       "version_minor": 0
-      },
+      "text/html": [
+       "<div><iframe src=\"http://127.0.0.1:7861/\" width=\"100%\" height=\"500\" allow=\"autoplay; camera; microphone; clipboard-read; clipboard-write;\" frameborder=\"0\" allowfullscreen></iframe></div>"
+      ],
       "text/plain": [
-       "Renderer(camera=PerspectiveCamera(children=(DirectionalLight(color='white', intensity=0.6, position=(0.3771645…"
+       "<IPython.core.display.HTML object>"
       ]
      },
      "metadata": {},
      "output_type": "display_data"
     },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "generated images <modelovae.Node object at 0x000001F3B8A18F10>\n"
-     ]
-    },
     {
      "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "2f112a2375f74531954a5bc3d4de338e",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "Renderer(camera=PerspectiveCamera(children=(DirectionalLight(color='white', intensity=0.6, position=(0.3513652…"
-      ]
+      "text/plain": []
      },
+     "execution_count": 10,
      "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "generated images <modelovae.Node object at 0x000001F3B8C315A0>\n"
-     ]
+     "output_type": "execute_result"
     },
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "69653cc85476484c9da5003f8dfaf59d",
+       "model_id": "65a760af40bf45d587e9431bfb1a1bb3",
        "version_major": 2,
        "version_minor": 0
       },
       "text/plain": [
-       "Renderer(camera=PerspectiveCamera(children=(DirectionalLight(color='white', intensity=0.6, position=(0.0813838…"
+       "Renderer(camera=PerspectiveCamera(children=(DirectionalLight(color='white', intensity=0.6, position=(0.4971382…"
       ]
      },
      "metadata": {},
      "output_type": "display_data"
     },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "generated images <modelovae.Node object at 0x000001F3B8C566E0>\n"
-     ]
-    },
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "a01e1fdbdf824caebcc93977b0c2bb4b",
+       "model_id": "4d4cd747115c4c67924a26739200aae7",
        "version_major": 2,
        "version_minor": 0
       },
       "text/plain": [
-       "Renderer(camera=PerspectiveCamera(children=(DirectionalLight(color='white', intensity=0.6, position=(0.1389834…"
+       "Renderer(camera=PerspectiveCamera(children=(DirectionalLight(color='white', intensity=0.6, position=(-0.000256…"
       ]
      },
      "metadata": {},
      "output_type": "display_data"
     },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "generated images <modelovae.Node object at 0x000001F3B8A70B20>\n"
-     ]
-    },
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "39ec4db0797340ad8a9ca1e3418a8a68",
+       "model_id": "48ea3f83128d4a6da225a5c9e53fcee6",
        "version_major": 2,
        "version_minor": 0
       },
       "text/plain": [
-       "Renderer(camera=PerspectiveCamera(children=(DirectionalLight(color='white', intensity=0.6, position=(0.0445088…"
+       "Renderer(camera=PerspectiveCamera(children=(DirectionalLight(color='white', intensity=0.6, position=(0.7299671…"
       ]
      },
      "metadata": {},
      "output_type": "display_data"
     },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "generated images <modelovae.Node object at 0x000001F3B8CE1270>\n"
-     ]
-    },
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "7ace303e1bf347d9be2fb076afdda5eb",
+       "model_id": "e0634d6c173b40959725335707e53957",
        "version_major": 2,
        "version_minor": 0
