diff --git a/anp-rnn_1d_regression.ipynb b/anp-rnn_1d_regression.ipynb index 0ac900b..0839863 100644 --- a/anp-rnn_1d_regression.ipynb +++ b/anp-rnn_1d_regression.ipynb @@ -8,7 +8,7 @@ "\n", "Results:\n", "\n", - "|model|test_loss|\n", + "|model|val_loss|\n", "|-----|---------|\n", "|ANP-RNN| -0.62|\n", "| ANP-RNN(impr)| -1.3217|\n", @@ -11638,10 +11638,11 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 20, "metadata": { "ExecuteTime": { - "start_time": "2020-02-15T23:08:51.400Z" + "end_time": "2020-02-16T01:15:32.866124Z", + "start_time": "2020-02-15T23:08:51.402519Z" } }, "outputs": [ @@ -12079,6 +12080,381 @@ "needs_background": "light" }, "output_type": "display_data" + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "9c570ff564f645a08bc2528c1871a9db", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=0.0), HTML(value='')))" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "train: 20000 loss=-0.2552 val_loss=-0.53\n" + ] + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "2d9e826627594526a68129549a202757", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=0.0), HTML(value='')))" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "train: 30000 loss=-0.3103 val_loss=-0.5162\n" + ] + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "1d5ea79756764e91af3eb20b204ba2e9", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=0.0), HTML(value='')))" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "train: 40000 loss=-0.873 val_loss=-0.6006\n" + ] + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "dd584a68729a407185f5ad66c3091558", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=0.0), HTML(value='')))" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "train: 50000 loss=-1.041 val_loss=-0.8277\n" + ] + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + }, + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX4AAAENCAYAAAAIbA6TAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjAsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+17YcXAAAgAElEQVR4nO3deXxU1cH/8c+ZZCAJIQiy7xBFyiqK4g6yisW6gEv9IdpqUWNtodhWn2KNYGtbpdSKaBV9RNCqYOtOBVQeLQiKCmERgQgiiyhbIPtkcn5/nEw2skxCJtv9vl+v+0qYmXvnJky+59yzXWOtRUREvMNX1ycgIiK1S8EvIuIxCn4REY9R8IuIeIyCX0TEYxT8IiIeo+CXBskY84Qx5t66Pg+RhkjBL7XOGLPTGDPyRI5hrb3NWjuzhs7HGmNOKfg+2RizsCaOW+z4NxljgsaY9GLbsGLPdzfGvG+MyTTGbCn9uzHGTDXGfGuMOWqMecYY07Qm9hXvUvBLvWOMia7rc6iuCs79I2ttfLFtRbHn/gl8DpwM/A5YbIxpU3C8McDdwAigG9ATuL+G9hWvstZq01ZrG7AAyAeygHTgN0B3wAI3A7uADwpeuwj4FkgDPgD6FjvOs8ADBd8PA3YD04DvgH3AT6pwThY4BbgEyAUCBee2vuD5FsDTBcfdAzwARBU8dxOwEpgNHAydU6nj3wT8t5z37gXkAM2LPfYhcFvB9y8Afyz23Ajg2xPdV5u3N9X4pVZZa2/Ahftl1tV8/1Ls6aHAD4AxBf9eApwKtAU+A56v4NDtcQHdCVeAPGaMaVnFc/sP8EfgpYJzG1jw1LNAHq5wGASMBm4ptusQ4CugHfCHcg4/yBhzwBiz1Rhzb7Erg77AV9baY8Veu77g8dDz60s9184Yc/IJ7isepuCX+iTZWpthrc0CsNY+Y609Zq3NAZKBgcaYFuXsGwBmWGsD1tq3cTX20070hIwx7YBLgSkF5/YdrnZ/XbGX7bXWPmqtzQudeykfAP1wBdh44MfArwuei8dd0RSXBjQv5/nQ981PcF/xMAW/1CffhL4xxkQZY/5kjEk1xhwFdhY81bqcfQ9aa/OK/TsTF3wnqhvgB/YZY44YY44A/8CF+HHnXRZr7VfW2h3W2nxr7QZgBjCh4Ol0IKHULgnAsXKeD31/7AT3FQ9T8EtdKG9J2OKPXw9cDozENeF0L3jcRO60jjsHcKGeA7S21p5UsCVYa/tWsE847xH6OTYBPY0xxWvhAwseDz0/sNRz+621B09wX/EwBb/Uhf24ESYVaY4L3INAHK7tvTbsB7obY3wA1tp9wFJgljEmwRjjM8YkGmOGhntAY8zYgiYjjDG9gXuB1wqOvxVYB9xnjIkxxlwJDABeKdj9OeBmY0wfY8xJwHRcn8MJ7SvepuCXuvAgML2g6eSucl7zHPA1bhTNZmB1dd/MGPM/xpglYb58UcHXg8aYzwq+nwQ0KTiPw8BioEMVTmEEkGKMyQDeBv5FyYLsOmBwwbH/BEyw1n4PhR3OfwHex3WKfw3cV0P7ikcZayNzIxZjzARcJ9ZgXHvoLgo+8KVGIYiISC2KZPCvxoX9a7gx1oNwIzO2AOdZa/Mj8sYiIlKhSAZ/m9AlZ7HHJgHzgRHW2vci8sYiIlKhiLXxlw79Ap8UfO0UqfcVEZGK1XbnbmgkxBe1/L4iIlIgYk09x72RMZ1wi0mtt9aOKuc1k4HJALGxsWd26dKlSu+Rn5+Pz6eBSlI39PmTulL8s7d169YD1to2Fb2+VoLfGBMPrAA6Amdba3dXts/gwYPt2rVrq/Q+K1asYNiwYdU5RZETps+f1JXinz1jzKfW2sEVvT7iy98aY2KBN3ATdoaGE/oiIhI5EQ1+Y4wfN9llMDCqYJ0SERGpQxEL/oIp788Dw4Fx1tpqz7wUEZGaE8ka/2PA1bj1yTOMMecUe263mnxEROpGJIcgjC34+jvgo1LbLeXtJCIikRWxGr+1tnukji0iItWnQcciIh6j4BcR8RgFv4iIxyj4RUQ8RsEvIuIxCn4REY9R8IuIeIyCX0TEYxT8IiIeo+AXEfEYBb+IiMco+EVEPEbBLyLiMQp+ERGPUfCLiHiMgl9ExGMU/CIiHqPgFxHxGAW/iIjHKPhFRDxGwS8i4jEKfhERj1Hwi4h4jIJfRMRjFPwiIh6j4BcR8RgFv4iIxyj4RUQ8RsEvIuIxCn4REY9R8IuIeIyCX0TEYxT8IiIeo+AXEfEYBb+IiMco+EVEPEbBLyLiMQp+