section plot now works

Gradio_app.ipynb

@@ -152,7 +152,28 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 64,
+   "execution_count": 75,
+   "metadata": {},
+   "outputs": [
+    {
+     "ename": "NameError",
+     "evalue": "name 't0s' is not defined",
+     "output_type": "error",
+     "traceback": [
+      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+      "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
+      "Cell \u001b[0;32mIn[75], line 2\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[39mimport\u001b[39;00m \u001b[39mpandas\u001b[39;00m \u001b[39mas\u001b[39;00m \u001b[39mpd\u001b[39;00m\n\u001b[0;32m----> 2\u001b[0m pd\u001b[39m.\u001b[39mdate_range(start\u001b[39m=\u001b[39mt0s[i], periods\u001b[39m=\u001b[39mwaveforms[i][\u001b[39m0\u001b[39m]\u001b[39m.\u001b[39mshape[\u001b[39m-\u001b[39m\u001b[39m1\u001b[39m], freq\u001b[39m=\u001b[39m\u001b[39m'\u001b[39m\u001b[39m1s\u001b[39m\u001b[39m'\u001b[39m)\n",
+      "\u001b[0;31mNameError\u001b[0m: name 't0s' is not defined"
+     ]
+    }
+   ],
+   "source": [
+    "import pandas as pd\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 114,
    "metadata": {},
    "outputs": [
     {
@@ -167,7 +188,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Running on local URL:  http://127.0.0.1:7914\n",
+      "Running on local URL:  http://127.0.0.1:7935\n",
       "\n",
       "To create a public link, set `share=True` in `launch()`.\n"
      ]
@@ -175,7 +196,7 @@
     {
      "data": {
       "text/html": [
-       "<div><iframe src=\"http://127.0.0.1:7914/\" width=\"100%\" height=\"500\" allow=\"autoplay; camera; microphone; clipboard-read; clipboard-write;\" frameborder=\"0\" allowfullscreen></iframe></div>"
+       "<div><iframe src=\"http://127.0.0.1:7935/\" width=\"100%\" height=\"500\" allow=\"autoplay; camera; microphone; clipboard-read; clipboard-write;\" frameborder=\"0\" allowfullscreen></iframe></div>"
       ],
       "text/plain": [
        "<IPython.core.display.HTML object>"
@@ -188,16 +209,9 @@
      "data": {
       "text/plain": []
      },
-     "execution_count": 64,
+     "execution_count": 114,
      "metadata": {},
      "output_type": "execute_result"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "torch.Size([256])\n"
-     ]
     }
    ],
    "source": [
@@ -205,6 +219,7 @@
     "\n",
     "import gradio as gr\n",
     "import numpy as np\n",
+    "import pandas as pd\n",
     "from phasehunter.model import Onset_picker, Updated_onset_picker\n",
     "from phasehunter.data_preparation import prepare_waveform\n",
     "import torch\n",
@@ -221,6 +236,7 @@
     "from obspy.clients.fdsn.header import URL_MAPPINGS\n",
     "\n",
     "import matplotlib.pyplot as plt\n",
+    "import matplotlib.dates as mdates\n",
     "\n",
     "def make_prediction(waveform):\n",
     "    waveform = np.load(waveform)\n",
@@ -328,7 +344,10 @@
     "                        continue\n",
     "\n",
     "                    if len(waveform) == 3:\n",
-    "                        waveform = prepare_waveform(np.stack([x.data for x in waveform]))\n",
+    "                        try:\n",
+    "                            waveform = prepare_waveform(np.stack([x.data for x in waveform]))\n",
+    "                        except:\n",
+    "                            continue\n",
     "                    \n",
     "                        distances.append(distance)\n",
     "                        t0s.append(starttime)\n",
@@ -346,19 +365,31 @@
     "    p_phases = output[:, 0]\n",