       },
       "text/plain": [
-       "Renderer(camera=PerspectiveCamera(children=(DirectionalLight(color='white', intensity=0.6, position=(0.5377806…"
+       "Renderer(camera=PerspectiveCamera(children=(DirectionalLight(color='white', intensity=0.6, position=(0.9065635…"
       ]
      },
      "metadata": {},
@@ -606,37 +425,22 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "generated images <modelovae.Node object at 0x000001F3B8CE1CC0>\n"
+      "ratio 13.516594372719945\n"
      ]
     },
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "e2719b6258844c46a006bede8f7b79ab",
+       "model_id": "d63e52ba4fc4439b8a0e4ecad9be258e",
        "version_major": 2,
        "version_minor": 0
       },
       "text/plain": [
-       "Renderer(camera=PerspectiveCamera(children=(DirectionalLight(color='white', intensity=0.6, position=(0.4734578…"
+       "Renderer(camera=PerspectiveCamera(children=(DirectionalLight(color='white', intensity=0.6, position=(0.0338801…"
       ]
      },
      "metadata": {},
      "output_type": "display_data"
-    },
-    {
-     "ename": "KeyboardInterrupt",
-     "evalue": "",
-     "output_type": "error",
-     "traceback": [
-      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
-      "\u001b[1;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
-      "Cell \u001b[1;32mIn[17], line 12\u001b[0m\n\u001b[0;32m      9\u001b[0m checkpoint \u001b[39m=\u001b[39m torch\u001b[39m.\u001b[39mload(\u001b[39m\"\u001b[39m\u001b[39m64.pth\u001b[39m\u001b[39m\"\u001b[39m, map_location\u001b[39m=\u001b[39mtorch\u001b[39m.\u001b[39mdevice(\u001b[39m'\u001b[39m\u001b[39mcpu\u001b[39m\u001b[39m'\u001b[39m))\n\u001b[0;32m     10\u001b[0m Grassdecoder\u001b[39m.\u001b[39mload_state_dict(checkpoint[\u001b[39m'\u001b[39m\u001b[39mdecoder_state_dict\u001b[39m\u001b[39m'\u001b[39m])\n\u001b[1;32m---> 12\u001b[0m gr\u001b[39m.\u001b[39;49mInterface( predict, \n\u001b[0;32m     13\u001b[0m     inputs\u001b[39m=\u001b[39;49m\u001b[39mNone\u001b[39;49;00m,\n\u001b[0;32m     14\u001b[0m     outputs\u001b[39m=\u001b[39;49mgr\u001b[39m.\u001b[39;49mPlot()\n\u001b[0;32m     15\u001b[0m )\u001b[39m.\u001b[39;49mlaunch(share \u001b[39m=\u001b[39;49m \u001b[39mTrue\u001b[39;49;00m)\n",
-      "File \u001b[1;32mc:\\Users\\User\\anaconda3\\envs\\py_torch\\lib\\site-packages\\gradio\\blocks.py:2005\u001b[0m, in \u001b[0;36mBlocks.launch\u001b[1;34m(self, inline, inbrowser, share, debug, enable_queue, max_threads, auth, auth_message, prevent_thread_lock, show_error, server_name, server_port, show_tips, height, width, encrypt, favicon_path, ssl_keyfile, ssl_certfile, ssl_keyfile_password, ssl_verify, quiet, show_api, file_directories, allowed_paths, blocked_paths, root_path, _frontend, app_kwargs)\u001b[0m\n\u001b[0;32m   2003\u001b[0m \u001b[39mtry\u001b[39;00m:\n\u001b[0;32m   2004\u001b[0m     \u001b[39mif\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mshare_url \u001b[39mis\u001b[39;00m \u001b[39mNone\u001b[39;00m:\n\u001b[1;32m-> 2005\u001b[0m         \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mshare_url \u001b[39m=\u001b[39m networking\u001b[39m.\u001b[39;49msetup_tunnel(\n\u001b[0;32m   2006\u001b[0m             \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mserver_name, \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mserver_port, \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mshare_token\n\u001b[0;32m   2007\u001b[0m         )\n\u001b[0;32m   2008\u001b[0m     \u001b[39mprint\u001b[39m(strings\u001b[39m.\u001b[39men[\u001b[39m\"\u001b[39m\u001b[39mSHARE_LINK_DISPLAY\u001b[39m\u001b[39m\"\u001b[39m]\u001b[39m.\u001b[39mformat(\u001b[39mself\u001b[39m\u001b[39m.\u001b[39mshare_url))\n\u001b[0;32m   2009\u001b[0m     \u001b[39mif\u001b[39;00m \u001b[39mnot\u001b[39;00m (quiet):\n",