ERGPUfCLiHiMgl9ExGMU/CIiHqPgFxHxGAW/iIjHKPhFRDxGwS8i4jEKfhERj1Hwi4h4jIJfRMRjFPwiIh6j4BcR8RgFv4iIxyj4RUQ8RsEvIuIxCn4REY9R8IuIeIyCX0TEYxT8IiIeo+AXEfEYBb+IiMco+EVEPEbBLyLiMQp+ERGPUfCLiHiMgl9ExGMU/CIiHqPgFxHxGAW/iIjHKPhFRDxGwS8i4jEKfhERj1Hwi4h4jIJfRMRjFPwiIh6j4BcR8RgFv4iIxyj4RUQ8RsEvIuIxCn4REY9pdMFvLaSnQ26u+15EREqKrusTqGl5ebB2rfveGPD73dakSdlb6Dm/371eRKSxazTBn5qayuzZs/nRj35Eeno6sbHxjBw5kWuumUanTolkZFS8f6iQKKtQCG0tWoCv0V0jiYjXNIrgX7JkCRMmTCAnJ4dgMAhAZuYx3nprHkuXzic5eTFDhoyt8BjWuuah3NzyXxMfD336QFxcTZ69iEjtavD119TUVCZMmEBmZmZh6IcEgwGyszNJTp7Anj2p1Tr+nj2pzJ6dxA9/mMDZZ/to0yaBm25KIjW1escTEalrDT74Z82aRSAQqPA1gUCARYtmV/nYa9Ys4ZZbBvDWW/PIzDyGtZbMzGMsXDiPfv0G8OabS6p72iIidabBB//ChQsrDf5gMMB//jOfdesOsn9/foXNOSF79qSSnDyB7OxMgsGSxw9dSVx99QTWrVPNX0Qalgbfxp+enh7W63Jy0pk6tTXwIXABMTHQtOkigsF/0a7dtXTrdgUtWkBcXDq5uSl88cXj5OZWfiVx//2z+dvf5tC1q0YFiUjD0OCDPz4+nmPHjoXxyih8vuY0b96WjAzIzobs7I+BF0lPH0hRk/0G4HzcxVB+hUcMBgMsXbqAHTvmcPgw/OAH0LTpifw0IiKR1+CDf+LEicybN6/C5p6oKD/jxk1mypQ5gBvBk5kJGzdOYsuWgbRseTqxsZCWBqmpPtasGcLhw2vCev+sLHfFsWlTKjNmzGLZsoVkZKQTHx/PxIkTmTZtGomJiSf+g4qI1JAGH/zTpk1j/vz5FQa/3+/n6qunFv7bGGjWDIYM6c+QIf1LvXoIsJof/jCBzMzKryRiYuJYs2YJyckTCAQChf0Bx44dY968ecyfP5/FixczdmzFw0lFJLK+/vprOnfuTFRUVF2fSp1r8J27iYmJLF68mLi4uOP+Q6Oi/MTExJGcvJhOnapW6x45ciJRUf5KXxcMBrjvvrI7gQOBAJmZmUyYMEHDP0Xq0KZNmzj77LO56aabCod979u3j7lz5/Ltt9/W8dnVvgYf/ABjx44lJSWFcePGkZCQgDE+4uISGDduMvPmpVQ6eass11wzDb+/suCPxu/vS15e5Z3As2dXfTipiNSMtLQ0MjMz2b9/f2HrwKJFi7jjjjtISkqq47OrfY0i+MHV/KdMmcKBA2l8+GGQt99OY8qUOVWu6Yd06pRIcvJiYmLijqv5R0X58fvjiIl5nYyMbcfV9EsLBAIsWLCgWuchjVdqaipJSUkkJCTg8/lISEggKUmTAyPhvPPO44MPPuC1114jJiYGgN69e3PZZZdx/fXXF75uw4YNDB06lOeee66uTrVWNPg2/tL8fjj/fPd9fr5btK2iLRgs/7khQ8Yyb14KixbNZtmyBWRluTWARo26gauvnkowmMiNN4Y3nDTcYafiDaFlRgKBQGENVP1CNWv9+vVkZ2czZMgQAAYNGlTi+dGjRzN69OgSjy1atIgPPviA3r17M2nSpFo719rW6IK/OJ+vaIG16rAW8vISGT9+Dnl5c9i5Ew4cKPma2Nh4srIq7wSOjY0nN7f65yKNR/FlRkoLFQRXXTWBZctS6NcvkWbNXIVGwrd9+3aGDx9OXl4eq1atom/fvmHtN23aNHr16kWfPn0KH8vLy8Pn8+FrRCs0Np6fJAJCK3bGxLgF2vr2hbZtS75m1KjKO4GjovyMHHkDn33mhpGKtz344KywJgc+/PBs1q2DlSth1SpYvx62boU9e9zQ41JLU0kx3bt3Z8SIEQwdOpRTTjkl7P1atGjBxIkTOeOMMwB455136N+/P4sWLYrUqdaJRl3jr2nGuElaUVGwb5977JprprF06fwK2/lDw0mzs+Hzz6FfP7fEs3hLbi6kpsI//7mw0gEBwWCAZcsWFM49Ca0ce/hw0Wt8PkhIgNatoU0bTR4Ed5WekwM5OdE88sjz5ORY9u1rUvj7y8srWm69aVO3Ff++9EjPr7/+mi1btvD0009z7bXX1s0PFQERDX5jTBdgNjAKMMByYIq1dlck3zeSjIHTToPoaPjmm6JO4NLj+EOioqJLDCcNBFz4t24N3bu7K4mGKjU1lVmzZrFw4ULS08ObtFadfRq6/HzYvRu+/trV0kOT/iqTlZVeYfNgfj4cOeK27dshNtYVBKEtPr5xLyOSn+/uthfa9u8/xksvPcUVV9xJdLQfqHr7WFRUyQJh+PCfMnduNDffPLHmf4A6ZGyE7k9ojIkD1gM5wHTAAg8AccAAa22Ft0YZPHiwXRu6lVaYVqxYwbBhw6p1vtWxc6fbwC3qVrwT2No4oBdDhszngQf6EV1OEXvyya4AaN68ds65ppTVOQnu6sbv95fZOVmdfRqSsj5/Bw+6UM7KKnos3MmBkABMJDb2ctq2HU2bNtCyJbRqBZ06QdeucMopbjJiWXw+F/7FC4OCAS0NVk4OfP+9244eLXl71dmzb+f1159gzJgb+dWvniU9HY4dKyoYjh1zS7W0bQvdurmrpHCb7Zs3h/bt3b71sb+l+GfPGPOptXZwRa+PZPD/EvgrcJq1dnvBYz2AbcBvrLV/rWj/hhD84Gr9ZY2+W7cO/ud/3B/8eefBffdV3LF78snuw5iQELlzrSmpqakMGDCgzM7JkLi4OFJSUgpr8eHsExMTw8aNGxtszb/45y8z030uDh4s+Zr8fPjVr5JYv34eUFFzj5+oqF4Eg5uAa4CXynyVMdCrF5x5JvTv775v1aqCo/pdYRAf78IsPt5dKdTnK4NQ2H/3nQv74nJzYccO+PJLWLXqfT777HasXUxeXr9KjxsTA126uK1bN1eQdu0KnTuX/beam5vNV1+tZ8SIIXTrVr8GalQ1+CPZ1PMjYHUo9AGstTuMMSuBy3GFQoPXpYtr9tm6tWTt4/TTYdYsuPtuWLUqwJ13rmX27HPLvXvXwYNua9XKfQjrcx9AuPdAuO+++7juuutITEzk0UcfrXSf7OxspkyZwhtvvFGTp1urgkF3Fbhnjwv54rZvz+Xhhw/w5ZfTgPlUFPwxMX7+8pcn+PzzFXTpchZdusDmzVv44IPHGTBgNvv2+fjqK/jqKxd6X35ZtO/JJ8Opp7qtVy/3tW1bF+6BgOsnKN5XEBXlrhpCBUJ8vPt3Xa5sUFHY79gB777rmky3bnXt9s7FwCYgiujoooKtWbOi75s2df1zu3a538G2bW4rzudztfuuXd3f4kUXQceOB7nttjM4duwwzz+fyrfftqFzZ/eahrgCRCRr/N8Cr1lrby31+Fzgamttm4r2byg1/pDvvoMvvigZ/gBbtmTy85+fTzD4BT17pjB7dq+wavUtW7omoPpSABRvmw9vNVRo2rQpOTk53H333Tz22GNh77d9+/YGWetfvnwFcXHDjrvfQ24uzJ+fzT//OR5rv6BFiw+49NIN/Pvfx/cLucmB/uNuF5qXF2DSpNPYt28Hd9wxmwkTpgDuivLDD7/hsccuAXoSCDxJVlaH484tIcEVAKec4rbExIpDyxh3JVC8MIiPj2wtt6KwP3AA3nsPli1zTWfFz7NTp+/4wQ/a0quX63/r2dPdHrWyq5ijR90V+65dJbe9e48vtE89FfLyLsXa3dx77/P07OnW+GrSxBUOHTvW7VVTfarxtwIOl/H4IaBlWTsYYyYDkwHatWvHihUrqvSG6enpVd6nJrmOu5KPde4MQ4Z04qOPvuerr+7miiuWY206sbGxDBs2kiuvvIYOHTodd6z0dPehjIpyH666rFWsWbOG5OTkgpAKfwxhbm4u55xzDvn5+VWawDZt2jSmTJlSnVOtE/n5ru04EEgnO3tFiedSUk7iscd6sXdvHvA9fv8R7rlnOX37dufii5/ktdcW8f77y8jKyiI2NpaLLx7F5ZdfTYcOsaSnlzzWHXf8nCVLXmfYsNNIT1/Bd9/tp2nTppx1Vhxdu/rZuPFNHn30HZo27U1qanNSU+PZvr05qanNOXrUz6efwqefFh3P78+na9cMevZMp3v3dHr0SKdHjwzi410V+tgxF8LFGeNqxFFRRV9PJPDcXBl3JVI6bDMzo/joo9asWNGO9etbYq17o2bNAlxwwfece+4BAoH3eOih33LJJT/jkksmFB4zo8IeRMfnc6HdrVvJxwMBw759sezeHceWLQksX96Bbdv8wAKaNWvG0qXfcsUV/y38PR065Jp2mzal3L68SKtq9kWyxp8L/NVae3epxx8A7rbWVvgramg1/pDDh2HjxpJjrFesWMyDD95YMHa78tpdWVq0cFcALcssMiMnnLb58iQkJJCWllb4fbg1/uL71Weh9uXQ0N709BXExw8D3Dj7J56A//zHPdetG9x22yHat/+W7t2LJgeddBL07u3am4NBF36VbcEgvPvuMm6//WrOPHMoM2e+yv7937J582ouvPDKwmPn5+fj8/mw1gX4tm2utuw2y/79ZSd2u3buiiC0nXIKdOhQfkdoVNTxVwahZqVQqIdmwxf/Pjf3+IDOy4NPPoHly938hZwc97jfD+ecA6NGwZAhRVceTz89nYUL/8Ctt85k6tTpJCQUNVOVV0BlZrpC7ejRos7f0oVO6f/n99+HV1+FLVvcY82bww03wOWXl7wKatXK/b7Ka9KNlPrUubsfeNUrTT3FHTkCGza4P9A9e1K55ZYBZGdX1KkZx7x5KWGtK5SQ4AqAijrwTlR1mnVK8/v9TJ48mTlz3Dj0pKQkHn/88bD29fl8VbqyqG3WFg3PLGpfdsHfrNkwli2DuXMhLe0IUVGvcOONN3PddSVHg/h8rkmic+fqncOuXbsYMGAAw4YN44UXXsDniyscvZKRAStWvMvf//5bpk9/gS5deh23/8yZP2bXru2MGzeHYHAIqamuM3rHDsq8NWlsrDvf4oVBjx7u8RNlrWsmXbbMBWzxMn/AABf2Q4eWHPkWF+eGRLdsCZ99toKLLjo/jEUVy3//0KifUGGQmXl8sy24St2TT6azYcNM4Od07NiFn/3MnV+oYPH5XN9fVdv/g8FgtZeMrk9NPZuAsuZJ9wE2R/B969xJJ7kP7IYN8PLLlXeEZmcH+Mc/ZjNjxpxKj330KKSkuD+C7t1dR15NKm/IZVVFR/uZPHkq1ro/iGnTpoUd/KGaan0c4x/qECzrAmjv3liefDLUnBIkLm4MmZkfEx+fjd9/R+HrWrRwtfwTCc2uXbvy2Wef0aNHD0xB4sTEuDC01nLrrfexdeunbN36EqNH30tGRlGhkJMDmzZ9xP79X3P66QmFTR0vvzyLd999kZEjf02rVtcUFgah0UmbNrktxLWvlywMEhPdMMmymn9Ca2fl5rqa/5Ej8MEHLvD37Cl6XbduLuxHjHCdrCFRUa6T+qGHkjj99D5cdtlltGzZjREjhlX/F1nwczRv7raOHd1jwaArANLSXJt/6MqjXz/o3v0uNmz4B3Fxqezdu5j774c+feD2293z+fmuUrBnj7tS6ty58sl1+fn5nH/++Zx33nkkJyeTEOHhfZGs8U8BHgZ6WWu/KnisO244593W2lkV7d+Qa/whR49Chw7hj9kePDiNSZPcsLxwNW/u/lBat672aRY6kWadkNLNV9HRLuhOOgluuWUcS5e+VaXj1Zcx/llZLgBLr9UELsxeegmeey5Ibm4UCQkuBHy+hSxcOJOHHlpGu3Zd8flcLblz58h3BKalpfHoo49y9913Ex0dTTAYZPHixVx99dXk5/vYsyeNjz/+nL59LyQ7O4qMDPjd78bz4Yf/4u675zNmjFugbN26FTzxxK8555xr6d//LlJTXVNRamrRhLTSmjd3NfJQs04o6Cu6iGvVCoYPD3LBBekMGNACYyAt7QD33DOO7OyjrFy5mbZtXfifd955fPTRR4wfP57FixdH6DdYJNRUtmuXKzi//343Dz44iZ/97CG2bTuTZ58tGiU1dChMnlxUgID7v27d2v2/lzdYY+XKlVxwwQV07tyZbdu2Fa4gGq761NTTDDeBK4uiCVwzgea4CVwV9vY1huAHCtpYw/kd+wD3lzFoENx4IwwcGP77xMUVtTUaU7QV/3dFz+3alcrtt1/O9u2bjj942OeQULhyaVnNVnv2pHLzzf3JyckqY+/Kjl1yXkBtCQbdH/w335TdDrxpkxu2u2OH+/eoUZCU5Ao6gNzcHJo0aUrz5m65j9pu+w1ZvXo15557Ltdeey0vvvjicc9bC/v3p7F69ed07tyHmBh3b+qnn/4TTz11D1deeSe/+MXfATh69BC/+c0Y+va9kLFj/1pYEIS20iNyivP7ITr6e3y+7cTEdODMM7szahQEAktITr6Ks84awwMPvEqTJtC6dZCBA+PJzs7myJEjtChIzeXLlxMMBhk6dGiVA/JEHTrk/q+Lt4BmZsKLL8LLL7srA78fxo+H//f/jp+Zn5DgCoXWrY/vCF63bh379+9nzJgxVT6vetPUY63NMMYMxy3ZsAC3ZMO7uCUbPLNGcbg3g4+Li2f8ePjXv9z45M8/d3MBbrzRfa1MZmb1F4AL3Tqyon6IikRF+bnkksncdVfFTVWdOiVy//2vlLu8RUVCN7MJ9RnUhu++c0EWuswvLj0dnnoK3njDhWbHjvDjHy/hvfceJjv7GcC1n8TENC2cHFSXw/2WLl1KdHQ0N9xwQ5nPGwPt27fgiiuGlXi8V68kLr/8HOLi2tCpk/u51637jC+/XEtUVDR33umaeADuvfdKhg5tz/jxD9KkyUlER8MrrySzY8en3HHHI3Tp0hNj4Kmn/soLL/yJa66ZwaRJ9wKwfXtHcnOzyco6RL9+rgnTmCg+/PBDunbtWqLpY+TIkRH5HYWjVSu37d/v5lDk5EB29n5++tN2XHYZzJsHS5e6guDNN11F4NJLi35HR4+6zedz/RPNmmXToUMMsbFwejh/6DUkYjX+E9VYavxJSUmV3gwe4OKLr+X3v3+RY8fglVdg8eKiEQ8DBrgCYNCgmg+PcDqfKxcHpNC5c2KJSUOnnlr2TOTjl7eoYEhFMbU12ic93bXjl/VW1rp26Ucfde3eUVFw7bUwaRL84Q/D+fDD9xk69GqSk18mPt615deH9Zjy8/M5ePAgbdpUOKYiLBkZGaxc+Qnp6UHOPHME6emwb99hRo1qRZMmMbz99jGiolydcsqUYaxf/388/PByzjxzBAD/+c98Xn11Dpdc8hOuuCKJ2Fho0yaP5s0zaNOmnkxcCUMwaElK+g3PPvsoc+d+TGLiAMBNpnv8cbeaasipp8IPf+j6LUKfh/37dzF58iAuvzyJW2+dQUyMwe93nehV/czUm6aeE9VYgj+cdnOfz8ezz35RYvRFenpRARAaAt+/vysAzjij5gqA2bOTeOuteVWqfRdxC2ElJCwmK2ssZZVt7dsXFQShr6WHpA4fHl5zmDGG/IrG3Z2gQKBoeGZZp/Pdd/DII26JZHDLdP/qV+4PFWD//jd5/vk3mTz5T/TrdxLdutXvpRBqUm5uLv/97ypSU3dz2WUTCQRCheQyMjMzOOOM82nVqg3WUrgZ4zqC68skxeq48847mTt3LtOn/53hw+8o8bnZvh3eftt1Xof+hps0cTOBL7gA9u17lH/84xeMGHE906c/X7jfgAFVH7Wn4K9nwQ/lj5QJdYTec88CLrroKgCysjJIS/ue9u27A+4D8+9/w6JFRe2Kffu6AmDw4BMPlvAXDDvesGF3cMMNU+nZM5FAwHX2bd3qastbt5bfTNKmjSsAzjvPXQpfeWX459CsWTMyMzNrdMSPtW7kxo4dJYdnhgSD7v/g6afdRK1mzeBnP4PLLoMjR76lZct2GGNIT19Bu3bD6N274S26J9WTlpZGamoqZ5xxBhkZ7jNUegBATg58+KErBD7/vOjx6Gjo2XMNZ53VhhEjetK9u/t7VvA3kuAHV/OfPXs2CxYsID09nbg4dwvH8eOLOkKttTz44I2sWvU6v//9S5x9dlEnT