     "    s_phases = output[:, 1]\n",
     "\n",
-    "\n",
-    "    print(p_phases.shape)\n",
-    "    # for i in range(len(waveforms)):\n",
-    "    #     current_P = P_batch[i::len(waveforms)].cpu()\n",
-    "    #     current_S_batch = S_batch[i::len(waveforms)].cpu()\n",
-    "    #     current_Pg_batch = Pg_batch[i::len(waveforms)].cpu()\n",
-    "    #     current_Sg_batch = Sg_batch[i::len(waveforms)].cpu()\n",
-    "    #     current_Pn_batch = Pn_batch[i::len(waveforms)].cpu()\n",
-    "    #     current_Sn_batch = Sn_batch[i::len(waveforms)].cpu()\n",
-    "    \n",
-    "    fig,ax = plt.subplots()\n",
-    "    ax.scatter(st_lats, st_lons)\n",
-    "    fig.canvas.draw()\n",
+    "    fig, ax = plt.subplots(nrows=1, figsize=(10, 3), sharex=True)\n",
+    "    for i in range(len(waveforms)):\n",
+    "        current_P = p_phases[i::len(waveforms)]\n",
+    "        current_S = s_phases[i::len(waveforms)]\n",
+    "        x = [t0s[i] + pd.Timedelta(seconds=k/100) for k in np.linspace(0,6000,6000)]\n",
+    "        x = mdates.date2num(x)\n",
+    "        ax.plot(x, waveforms[i][0, 0]+distances[i]*111.2, color='black', alpha=0.5)\n",
+    "        ax.scatter(x[int(current_P.mean()*waveforms[i][0].shape[-1])], waveforms[i][0, 0].mean()+distances[i]*111.2, color='r')\n",
+    "        ax.scatter(x[int(current_S.mean()*waveforms[i][0].shape[-1])], waveforms[i][0, 0].mean()+distances[i]*111.2, color='b')\n",
+    "        ax.set_ylabel('Z')\n",
+    "\n",
+    "        ax.xaxis.set_major_formatter(mdates.DateFormatter('%H:%M:%S'))\n",
+    "        ax.xaxis.set_major_locator(mdates.SecondLocator(interval=10))\n",
+    "\n",
+    "        # for a in ax:\n",
+    "        #     a.axvline(current_P.mean()*waveforms[i][0].shape[-1], color='r', linestyle='--', label='P')\n",
+    "        #     a.axvline(current_S.mean()*waveforms[i][0].shape[-1], color='b', linestyle='--', label='S')\n",
+    "\n",
+    "        # ax[-1].set_xlabel('Time, samples')\n",
+    "        # ax[-1].set_ylabel('Uncert.')\n",
+    "        # ax[-1].legend()\n",
+    "\n",
+    "        plt.subplots_adjust(hspace=0., wspace=0.)\n",
+    "        \n",
+    "    fig.canvas.draw();\n",
     "    image = np.array(fig.canvas.renderer.buffer_rgba())\n",
     "    plt.close(fig)\n",
     "\n",
@@ -470,10 +501,34 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 105,
    "metadata": {},
-   "outputs": [],
-   "source": []
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "DatetimeIndex(['2019-07-04 17:33:49', '2019-07-04 17:33:50',\n",
+       "               '2019-07-04 17:33:51', '2019-07-04 17:33:52',\n",
+       "               '2019-07-04 17:33:53', '2019-07-04 17:33:54',\n",
+       "               '2019-07-04 17:33:55', '2019-07-04 17:33:56',\n",
+       "               '2019-07-04 17:33:57', '2019-07-04 17:33:58',\n",
+       "               ...\n",
+       "               '2019-07-04 19:13:39', '2019-07-04 19:13:40',\n",
+       "               '2019-07-04 19:13:41', '2019-07-04 19:13:42',\n",
+       "               '2019-07-04 19:13:43', '2019-07-04 19:13:44',\n",
+       "               '2019-07-04 19:13:45', '2019-07-04 19:13:46',\n",
+       "               '2019-07-04 19:13:47', '2019-07-04 19:13:48'],\n",
+       "              dtype='datetime64[ns]', length=6000, freq='S')"
+      ]
+     },
+     "execution_count": 105,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "pd.date_range(start=obspy.UTCDateTime(\"2019-07-04 17:33:49\").timestamp*1e9, periods=6000, freq='s')"
+   ]
   }
  ],
  "metadata": {