-      "File \u001b[1;32mc:\\Users\\User\\anaconda3\\envs\\py_torch\\lib\\site-packages\\gradio\\networking.py:228\u001b[0m, in \u001b[0;36msetup_tunnel\u001b[1;34m(local_host, local_port, share_token)\u001b[0m\n\u001b[0;32m    224\u001b[0m     remote_host, remote_port \u001b[39m=\u001b[39m payload[\u001b[39m\"\u001b[39m\u001b[39mhost\u001b[39m\u001b[39m\"\u001b[39m], \u001b[39mint\u001b[39m(payload[\u001b[39m\"\u001b[39m\u001b[39mport\u001b[39m\u001b[39m\"\u001b[39m])\n\u001b[0;32m    225\u001b[0m     tunnel \u001b[39m=\u001b[39m Tunnel(\n\u001b[0;32m    226\u001b[0m         remote_host, remote_port, local_host, local_port, share_token\n\u001b[0;32m    227\u001b[0m     )\n\u001b[1;32m--> 228\u001b[0m     address \u001b[39m=\u001b[39m tunnel\u001b[39m.\u001b[39;49mstart_tunnel()\n\u001b[0;32m    229\u001b[0m     \u001b[39mreturn\u001b[39;00m address\n\u001b[0;32m    230\u001b[0m \u001b[39mexcept\u001b[39;00m \u001b[39mException\u001b[39;00m \u001b[39mas\u001b[39;00m e:\n",
-      "File \u001b[1;32mc:\\Users\\User\\anaconda3\\envs\\py_torch\\lib\\site-packages\\gradio\\tunneling.py:59\u001b[0m, in \u001b[0;36mTunnel.start_tunnel\u001b[1;34m(self)\u001b[0m\n\u001b[0;32m     57\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39mstart_tunnel\u001b[39m(\u001b[39mself\u001b[39m) \u001b[39m-\u001b[39m\u001b[39m>\u001b[39m \u001b[39mstr\u001b[39m:\n\u001b[0;32m     58\u001b[0m     \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mdownload_binary()\n\u001b[1;32m---> 59\u001b[0m     \u001b[39mself\u001b[39m\u001b[39m.\u001b[39murl \u001b[39m=\u001b[39m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49m_start_tunnel(BINARY_PATH)\n\u001b[0;32m     60\u001b[0m     \u001b[39mreturn\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39murl\n",
-      "File \u001b[1;32mc:\\Users\\User\\anaconda3\\envs\\py_torch\\lib\\site-packages\\gradio\\tunneling.py:95\u001b[0m, in \u001b[0;36mTunnel._start_tunnel\u001b[1;34m(self, binary)\u001b[0m\n\u001b[0;32m     93\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mproc\u001b[39m.\u001b[39mstdout \u001b[39mis\u001b[39;00m \u001b[39mNone\u001b[39;00m:\n\u001b[0;32m     94\u001b[0m     \u001b[39mcontinue\u001b[39;00m\n\u001b[1;32m---> 95\u001b[0m line \u001b[39m=\u001b[39m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mproc\u001b[39m.\u001b[39;49mstdout\u001b[39m.\u001b[39;49mreadline()\n\u001b[0;32m     96\u001b[0m line \u001b[39m=\u001b[39m line\u001b[39m.\u001b[39mdecode(\u001b[39m\"\u001b[39m\u001b[39mutf-8\u001b[39m\u001b[39m\"\u001b[39m)\n\u001b[0;32m     97\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39m\"\u001b[39m\u001b[39mstart proxy success\u001b[39m\u001b[39m\"\u001b[39m \u001b[39min\u001b[39;00m line:\n",
-      "\u001b[1;31mKeyboardInterrupt\u001b[0m: "
-     ]
     }
    ],
    "source": [
@@ -648,7 +452,7 @@
     "\n",
     "Grassdecoder.eval()\n",
     "\n",
-    "checkpoint = torch.load(\"64.pth\", map_location=torch.device('cpu'))\n",
+    "checkpoint = torch.load(\"modelo-bet-nuevacc.pth\", map_location=torch.device('cpu'))\n",
     "Grassdecoder.load_state_dict(checkpoint['decoder_state_dict'])\n",
     "\n",
     "gr.Interface( predict, \n",