0aGm0Dy8stFnWd9+rgmhrPPDq8AsNb1Axw65LaDB2HmTB+u373q3n+/4v1CHaOhgiC0Lkrx2c0tW0KHDkls3vxElc+jJkb8HDnizqm8mZ7btsHDD7vzB1dbu/NO1zm3detn/PrXo7nqql9w002/JydnBSNHDgt7xUdpfKy1HDtm2LGj5HpIIXv3urkKa9a4gQHFL2Dj411fwDnnuAECVRndV286d6WkxMRE5syZU6JzMi/PTQTavTs0qzGXzMyj5OXl0qZNyZk9zZq5UQJXXgmvveaGD27e7BaB690bJk50HWKhQA+F++HDRd8fOlS6Bn4IiAeqXuPv3r3y1Q+jotzwxR49IHRr0/x8N745JcUVZNu3w+HD04DwxvgXF7qCmjBhQrVG/KSmutE6ZcnKgmefdU1t+flu/Pgvf+muUkL27fuK9PTDbNv2Mf3755GSEv4yv9L4rFmzhilTpvDCCy8wcGAPjh0rWnsoO9u9pmNHiI19lF/84iLath3IJ5/Axx+7uR8HD7qlH9atczOCqxL8VaUafz1QvAAIBCw7d26mR4+iuW+hm7wXl5VVVAAcOb4NSpUAAA0aSURBVBL+e8XEuMvIvLw/cODAH2nf/lL27XsNa6vWxj9r1rucccbwKu1TmrXuQ79gAWzaVP0rD4B+/frx6quvhhX++flu6n3pdWhCVq+Gv/3Njdzw+eCqq+AnPzl+KKYx8M037zFhwgXExDRpsJ8/qRkXX3wxK1as4Ne//jV/+ctfCh93a1X5+P572LHjKGPGtCImJo4XXviKFi3cBBxrXcVs2zbXpDt1atXm5qjG3wBFR7tZuJ07w+7dBr+/b2Fb86efLmfmzB/z29/+L+eeO65wn9hYuO46VzN44w23JozfXzTcrLwtNFv0sccOsnhxJhde2J3XXvOTnR1+8P/kJ/dz/vnDC5e8bdasaIG6rCzXnJSVVfk9YY1x666cfTaMHRtPTk71+hoANm7cyIABAypt9snLc9PuyyosDx1yo3VCa12dcgrcdZdb8THk88/fp127biQm9qR3bxg69MQKP2k8/v3vf/PnP/+Z+++/H4CdO3cyfvx4jDGsXbuWhARITEygb9/+XH75DfTo0ZojR1xFxBh3xX7yydVr468qBX89UrwA2LPHNUMsX/48aWkH2LLlkxLBHxIbC9dc47aKpKUd4ODBw3TufCoAP/3pDC644AoGDryIQYOGhzW2vlevfjz88CNceunwsNYgyc0tKgRKFwrF2zaNgTFjJp7A6CInMzOTCRMmsHJlCl27JhYuaBZa/CwYdLX80u35+fnw1lvw5JOuMz0mxtXwx48vudbKhg3/5be/HUvbtu355JOPadGibbXPVRqfk046iQcffLDw3+3atSMlJQVjTOHqqwDvvPM2HTq4pbODQVfhOHDANfWUNbggEhT89VB0tFuGoVMnmDv3aZ59dhQXXVR0o2drbeH6LOHYvHkN99xzKe3b92Du3DVERUURGxvPwIEXATBkyFjmzUspMbY+NjaeceNuYNq0qQwenFittusmTdwWmsVaXHa2+6Dv2+fCNpyb1ocjJydAcvLswpuUV2bnTjfzduNG9+8hQ2DKlJJrxIT06TOQgQNP56yzzqBduxpYI0MatdjYWFatWkXv3r0LQx8oDH1wFYs2bdxmrbsSrYmF7yqjNv4GIC/PXQHs3g1Hj2Zwzz0/5Prr7+Hss8ewZ08qL788i+XLFxYG9siRE7nmmmmFo4WystK56aY+dOlyGvfe+yItWpS/sltoIaxOnWpv4tHhw+7qZsmSJdWa1VtaXFwCb71V8USvvDzXt/DCC+77li3daJ1hw8oeIdWxo1uALDs7g7i4uDIL3sb6+ZP6T238jVDoCqBzZ7jvvidYv/7/SEs7QDAYYMaMa0sEZWbmMd56ax5vv/00M2a8wrnnjiM2Np45cz6ideuOJQIrOrro3quhW9PVxVoyLVu6LTFxLIMGpfD447NZutRdeTRtGkdubjb5+eFfA2dlVbwiyP79MHNm0UqT48a5hbXKGnv/3nvPkZW1nYcemgG4eQQiDZ2CvwGJioIZM6YSE5ND166DuPXWstfXCRUCv//9eJ59djOdOiXSoUOnehHyFWnWDEaPTmTo0Dns2jWHfftc+/uaNUu4774J5OSEt6xEbGw8gUDJ9e9DVq6EP//ZjZxo3Rp+97vy10LKz9/Fn/50C4FAgB/9aBQXXnjhCfx0IvWHgr+B8fl8TJ/+PyQlJVXaHGJtPu++O5vHHptTK+2GNaVpUzezt3NnN+xyyJCxPP10CtOnX8HOnRsr3T8zczijR7vgj4tzW2ys628I3UHpnHPcHIiylgto0sSN5Dn55K4899xzfPPNNwp9aVQU/A3UwoULK134LRjM45VXFvDMM7W3omVNio11C9Pt2QNRUYk88MCrYS0o16TJRYW3+UtLK7nYWlSUa9aZMKHsyVbt2kGbNkc4+WTXI33dddfV5I8kUi9onmEDFe7Ny6tyk/P6qlMnOOss6NcvkeTkxcTExBEVVbIdJyrKT9OmcUycOJ133pnK8uVubsMrr1hmzUpl6NAkmjZNIBj0MX9+Ao88ksSePamF+zdpAn37Wl5++X4GDerPjtAC+yKNkGr8DVS46/zH14c1gWtATIy7MU27dmPp3j2Ff/6z5NDT0jeAMQaOHt3NlClDOXBgL8FgsEQH+OuvP8Hrrz8OGJo1cwu+/fKXd7Js2TL27t3L2rVr6dGjRx3+xCKRo+BvoCZOnFjpOv9+v7/cG280VO3bw1VXJTJo0Bz27ZtT5vLJIU899Tv27v2qnGdt4deMjGM888w8FiyYz3PPPcf06dO55JJLavrUReoNNfU0UNOmTcNf1rCVYvx+P1OnTq2lM6o9fr9b23/wYNcJ27WrmwATH19ypm1MTCzGhPcRDwQCZGZmMmnSJE499dQInblI/aAafwOVmJjI4sWLy1znv/hyxbV9j9raFFonqLTQUhHvv/9C2Hf3CqmLWzyK1DbV+BuwsWPHkpKSwuTJk0lISMDn85GQkMDkyZNJSUmp9hr1DV1omYiMjKp3bAcCARYsWBCBsxKpP1Tjb+DKWudfnHA7wEtrDCOhRCqiGr80WhMnTqy0H6QsjWUklEh5FPzSaIXTAV5aYxwJJVKagl8arVAHeFxcXNgFQGMdCSVSnIJfGrXSHeCh1UlLL6vs9/uJi4tr9COhREDBLx4Q6gBPS0sjPz+f7du3k5SUpJFQ4lka1SOeo5FQ4nWq8YuIeIyCX0TEYxT8IiIeo+AXEfEYBb+IiMco+EVEPEbBLyLiMQp+ERGPUfCLiHiMgl9ExGMU/CIiHqPgFxHxGAW/iIjHKPhFRDxGwS8i4jEKfhERj1Hwi4h4jIJfRMRjFPwiIh6j4BcR8RgFv4iIxyj4RUQ8RsEvIuIxCn4REY9R8IuIeIyCX0TEYxT8IiIeo+AXEfEYBb+IiMco+EVEPEbBLyLiMQp+ERGPUfCLiHiMgl9ExGMU/CIiHqPgFxHxGAW/iIjHKPhFRDxGwS8i4jEKfhERj1Hwi4h4jIJfRMRjFPwiIh6j4BcR8RgFv4iIxyj4RUQ8RsEvIuIxCn4REY9R8IuIeIyCX0TEYxT8IiIeo+AXEfEYBb+IiMco+EVEPEbBLyLiMQp+ERGPUfCLiHiMgl9ExGMU/CIiHqPgFxHxGAW/iIjHKPhFRDxGwS8i4jEKfhERj1Hwi4h4jIJfRMRjFPwiIh6j4BcR8RgFv4iIxyj4RUQ8JiLBb4zpZYx5xBiTYoxJN8bsM8a8bowZGIn3ExGR8EWqxj8auBiYD1wGJAFtgNXGmDMj9J4iIhKG6Agd90XgMWutDT1gjHkP2An8EpgUofcVEZFKRCT4rbUHyngszRizFegUifcUEZHw1FrnrjGmFdAP+KK23lNERI4XqaaesjwKGOBv5b3AGDMZmFzwz3RjzJdVfI/WwHFXGyK1RJ8/qSvFP3vdKnuxKdYMX/6LjBkJLAvjzf/PWjusjP3vAf4I3GytfSaM41SLMWattXZwpI4vUhF9/qSuVPWzF26NfxXwgzBel1nGCd2GC/3pkQx9EREJT1jBb63NBLZU9eDGmBuAucAsa+0fqrq/iIjUvIh