app.py

@@ -2,6 +2,7 @@
 
 import gradio as gr
 import numpy as np
+import pandas as pd
 from phasehunter.model import Onset_picker, Updated_onset_picker
 from phasehunter.data_preparation import prepare_waveform
 import torch
@@ -18,6 +19,7 @@ from obspy.taup.helper_classes import SlownessModelError
 from obspy.clients.fdsn.header import URL_MAPPINGS
 
 import matplotlib.pyplot as plt
+import matplotlib.dates as mdates
 
 def make_prediction(waveform):
     waveform = np.load(waveform)
@@ -125,7 +127,10 @@ def predict_on_section(client_name, timestamp, eq_lat, eq_lon, radius_km, source
                         continue
 
                     if len(waveform) == 3:
-                        waveform = prepare_waveform(np.stack([x.data for x in waveform]))
+                        try:
+                            waveform = prepare_waveform(np.stack([x.data for x in waveform]))
+                        except:
+                            continue
                     
                         distances.append(distance)
                         t0s.append(starttime)
@@ -143,19 +148,31 @@ def predict_on_section(client_name, timestamp, eq_lat, eq_lon, radius_km, source
     p_phases = output[:, 0]
     s_phases = output[:, 1]
 
-
-    print(p_phases.shape)
-    # for i in range(len(waveforms)):
-    #     current_P = P_batch[i::len(waveforms)].cpu()
-    #     current_S_batch = S_batch[i::len(waveforms)].cpu()
-    #     current_Pg_batch = Pg_batch[i::len(waveforms)].cpu()
-    #     current_Sg_batch = Sg_batch[i::len(waveforms)].cpu()
-    #     current_Pn_batch = Pn_batch[i::len(waveforms)].cpu()
-    #     current_Sn_batch = Sn_batch[i::len(waveforms)].cpu()
-    
-    fig,ax = plt.subplots()
-    ax.scatter(st_lats, st_lons)
-    fig.canvas.draw()
+    fig, ax = plt.subplots(nrows=1, figsize=(10, 3), sharex=True)
+    for i in range(len(waveforms)):
+        current_P = p_phases[i::len(waveforms)]
+        current_S = s_phases[i::len(waveforms)]
+        x = [t0s[i] + pd.Timedelta(seconds=k/100) for k in np.linspace(0,6000,6000)]
+        x = mdates.date2num(x)
+        ax.plot(x, waveforms[i][0, 0]+distances[i]*111.2, color='black', alpha=0.5)
+        ax.scatter(x[int(current_P.mean()*waveforms[i][0].shape[-1])], waveforms[i][0, 0].mean()+distances[i]*111.2, color='r')
+        ax.scatter(x[int(current_S.mean()*waveforms[i][0].shape[-1])], waveforms[i][0, 0].mean()+distances[i]*111.2, color='b')
+        ax.set_ylabel('Z')
+
+        ax.xaxis.set_major_formatter(mdates.DateFormatter('%H:%M:%S'))
+        ax.xaxis.set_major_locator(mdates.SecondLocator(interval=10))
+
+        # for a in ax:
+        #     a.axvline(current_P.mean()*waveforms[i][0].shape[-1], color='r', linestyle='--', label='P')
+        #     a.axvline(current_S.mean()*waveforms[i][0].shape[-1], color='b', linestyle='--', label='S')
+
+        # ax[-1].set_xlabel('Time, samples')
+        # ax[-1].set_ylabel('Uncert.')
+        # ax[-1].legend()
+
+        plt.subplots_adjust(hspace=0., wspace=0.)
+        
+    fig.canvas.draw();
     image = np.array(fig.canvas.renderer.buffer_rgba())
     plt.close(fig)
 

requirements.txt

@@ -12,4 +12,5 @@ torchvision==0.15.1
 tqdm==4.65.0
 webdataset==0.2.48
 obspy
+pandas
 git+http://github.com/nikitadurasov/masksembles