app.py

@@ -119,6 +119,14 @@ def tr2(root):
 use_gpu = True
 device = torch.device("cuda:0" if use_gpu and torch.cuda.is_available() else "cpu")
 
+def traversefeatures(root, features):
+       
+    if root is not None:
+        traversefeatures(root.left, features)
+        features.append(root.radius.tolist()[0][3])
+        traversefeatures(root.right, features)
+        return features
+    
 def my_collate(batch):
     return batch
 
@@ -194,12 +202,16 @@ def predict():
         
         z = torch.randn(1, latent_size)
         generated_images = decode_testing(z, 100, Grassdecoder, mult, 1)
-        print("generated images", generated_images)
         count = []
         numerar_nodos(generated_images, count)
+        plotTree(generated_images, True)
         #tr(generated_images)
         generated_images.toGraph( G, 0, True, flag = 0)
-
+        r_list = []
+        r_list = traversefeatures(generated_images, r_list)
+        max_radius = max(r_list)
+        min_radius = min(r_list)
+        
         '''
         #generated_images = iter(data_loader).next()[0]
         generated_images = next(iter(data_loader))[0]
@@ -221,36 +233,48 @@ def predict():
         #nx.write_gpickle(graph, "grafo0.gpickle" )
         for arista in G.edges():
             nx.set_edge_attributes( G, {arista : {'procesada':False}})
-        graph_resampled = resamplear(G, puntosPorUnidad=5)
+        #graph_resampled = resamplear(G, puntosPorUnidad=5)
         graph_resampled = G
         prepararAristas(graph_resampled)
         graphOfCenters = GrafoCentros(graph_resampled)
         
-
-        try:
-            graphOfCenters.tile()
-            mesh = tm.Trimesh( graphOfCenters.getVertices(), graphOfCenters.getCaras() )
-            mesh_o3d = mesh.as_open3d
-            m2 = mesh_o3d 
-            o3d.io.write_triangle_mesh("sinsub.obj", m2)
-            mesh_o3d = mesh_o3d.subdivide_loop(3)
+        if abs(max_radius/min_radius)<10:
+            try:
+                graphOfCenters.tile()
+                mesh = tm.Trimesh( graphOfCenters.getVertices(), graphOfCenters.getCaras() )
+                mesh_o3d = mesh.as_open3d
+                m2 = mesh_o3d 
+                o3d.io.write_triangle_mesh("sinsub.obj", m2)
+                mesh_o3d = mesh_o3d.subdivide_loop(3)
             
-            v_matrix = np.asarray(mesh_o3d.vertices )
-            f_matrix = np.asarray(mesh_o3d.triangles )
-            v_matrix_2 = np.asarray(m2.vertices )
-            f_matrix_2 = np.asarray(m2.triangles )
-            success = True
+                v_matrix = np.asarray(mesh_o3d.vertices )
+                f_matrix = np.asarray(mesh_o3d.triangles )
+                v_matrix_2 = np.asarray(m2.vertices )
+                f_matrix_2 = np.asarray(m2.triangles )
+                success = True
             
-        except Exception as e:
-            print("No se pudo generar")
-            print(str(e))
+            except Exception as e:
+                print("No se pudo generar")
+                print(str(e))
+                pass
+        else:
+            print("ratio", max_radius/min_radius)
             pass
-    
+    '''
     M = max((max(v_matrix[:,0]) - min(v_matrix[:,0])), (max(v_matrix[:,1]) - min(v_matrix[:,1])), (max(v_matrix[:,2]) - min(v_matrix[:,2])))
     x = (v_matrix[:,0] - min(v_matrix[:,0]))/M
     y = (v_matrix[:,1] - min(v_matrix[:,1]))/ M
     z = (v_matrix[:,2] - min(v_matrix[:,2]))/M
+    '''
 
+    M = max((max(v_matrix[:,0]) - min(v_matrix[:,0])), (max(v_matrix[:,1]) - min(v_matrix[:,1])), (max(v_matrix[:,2]) - min(v_matrix[:,2])))
+    x = (v_matrix[:,0] - np.mean(v_matrix[:,0]))/M
+    y = (v_matrix[:,1] - np.mean(v_matrix[:,1]))/ M
+    z = (v_matrix[:,2] - np.mean(v_matrix[:,2]))/M
+
+    minimo = min(min(x), min(y), min(z))
+    maximo = max(max(x), max(y), max(z))
+  
     fig = go.Figure(go.Mesh3d(
             x=x, y=y, z=z, 
             i=f_matrix[:,0], j=f_matrix[:,1], k=f_matrix[:,2], 
@@ -263,9 +287,9 @@ def predict():
     
     fig.update_layout(
         scene = dict(
-            xaxis_range=[0,1],
-            yaxis_range=[0,1],
-            zaxis_range=[0,1],
+            xaxis_range=[minimo, maximo],
+            yaxis_range=[minimo, maximo],
+            zaxis_range=[minimo, maximo],
             aspectratio=dict(x = 1, y = 1, z = 1),
             xaxis = dict(visible=False),
             yaxis = dict(visible=False),
@@ -281,9 +305,10 @@ def predict():
     #o3d.io.write_triangle_mesh("generadas/"+filename.split("_")[0]+".obj", mesh_o3d)
     return fig
 