17hpjrgT+F5hnrb0rUu9TypO19D4iZdHnT+pKlT57YbXxV5Ux5iJgKbAJuBPIL/Z0jrX28xp/UxERCUukRvUMB5oCZwArSz33NdA9Qu8rIiKViEiNX0RE6q9GuTqnFomT2mKM6WKMWWyMSTPGHDXG/MsY07Wuz0saN2PMBGPMK8aYr40xWcaYL40xDxpjmoe1f2Os8Rtjfo6bCDYf+Aw4CfgNcDpwgbX20zo8PWkkjDFxwHogB5gOWOABIA4YYK3NqMPTk0bMGLMa2AW8BuwGBgHJuNGX51lr88vfu/EGf2vgYKlF4lrgFol7w1qrReLkhBljfgn8FTjNWru94LEewDbgN9bav9bl+UnjZYxpY639vtRjk3CV3RHW2vcq2r9RNvVYaw/YUiWatTYN0CJxUpN+BKwOhT6AtXYHbkDD5XV2VtLolQ79Ap8UfK004xpl8JdFi8RJBPQFNpbx+CagTy2fi8jQgq+VZpxngp8wFokTqaJWwOEyHj8EtKzlcxEPM8Z0AmYAy621ayt7fYMIfmPMSGOMDWNbUc7+9wDXAz8vflkuItLQGWPicZ28ecBPwtmnNpdlPhFaJE7qo8OUXbMv70pApEYZY2KBN4CewFBr7e5w9msQwa9F4qSe2oRr5y+tD7C5ls9FPMYY4wcWA4OBUdbaDeHu2yCaeqqjjhaJE295HTjHGNMz9IAxpjtwfsFzIhFhjPEBz+OWx7nCWru6Svs30nH8WiROIs4Y0ww3gSuLoglcM4HmuAlc6XV4etKIGWMeB24D/gC8Werp3ZU1+TTW4E8G7ivn6a+ttd1r72ykMStYnmE2MAo3auxdYIq1dmddnpc0bsaYnZR/i8X7rbXJFe7fGINfRETK12jb+EVEpGwKfhERj1Hwi4h4jIJfRMRjFPwiIh6j4BcR8RgFv4iIxyj4RUQ85v8DRattN4ZMCzQAAAAASUVORK5CYII=\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "3516182ae8d74863b716d0aa53742fae", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=0.0), HTML(value='')))" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "train: 60000 loss=-0.6 val_loss=-0.8501\n" + ] + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + }, + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX4AAAENCAYAAAAIbA6TAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjAsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+17YcXAAAgAElEQVR4nO3deXyU1aH/8c+ZEEJCCBgIoIAsQUQRFAVRsbIIKBWLS3C7gD+RYsW2xov71RpRa6sireJye7EFUS9qVKwLtQJFr6AoUg0iCIkigoCCELJnkjm/P85MNrJMgMn2fN+v17ySzDzPzJNk5nvOc7bHWGsRERHv8DX2AYiISMNS8IuIeIyCX0TEYxT8IiIeo+AXEfEYBb+IiMco+KVZMsY8bYy5u7GPQ6Q5UvBLgzPGbDXGjDmc57DW/spae98ROh5rjOkb/D7NGPPckXjeKq/RxxjzpjEmxxizxxjzUIXHEo0xrxlj8owx3xpjrqqy71XB+/OMMUuMMYlHYl/xLgW/NDnGmFaNfQyHqrpjN8a0Bt4FVgBdge5AxcLlCaAY6AL8B/CUMWZAcN8BwH8DU4KP5wNPHqF9xaustbrp1mA3YBEQAAqAXOBWoBdggWuBbcD7wW1fBnYB2cD7wIAKz7MAuD/4/UhgOzAL+AHYCVxTj2OyQF/gfFyI+oPH9nnw8fbAM8Hn3QHcD0QFH/t/wCpgLrA3dExVnn8G8H81vHbb4Gv2q/I3+kPw+98DL1R4LDm4fbvD2bex3we6Ne5NNX5pUNbaKbhwv9BaG2+tfajCwyOAE4Dzgj8vBY4DOgPrgOdreequuIDuhitAnjDGHFXPY/sHLixfDB7bycGHFgAluMJhMDAOmF5h12HA17ha9QPVPPUZwFZjzNJgM89KY8zA4GP9gBJr7eYK238ODAh+PyD4c+gYswiG/WHuKx6m4JemJM1am2etLQCw1v7VWptjrS0C0oCTjTHta9jXD8y21vqttW/jauzHH+4BGWO6AD8HUoPH9gOudn9Fhc2+t9Y+bq0tCR17Fd2D2z8GHAO8BbwebAKKBw5U2T4bV6Mn+Hh2DY8fzr7iYc22LVVapO9C3xhjonC150lAEq55CKATB4cZwF5rbUmFn/NxwXe4egLRwE5jTOg+X8VjrfJ9dQqAD6y1SwGMMY8Ad+HObnKBhCrbJwA5we9rezxwGPuKh6nGL42hpiVhK95/FTARGINrwukVvN8QWVWP7TugCOhkre0QvCVYawfUsk9VGbVssxloZYw5rsJ9JwMbgt9vCP4MuNFBQExwv8PZVzxMwS+NYTfQp45t2uECdy8Qh2t7bwi7gV7GGB+AtXYn8E9gjjEmwRjjM8YkG2NG1OM5nwPOMMaMCZ7JpAJ7gI3W2jzgVWC2MaatMWY4rsBbFNz3eeBCY8zPjDFtgdnAq8EmsEPe91D/ONIyKPilMTwI3GWM2W+MubmGbZ4FvsWNovkS+OhQX8wYc6cxZmmYm78c/LrXGLMu+P1UoHXwOPYB6cDR4b6+tfYrYDLwdHD/icAvrLXFwU1mArG4EUn/C1xvrd0Q3HcD8CtciP+AKxBnVnj6w9lXPMpYG5kLsRhjUoArgSG4URnbcLWT36vGISLSeCIZ/B/hwv513BjrwbiRGZuAs6y1gZr3FhGRSIlk8CdZa3+sct9UYCFwrrV2RUReWEREahWxNv6qoR/0SfBrt0i9roiI1K6hO3dDIyE2NvDriohIUMSaeg56IWO6Af/GrX8ytoZtZuDWNSE2Nva0Hj161Os1AoEAPp8GKknj0PtPGkvF997mzZv3WGuTatu+QYLfGBMPrMRNVz/dWru9rn2GDBli165dW6/XWblyJSNHjjyUQxQ5bHr/SWOp+N4zxnxqrR1S2/YRX7LBGBMLvIGbsDMinNAXEZHIiWjwG2OicZNdhgBjrbXrI/l6IiJSt4gFf3DK+/PAaGCCtfaQZ16KiMiRE8ka/xO4lRUfAPKMMWdUeGy7mnxERBpHJIcgjA9+/S/gwyq36TXtJCIikRWxGr+1tleknltERA6dBh2LiHiMgl9ExGMU/CIiHqPgFxHxGAW/iIjHKPhFRDxGwS8i4jEKfhERj1Hwi4h4jIJfRMRjFPwiIh6j4BcR8RgFv4iIxyj4RUQ8RsEvIuIxCn4REY9R8IuIeIyCX0TEYxT8IiIeo+AXEfEYBb+IiMco+EVEPEbBLyLiMQp+ERGPUfCLiHiMgl9ExGMU/CIiHqPgFxHxGAW/iIjHKPhFRDxGwS8i4jEKfhERj1Hwi4h4jIJfRMRjFPwiIh6j4BcR8RgFv4iIxyj4RUQ8RsEvIuIxCn4REY9R8IuIeIyCX0TEYxT8IiIeo+AXEfEYBb+IiMco+EVEPEbBLyLiMQp+ERGPUfCLiHiMgl9ExGMU/CIiHqPgFxHxGAW/iIjHKPhFRDxGwS8i4jEKfhERj1Hwi4h4jIJfRMRjFPwiIh6j4BcR8RgFv4iIxyj4RUQ8RsEvIuIxCn4REY9R8IuIeIyCX0TEYxT8IiIeo+AXEfEYBb+IiMco+EVEPEbBLyLiMQp+ERGPUfCLiHiMgl9ExGMU/CIiHqPgFxHxGAW/iIjHKPhFRDxGwS8i4jEKfhERj2kxwZ+VlcXcuXNJSEjA5/ORkJDAzJkzycrKauxDExFpUlpE8C9dupRBgwbx1ltvkZOTg7WWnJwc5s+fz6BBg1i6dGljH6KISJPR7IM/KyuLlJQU8vPzKS0trfSY3+8nPz+flJSUQ675Z2VlMXPmTJ1JiEiL0eyDf86cOfj9/lq3KS7288ADc8nJgZKS8J87dCYxf/58nUmISIvRqrEP4HA999xzdQZ/SYmfxYsXMnHivbRrdxQxMT5iYyE2Ftq0oez72Fho3drtU/FMoiq/34/f7yclJYWMjAySk5Mj8auJiEREs6/x5+bmhrVdQUEuF13UiQ0bVuP3w4ED8OKLLzN9+pXMn7+Ef/8bVq+Gd97J5ZlnVpOamkZxce0Fit/vZ+7cuUfi1xARaTDNPvjj4+PD2s7niyI+vgMdOnQuu2/Tpo9ZsWIx27ZtKrsvM3M906cP5+23X6CkpO7gX7Ro0aEduIhII2n2wT958mSio6Pr2Cqao4/+FRMm7GPt2n6sXg1ZWXD22VO5885FnHnmhLItjfFxwgnDCAQCYb1+6IxDncAi0lwYa21jH0O1hgwZYteuXVvndllZWQwaNKjatvhycUAGcHBbfNu20Lmzu3Xp4m6dO8MjjyRQVJRT5+vHx8fz0ksvkZKSUtb2HxIdHU10dDTp6emMHz++zueS5m3lypWMHDmysQ9DavDtt9/SvXt3oqKiGvtQjriK7z1jzKfW2iG1bd/sO3eTk5NJT08nJSWFoqKiSkM6o6KiadUqmquvTqdjx2R++AF273a30Pd5efDNN+5W2WRgPlD3iCF1Aos0bRs2bGD06NGMGzeOBQsWEBUVxc6dO3nttde45JJL6Nq1a2MfYoNq9sEPMH78eDIyMpg1axb/+te/yMnJJTY2nrFjpzBp0k1061Z94FrrOnmrKxC++24WX3+9kNqCPyqqFb16DeDrr9fXenyhTuB58+Ydzq8pIocoOzub/Px8du/ejd/vJyoqipdffpkbb7yRZcuW8eqrrzb2ITaoZt/GH5KcnExqaip79mSzYkUpb72VTWrqvBpDH8AYaN8ejjsOzj4bLr0Urr8e0tLgmWeS+cMf0omJiSMqqmofQjStW8fxwAN/Z/v2LeoElkOifqGGc9ZZZ/H+++/z+uuv06ZNGwD69+/PhRdeyFVXXVW23fr16xkxYgTPPvtsYx1qg2gxwR/i80GPHq6dvkMHiIuDQ23SGzZsPM88k8GECTOIi0vAGB8+XwIwg0Aggz17xlNQEN5w0nCHnYo3aHJg5H3++eesWbOm7OfBgwcTGxtb9vO4ceP4+9//TkpKStl9L7/8Mu+//z6rVq1q0GNtaC2iqaeiqCiorim9tBSKi6GoyH0N3Sr+XFR08Mzebt2SSU2dR2qqa6bx++GJJ+D11+GRR6BVq3hKSsLrBBYBTQ5sCJmZmYwePZqSkhJWr17NgAEDwtpv1qxZ9OvXjxNPPLHsvpKSEnw+Hz5fy6knt7jgr0lUVPns3NoEAtUXCMXFsGeP2yY1Ffr3h0cfBb+/7k7g6OhopkyZcsR+F2newllmRP1Ch6dXr16ce+65FBYW0rdvX8D16RUUuK8+X+WbMe5r+/btmTx5ctnzvPPOO6SmppKWlsbll1/eWL/OEeeZ4A+Xz+eWcQg2A1aSlwcZGa4gOP986N0b7rxzFj/9VHsncKtW0dx4402RO2hpVsJZZsTv97NgwSJ+/et5JCbCUUdBndNVpEyrVq3429+eJzvbsmNHaw4cgJwcd+Zfm6oFwnvvfcumTZuYP/+ZFhX8ET13Mcb0MMakG2OyjTEHjDGvGmOOjeRrRlLbtnDaadCunfv5+ONdJ3CfPum4uQIHfzKjolpxzz3p7N2bzO7dDXq4EXconZNe79AsKQm/vyc/P5ddu+DLL2HVKli7Fr7+2gWYHCw7O4fZsx8lI8PPRx/BJ59Es3lza7Ztg2+/hQ8/hBdfhFdegX/8A957z/1Nv/wStm6FH390f9vQmX5hIZx77jRuueUZ7rjjTTIz3X0tQcQmcBlj4oDPgSLgLsAC9+MScpC1Nq+2/cOdwFVRQ02gKS11b5a9e8t/fvTRLN5+ey6wCMilTZs4evTox+23L6RPn5PK9m3XDvr0cTW45mzp0qX1nrR2KPs0J3W9/3bvdjPGx45NID+/7vSOi0tg7NjJDB8+kaFDx1V6LDa2fOJh27aHe+TNl7Wwf7/7295++/W8/vrTjBlzNb/4xQI2baLs9v339Xve2Fj3d42Ph1Gj3Ii/tm1dk1CXLu5sPyYmMr/ToajvBK5IBv+NwKPA8dbazOB9vYEtwK3W2kdr278pBz+4N1xmJuzYUX7f8uUwZ45rR4yPd0NDx493b5aq2reHXr2aZwEQzmzpuLi4Sp2T4ezTpk0bvvjii2bboVnT+6+gADZvhn373M9z587krbfmU1paW3NPNFFR/Sgt3UCvXpdx3XUv0ru3C/qq76cOHaBbN/deauWRxtuiIti50wV6cTHk58OiRf/itdeup7g4HWtPqrR9TAz06wd9+7rPbn6+u+Xluf9P1a9VJSRASkohJ530OYMHDyMqyoV/t27Vf74bWlOaufsL4KNQ6ANYa78xxqwCJuIKhWbLGDf+v00bV4sDOPdcOOkkmDsX1qyBhx92hcGNN/rJyVnLgAFnlu2fnQ2ff948C4BwOyfvuecerrjiCpKTk3n88cfr3KewsJDU1FTeeOONI3m4jSYQgG3b3C0QAL+/mOzsPYwZM4s336y9XwiiKS19GljJ1q1DueMOgE3ExDzFkCFzOeUUH4MHu/DZv9/djHE11YQEd2aZkOAqIE0hmA5XIOA+Mz/95G55ee5Me+1a+Oc/XVNYUdEoYAM+XxR9+rgBGCec4L726hX+sO5AwIV/fr5rAlq0CNav38tf/3oqsI/p07O4/PIkMjNh1y5XoCQkRO53j4RI1vh3Aa9ba6+rcv+TwCRrbVJt+zf1Gn9FP/4IGze6Nwy4GsXy5fD443DgQD7GDMfn28gzz2TQs2e/ap8jIcG9ORMTG+646yMrK4uHH57D888/R25ueI3MMTExFBUVcfvtt/PEE0+QE2bjdGZmZrOs9Vd8/+3f72r5oROc4uJCfve7S9m0aSPFxe9TULAeSMHn8xMIlBcAUVGu2et3v0une/fxbN7smioyM/18/vnxWPsNMBdIBVyFYejQ71i//nx69uzDzTf/hY4djy57Pp/PhX+oEIiNdXNbmkNHcX6+O0v66Sf39wx1zO7dC++/D+npoSacH4DODBoE48bBiBHudw2Ji3MFYbt27ntXCNd+qzis21pXwNx778/Jy9sOPE+vXgO55RY48URXsB59tGvCbawzrqbU1FMMPGqtvb3K/fcDt1trD/oTGWNmADMAunTpctrixYvr9Zq5ubmNNl4+VEuo+OfMzo7mf/6nL++9dzuwjLZtB1NauoKionzatIll5MgxXHzxZRx9dLeyfXw+d1ralNaR+vDDNdx7bxolJf6DLm9ZG2MMw4YN45xzzuHhhx8m3PfaxIkTSU1NPdTDbTS5ubm0bRtf7XyQHTv8zJp1G3l524DlDB3ajUmTVvDeey/wr3+9S0FBAbGxsYwaNZaJEydVek+EfPbZOpYseZvTT3+IjRuP4bPPcti/vwOQAJwHvEf//qsYMKA7ffrk0rt3Lscck1/je6nqCJbQrTGVlLiALykp/yxZC9u2xbFmTSfWrOnE5s3l1evExLfJzr6ElJRfMXnyRYAL36go97sczufIWncrKnLHlJ2dzRdf9GLhwn7s3BmLMZYJE3YwZco3xMaWYoz77DZG+FfMvlGjRjWv4K+oOdX4QwoL3XDPqs3Yf/lLOv/7v1fjTu0Prt2lpaUzbFjlTs2EBOjZEzp2jPxx16SgAJYty+KyywZRWFjb6qfVa9s2gQ8/zKZLF+jbNyHsGn9CQgLZ2dn1fr3G9u67K4mLG0nFFi1rYdkyd/aXk/MTsbG7uPHGExk3rrwJxudzNXBjymujdX0s1659l7S0SfTtO4K+fZewbNkusrM/Ai6usFWANm18JCe7tu3kZDervXt3SEy0+HwHtwEZ45ov4+LKzw5Ct9DV6Y6k3Nzy5pvs7PLfu7QU1q93TTirV1funG3dGoYMgbFjYcuWu3jhhQe44Yb7uPvuu0hKikzhtXev69MrKHAFwcKFboRQIOA6e//zP+H00922HTq4ZuCG7HRvSm38+4DqWq4Tg4+1OG3awKmnwldfueYfgB07snjttauBg4OztNRPaamftLQU5s/PqLSu0IED7o3frp1rAmrIAiArK4vZs+eQnv