+
 a = [1.,1.,1.]
 mult = torch.Tensor(a)
-latent_size = 64
+latent_size = 32
 Grassdecoder = GRASSDecoder(latent_size=latent_size, hidden_size=256, mult = mult)
 Grassdecoder = Grassdecoder
 

meshSubplot.py

@@ -0,0 +1,111 @@
+import uuid
+from ipywidgets import Output, HBox
+import meshplot as mp
+rendertype = 'JUPYTER'
+import numpy as np
+import networkx as nx
+
+class Subplot():
+    def __init__(self, data, view, s):
+        if data == None:
+            self.rows = []
+            self.hboxes = []
+        else:
+            self.rows = data.rows
+        if s[0] != 1 or s[1] != 1:
+            if data == None: # Intialize subplot array
+                cnt = 0
+                for r in range(s[0]):
+                    row = []
+                    for c in range(s[1]):
+                        row.append(Output())
+                        cnt += 1
+                    self.rows.append(row)
+
+                for r in self.rows:
+                    hbox = HBox(r)
+                    if rendertype == "JUPYTER":
+                        display(hbox)
+                    self.hboxes.append(hbox)
+        
+            out = self.rows[int(s[2]/s[1])][s[2]%s[1]]
+            if rendertype == "JUPYTER":
+                with out:
+                    display(view._renderer)
+            self.rows[int(s[2]/s[1])][s[2]%s[1]] = view
+
+    def save(self, filename=""):
+        if filename == "":
+            uid = str(uuid.uuid4()) + ".html"
+        else:
+            filename = filename.replace(".html", "")
+            uid = filename + '.html'
+
+        s = ""
+        imports = True
+        for r in self.rows:
+            for v in r:
+                s1 = v.to_html(imports=imports, html_frame=False)
+                s = s + s1
+                imports = False
+
+        s = "<html>\n<body>\n" + s + "\n</body>\n</html>"
+        with open(uid, "w") as f:
+            f.write(s)
+        print("Plot saved to file %s."%uid)
+
+    def to_html(self, imports=True, html_frame=True):
+        s = ""
+        for r in self.rows:
+            for v in r:
+                s1 = v.to_html(imports=imports, html_frame=html_frame)
+                s = s + s1
+                imports = False
+
+        return s
+
+def subplot(f, c = 'red', uv=None, n=None, shading={}, s=[1, 1, 0], data=None, **kwargs):
+    
+    shading={'point_size':0.05, "point_color": c, "line_color": c, "width":400, "height":400}
+    view = mp.Viewer(settings = {"width": 500, "height": 500, "antialias": True, "scale": 1.5, "background": "#ffffff",
+                "fov": 30})
+
+    #obj = view.add_points(np.array([ f.nodes[v]['posicion'] for v in f.nodes]), shading=shading)
+    obj = view.add_points( np.array([f.nodes[v]['posicion'] for v in f.nodes if f.nodes[v]['root'] == True]), shading={'point_size':.02, 'point_color':'red'})
+    if len(np.array([ f.nodes[v]['posicion'] for v in f.nodes if f.nodes[v]['root'] == False])) != 0:
+        obj = view.add_points( np.array([f.nodes[v]['posicion'] for v in f.nodes if f.nodes[v]['root'] == False]), shading={'point_size':.02, 'point_color':'black'})
+    for arista in f.edges:
+        obj = view.add_lines( f.nodes[arista[0]]['posicion'], f.nodes[arista[1]]['posicion'], shading  = shading)
+   
+    subplot = Subplot(data, view, s)
+    return subplot
+
+def plotTree( root, dec ):
+    graph = nx.Graph()
+    root.toGraph( graph, 0, dec, 0)
+    edges=nx.get_edge_attributes(graph,'procesada')
+
+    p = mp.plot( np.array([ graph.nodes[v]['posicion'] for v in graph.nodes if graph.nodes[v]['root'] == True]), shading={'point_size':0.05, 'point_color':'red'}, return_plot=True)
+    if len(np.array([ graph.nodes[v]['posicion'] for v in graph.nodes if graph.nodes[v]['root'] == False])) != 0:
+        p.add_points( np.array([graph.nodes[v]['posicion'] for v in graph.nodes if graph.nodes[v]['root'] == False]), shading={'point_size':.05, 'point_color':'black'})
+    for arista in graph.edges:
+        p.add_lines( graph.nodes[arista[0]]['posicion'], graph.nodes[arista[1]]['posicion'])
+
+    return 
+
+
+def sTree( root, dec, s, c, d=None):
+    "plot trees next to each other"
+    graph = nx.Graph()
+    root.toGraph( graph, 0, dec, 0)
+  
+    if d:
+        subplot(graph, c=c, s=s, data = d)
+    else:
+        
+        d = subplot(graph, c=c, s=s)
+
+    return d
+
+
+

modelo-best-nueva.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9287134af35df22f84fb65f5b6ec96ea8ed6cfc5bbb48a823524cc85a09bda97
+size 17540213

modelo-best-nuevacc.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1ca26f006816c788f1aa282341630318dfac93953e1f0608d15ee837efc6afa7
+size 6730273

sinsub.obj

@@ -1,35 +1,19 @@
 # Created by Open3D 
 # object name: sinsub
-# number of vertices: 28
-# number of triangles: 48
-v 0.25604 0.465126 0.574021
-v 0.252748 0.447025 0.598665
-v 0.225432 0.459906 0.604476
-v 0.228725 0.478007 0.579832
-v 0.166668 0.321205 0.455716
-v 0.167404 0.313618 0.490033
-v 0.133924 0.323905 0.493025
-v 0.133188 0.331492 0.458709
-v 0.152542 0.267385 0.451735
-v 0.158277 0.248919 0.496876
-v 0.139361 0.289971 0.516073
-v 0.133626 0.308438 0.470932
-v 0.0874069 0.321796 0.429109
-v 0.0589153 0.30421 0.465712
-v 0.0862486 0.328973 0.498885
-v 0.11474 0.346559 0.462283
-v 0.0626304 0.383599 0.434658
-v 0.0311964 0.366872 0.468706
-v 0.0673745 0.366957 0.502148
-v 0.0988085 0.383684 0.4681
-v 0.0594784 0.458841 0.463877
-v 0.0233739 0.439053 0.49896
-v 0.0635398 0.41784 0.52833
-v 0.0996443 0.437628 0.493247
-v 0.0771643 0.546611 0.530904
-v 0.0343078 0.521689 0.570211
-v 0.0797324 0.487653 0.598157
-v 0.122589 0.512575 0.558851
+# number of vertices: 12
+# number of triangles: 16
+v 0.449378 0.00737204 0.0931694
+v 0.453679 -0.00822981 0.0918463
+v 0.443061 -0.0101272 0.0797087
+v 0.43876 0.00547462 0.0810317
+v 0.540627 0.0390201 0.0148946
+v 0.554566 0.0244183 0.00205357
+v 0.538313 0.0243431 -0.0155028
+v 0.524375 0.0389449 -0.00266178
+v 0.539596 0.11558 0.00564692
+v 0.54497 0.11561 -0.000638385
+v 0.538685 0.115594 -0.00601316
+v 0.53331 0.115563 0.000272139
 f 5 1 2
 f 6 2 3
 f 7 3 4
@@ -38,22 +22,6 @@ f 9 5 6
 f 10 6 7
 f 11 7 8
 f 12 8 5
-f 13 9 10
-f 14 10 11
-f 15 11 12
-f 16 12 9
-f 17 13 14
-f 18 14 15
-f 19 15 16
-f 20 16 13
-f 21 17 18
-f 22 18 19
-f 23 19 20
-f 24 20 17
-f 25 21 22
-f 26 22 23
-f 27 23 24
-f 28 24 21
 f 2 6 5
 f 3 7 6
 f 4 8 7
@@ -62,19 +30,3 @@ f 6 10 9
 f 7 11 10
 f 8 12 11
 f 5 9 12
-f 10 14 13
-f 11 15 14
-f 12 16 15
-f 9 13 16
-f 14 18 17
-f 15 19 18
-f 16 20 19
-f 13 17 20
-f 18 22 21
-f 19 23 22
-f 20 24 23
-f 17 21 24
-f 22 26 25
-f 23 27 26
-f 24 28 27
-f 21 25 28