5cWKNPqhMVFc2YMVPYu9d9aM49dzJLljwV1r7NbXmLvDzXrFNUVLkJ5bvv9jN79itkZl4LwLBhicyalUhShUbODh1cO3TFESLWljc5FBW5CkVhoevELChwtx49jgcgPt4wfXoBN9xwNLt3X0xmJmzZAmvXLmfz5tsoLHyBDRv6sWFD5WP2+a4iOjqTAQPmceKJw+je3RUK3bq5fqfqOjhbtTq4MAgVEHWFbWFheYdqqFM1P59KhWRBAXz8sQv6jz5yn4GQ9u3hzDNh+HA3pDp0mdRrrrmfq68ew6hRwyPafNWxo2tW274dNm7Mxdr7eOCBX/PXv/Zgyxa47TZXEN1wg9t+7Vro2tX9TePiwnuN0tLSBlsyOpLBvwGobp70icCXEXzdRtWqFQwY4GoomZnw0kt1d4QWFflZvHgus2YdPEszJ6fhCoCSEnjttaVMneqGXNY+6qR20dHRTJpUPmktJWVW2MEfCATw+XzEx8czefJkZs2a1STb/AMB1/n33XcH19CXLSvlwQfPIxD4mOjoQlJTb6g0wssY9/889tiDO1+NcaHWunXNtcaSkmPp338dHTv25qefDPv2uaDp2hWGD7esW3cPJSWfcuWVL3LaaXeTmenmAOzY4Y73wIEPKSr6lnXrEli3LvSsc4DFxMbeQu/el9G9O2W3UKFQUnLwPIJQE0fFgqC0tDzc8/Nrnji1d68L+lWrYN26ygVB9+4u6IcPd23pUVHutZ54YiYnn3wiV155IT179qRnz5F1/auOCJ/P/b/uu+9mFi/+b77/PounnkrnpZdgwQJ4911XYF1zDfziF27U0c6d0KmT+/u1b1/zcwcCAYYPH85ZZ51FWloaCRHuLY5kU08q8AjQz1r7dfC+XrjhnLdba+fUtn9zbOqpKjcXunQJb8w2JDBzZjYXXlj9rOGQdu1cE1CnTod3bH6/+wDn5LjjzMlxNf3p0w+tWaecuwbCb3+bzoQJ4yuF2h13TOCjj96q37M10TH+P/3kavkVJ/Tk5q6kpGQkf/qTmxwEzxETcx9z5rzLgAHl8xbbtHFBdiQ/26FADtWmd+3KZtGix7niituJimpFaWkp772XzsiRk/D5fOzcmc2aNf+mTZuf8f33UWzfDh9/fCl5ea8CC4GpwWdeCdwCXA7cTKdOLpC7dStvNurRw3Vu1lTjLi11Z8C7dpWH4a5dblLV5s2VtqR//1zOOac9w4dD+/Z7uOOOCeTnH2DJki/p3BmSkmDEiLP48MMPufTSS0lPTz9yf8Qwbd++nalTp3LrrQ/Ttu1plJa6M4E//9nV9ME1q/3613DKKeX7JSS4gqNjx4ML+1WrVnH22WfTvXt3tmzZUraCaLiaUuduW9wErgLKJ3DdB7TDTeCq9Xy+JQQ/gM/nC7NT0weUctRRcMUVrsZwpAqAoqLKAZ+b6+6raMeOLO66ayJbt26o/knCkgBMAW4Ckg+qse3alcW0aQMpLq6mHaEOVecFNJbiYteUEmrKq+idd77g6adPYv9+V+u9/noYN66ImArtOJ07u+F/DdUBWFTkCoP/+7+PmDjxTMaNu5x77llMcfHB2+bmZrNly79p3/5EDhzozPbt8O67fyAj4w7atfsNhYWPBWvkP+E6k39GaFS2z+fONkKFQmic/a5d7hoXlWv8PwKZwNG0bt2LIUOga9elvPnmJQwdeh73378EYyA+vpThw+MpKipk//79tA9WmZctW0ZpaSkjRoyod0AeaYWFrml33z531vfBB24Rx9As/RNOgJQUOOec8v95bKzrF+jatfJn/LPPPmP37t2cd9559T6OJhP8wQM4Fjf2bCxggOVAqrV2a137tpTgT0gIr1MzJiaBXr2y+eor93O4BUB8vGsyCBUABQWVAz43l2o/5BWtWbOUtLSUQ67pR0VFM2HCDK67bh5r17o3f9U22g4d4IwzIClpKS+9lBIcIRR+U1J0dDQzZsxotEXLrHXNd998c/CInexseOwxWLHiW2AaAwb8lbvu6knFizpFRbkOv8a60NPs2bO57777WLJkCRdccAGlpeX9BVVvFSsFeXkH2LJlHe3bJ3HssQP44QdYsWIZ8+eP5aijzqBv3w/Zvt0FXSBwMdAVeBDoEHyGNOBTOnT4M9269eHoo2HnzjvYsOEPnH/+bH7727uJjYXMzM/55S9PYfDgn/HGG++TlOSautauXcuxxx5LUlISpolOSNi9ezelpV3IynIFXGGh6/h99dXyz0BSElx8MVxwQeUzvVatCjnmmDYkJrrP8qFWCJpU8B+OlhL8M2fOZP78+XW2848adTl3372YNWvciIFNm9z9oQLgwgtrX1k0Nvbg8cfh2LHj8Jt32rSJO6hzuvZRGVkcddRcsrMXUVSUi7XhXdi+sUb75Oa6Wl115fcHH7hVWvftA5/vcgKBlxgxYhJpaS+VbdOunTvjqWtl2EgKBALs3buXpKRap88A5eEVKgjy8ysXCgUFeXz11SeUlpZy2mnnArB37z5SUhJp1aoN116bQ2xsK7p2hQULRrJp03s88siysm3/8Y+FLFkyj/PPv4aLLppJ+/aQmFhCXFweSUm1NIQ3MdZabr31Vh5//HE+/vhj+vUbRFZW+dlgYaGbXPbKK24SH7i+kHHj3CKPiYnbuO66wUycOJNrrpmNMYboaDcJtLb+gOoo+JtY8IezVIHP52PBgo306OEmd1nLIRcA9RXe8gHVq204akXWuk7QVavcLfQ7Afz857B0aXjNYcYYAoHwCokjobTU1fB37Di48/bAAZg3z3XoAQwaBNdeu4xly9KZMeMPxMe7Gm+PHm5iTxOtrNZbaL5K1duBA8V8+ulqfvxxO2PHli9rvHbtuxQW5nHqqcPp0iWprNM6JsbdEhOb1po39fWb3/yGJ598kscee4wbgkN69u93AztCg9MCAdf2/8orbtRSSELC4xw48FsGD76Khx9+vmzOwaBB9Z/IqeBvYsEPNS9O1qqV6wi9445FnHPOJYCrTWVn/0jXrr2w1r1RFiwoD8sOHcqbgA61APD73QiPzZvhsccSwrqQTHUmTryh1msa12TPHli61E2F9/shKiqB0tLwjqFt27bk5+dHfMTPnj2uLb9qXwi4VR7nzHEjUlq33sUvf9mFSy4x5OevJD5+JODC7YQTmtdSHIcrECg/UwiNTIqJaR6zhA9VdnY2WVlZnHrqqZXut9bV/HfscE2BId9+C2+9BStWhBZ5XAMk0b59H844w3UG/8d/wMCB9TsOBX8TDH5wNf+5c+eyaNGisll2U6ZM4frrbyIqKjnYRmp58MGrWb367/zudy9y+umuk+dwCoCSEldr3bzZNVd89ZX7ubz88eH63evv669t2bowRUWuprNvn3ujhzPBNzPTTXzJyZkJPF3v44jEiJ/CQhf4oZVXK8rNdR13//iH+zk5eR27d48jJeW3XH3178jNdcHfsaNbH6YlB55Uz1p7UF9Ebq4rACp2codW+F2zxi0/8d13lZ9n4UKYOpWwNaUJXFJBcnIy8+bNq7FzsmdP2LKlmPz8A5SUFJOU1L3sMWNg2DA3M7BiAfD007B4cXkBEB3tmlRCIb95s1tArmr3gjGhJoifWL06Hr+//jX+k046id69y39u3doVAD16uIIqJ6d88bCaCoK+feGPf4SbbppFUVF4Y/wrOpKXKbTWDcnburX6Y/34Y7fo3p497nedPh06dfqa++/fx6ZNH1NaWlL2O3XvfvD+0vKtWbOG1NRUXnjhBXpX+HDEx7trdyQnu1FOf/rT4/Tvfw4DB57MwIFw7bWuMvbpp/DZZ65vbNiwyB6ravxNTH6+ZfnyL0lIGFDWrlxQkEtsbPkaRNbCJ5+4AmDjRndf27Yu4KsbwdO9uxtCePzx7utxx8Frrz3A88//nmHDfs4HH7xe7zb+5cuXM3r06LC2rasg+OwzuOmmQz/zAFcQLVmy5JDC/8ABV0hWN2E4Lw+eesqdnoPrpL3tNjceG2DduhUMHHg27du3Jjd3JaNHjzzUX0GauVGjRrFy5UpuueUWHnroobL7QxMSAQ4cOEBiYiJxcXGsXPk10ImcnMp9SCee6EYB1adfSDX+Zi4uznDhhQMoKHC1z7ffXsZ9913Jbbf9jTPPnAC4N8Tpp8PQoQcXAN26HRzy1a1bl529l8LCfA7gI30AAAyQSURBVLp27UV0dHS9gv/ee+8NO/RDx5uQUD6BxVoXtl9/7QqBU06BmJh4iooO/dJSX3zxBYMGDapXs09pqTsj2rmz+rVx1q51tfwffnBnU9OmwXHH/YtWrXoCfQA49dTRHHOMq+m///4hH760AK+99hp//OMfuffeewHYunUrl156KcYYQpXYhIQEBg4cyJQpUzj1VDcGu+Kch8JCN6Qz0oMBFPxNVGys6xz8/e+fJzt7D1999UlZ8IdULAC2b3ft/qHLQlaVnb2HnJx9dO9+HADTps3m7LMv4uSTz2Hw4NGkpdW9TMNJJ53En//853qFfnWMccPVBg924b9tG5x33mTefHN+pSWK6ys/Pz/sZp9AwJ1S799f3fO4ZrTQZQH693e1/JycD5g1azyJiV156qmP6dy5M/36QRgjJMUDOnTowIMPPlj2c5cuXcjIyMAYU7b6KsDbb7/N0UeXL50dE+Pm4vTq5d6PDTHKqZEXYZW6LFz4DM8//zyPP/47unRxoVm1eS7UZl9T6H/55RqmTj2e++67smxZ5djYeE4++RwAhg0bz8KFGUyaNIP4+PJr4d5www1kZmZircVay/r16w879Kvq0weOOQYuu2wWrVsffm+o3+9n7ty5tW5jLWzYUH3or1vnavZvvOFqXtOnu2GbvXpBcvLJ9O17CsOG/ZyePTsxZIhCX2oWGxvL6tWr2bt3b1noA5VCv6oOHRpmvodq/E2cz+fjqquuAtwZQKdOeVxwwQVMmnQHQ4eex44dWbz00hyWLXuurC9gzJjJXHbZrLJhlr17D6BNm7a0a3cUubn76dixY9kVmkIXqIiJSeayy+YBDT8z9rjjwO9PJi0tnbS0FIqK/Fh7aDV/v9/PokWLauxEt9Y1i1UdtVNQAH/5CyxZUn5Mt9/uCqaQuLh2PProcvr3j6NnT9NixuZL5AwdOrSxD6FaCv5m5tlnn+aTT94jN3cPbdv6mTXr8kpNNPn5Obz11nzefvsZZs9+hbPPnkDXrvG88sqH9O17DO3bm0adQVodY1yhVlIynu7dM3j55bm89dYiSkpygTh8vkICgfCnJFdd1rm0tPy2bZtrs6/o88/hoYfc7OKoKDeM7qqrXI3/nXeeZceOTKZNmx1cdrtts7vMnkhVCv5m5qabbqKoqIjBgweTklL9+jqhQuCeey7lyy+/pG/fZODgKzo1JT6fm6peUpJMauo8brxxHk8+6S6vB0uJjk7B7w9vWYnY2HhWry4P+5oUFsL8+W5NFWvdcLvbb3cdtQC7d2/jkUemU1LiZ+zYsVx55c88czFzadnUxt/M+Hw+7rzzTt5444061/+xNsCf/lR7e3dTEhXlpqvHxbmzgJkz4fLLIRAYT0lJBklJJ4X1PIMHj6a4uPbQ37ULZsxw0+iNcbX8p54qD32ALl2O5b/+61nuvvshpkxR6EvLoeBvpp577rk6g7+kpIRFixY10BEdGdHRLvxbt3aBfN11rtnF2mT27FlCdHTdlzMaNOicWh/fvh1uvNHNluzZ0wX+NdeUz7TNzXW9vu3awW23XcHs2bcc9u8l0pQo+JupcC9P2NwuYwhuGeqBA8uvuDR9OkyZ4sK/pCSd6Og4oqIqjwCKioomJiaOyZPvqnTlL2stO3ZkMXfuTC64IIFRo3xMmZLADz/M5Ljjspg3z813CG27cOG9TJs2EPiGU09t3BU1RSJFJ6/NVHx8fFjr/MdXN3urGWjXzrX5r1/vxtxPm+YKggULxuP3Z5CYOJe8vEUUF+cSExPP6adPYdSom+jYMZmMDLfPvn3beeKJEWRnf08gUFphjkAO8DRbtjzFhRca4uLcSKhLLvkN69a9y9693/Pjj2sxpndthyjSbCn4m6nJkyfXuc5/dHQ0U6ZMacCjOrKOOsqF/xdfuCC/+mq3NMWCBcn89FP50NPCQjdr9uCZs/8FfF3Ds9uyr6GRUP/850IWLHiW9u3v4vzzz4/I7yTSFCj4m6lZs2axcOHCOoP/pptuqvHx5iAx0V28fsMGF/4pKe4qRu+8A8uXu+nuPp9rEqr6dceOWPbu9QF1r+FfWuqGxE6bNpWMjIzI/2IijUjB30wlJyeTnp5e7Tr/FZcrbuxr1B4JHTu6has2bHDDLmNj4aKL3K02F1zwAuGEfkWhmb+NdYlHkYagzt1mbPz48WRkZDBjxgwSEsqXWpgxYwYZGRlHbI36pqBTJxf+9ZktW1BQ/47t0MxfkZZMNf5mrq51/luSpCQ3w3fjxupX06wqNjae/Pz6r/jZHEdCidSHavzSrHTuDKedFt565WPGTD5o2Gc4mutIKJFwqcYvzU58vOvwLSx0nbuBQPW3m2+exbvvLqSgIPwF35r7SCiRcCj4pdlq08bdatKtWzKvvFJ9B3hNWsJIKJG6qKlHWrSqHeChC2FXvSB2dHQ0cXFxLWYklEhtFPzS4oU6wLOzswkEAmRmZjJz5swWPxJKpCZq6hHP8dJIKJHqqMYvIuIxCn4REY9R8IuIeIyCX0TEYxT8IiIeo+AXEfEYBb+IiMco+EVEPEbBLyLiMQp+ERGPUfCLiHiMgl9ExGMU/CIiHqPgFxHxGAW/iIjHKPhFRDxGwS8i4jEKfhERj1Hwi4h4jIJfRMRjFPwiIh6j4BcR8RgFv4iIxyj4RUQ8RsEvIuIxCn4REY9R8IuIeIyCX0TEYxT8IiIeo+AXEfEYBb+IiMco+EVEPEbBLyLiMQp+ERGPUfCLiHiMgl9ExGMU/CIiHqPgFxHxGAW/iIjHKPhFRDxGwS8i4jEKfhERj1Hwi4h4jIJfRMRjFPwiIh6j4BcR8RgFv4iIxyj4RUQ8RsEvIuIxCn4REY9R8IuIeIyCX0TEYxT8IiIeo+AXEfEYBb+IiMco+EVEPEbBLyLiMQp+ERGPUfCLiHiMgl9ExGMU/CIiHqPgFxHxGAW/iIjHKPhFRDxGwS8i4jEKfhERj1Hwi4h4jIJfRMRjFPwiIh6j4BcR8ZiIBL8xpp8x5s/GmAxjTK4xZqcx5u/GmJMj8XoiIhK+SNX4xwGjgIXAhcBMIAn4yBhzWoReU0REwtAqQs+7GHjCWmtDdxhjVgBbgRuBqRF6XRERqUNEgt9au6ea+7KNMZuBbpF4TRERCU+Dde4aYxKBk4CNDfWaIiJysEg19VTnccAAf6ppA2PMDGBG8MdcY8xX9XyNTsBBZxsiDUTvP2ksFd97Peva2FRohq95I2PGAO+G8eLvWWtHVrP/HcDvgWuttX8N43kOiTFmrbV2SKSeX6Q2ev9JY6nvey/cGv9q4IQwtsuv5oB+hQv9uyIZ+iIiEp6wgt9amw9squ+TG2OmAE8Cc6y1D9R3fxEROfIi1rlrjLkY+Bsw31p7c6Rep4q/NNDriFRH7z9pLPV674XVxl9fxphzgH8CG4DfAIEKDxdZa/99xF9URETCEqlRPaOBGOBUYFWVx74FekXodUVEpA4RqfGLiEjT1SJX59QicdJQjDE9jDHpxphsY8wBY8yrxphjG/u4pGUzxqQYY14xxnxrjCkwxnxljHnQGNMurP1bYo3fGPNr3ESwhcA6oANwK3AKcLa19tNGPDxpIYwxccDnQBFwF2CB+4E4YJC1Nq8RD09aMGPMR8A24HVgOzAYSMONvjzLWhuoee+WG/ydgL1VFolrj1sk7g1rrRaJk8NmjLkReBQ43lqbGbyvN7AFuNVa+2hjHp+0XMaYJGvtj1Xum4qr7J5rrV1R2/4tsqnHWrvHVinRrLXZgBaJkyPpF8BHodAHsNZ+gxvQMLHRjkpavKqhH/RJ8GudGdcig786WiROImAA8EU1928ATmzgYxEZEfxaZ8Z5JvgJY5E4kXpKBPZVc/9PwFENfCziYcaYbsBsYJm1dm1d2zeL4DfGjDHG2DBuK2vY/w7gKuDXFU/LRUSaO2NMPK6TtwS4Jpx9GnJZ5sOhReKkKdpH9TX7ms4ERI4oY0ws8AbQBxhhrd0ezn7NIvi1SJw0URtw7fxVnQh82cDHIh5jjIkG0oEhwFhr7fpw920WTT2HopEWiRNv+TtwhjGmT+gOY0wvYHjwMZGIMMb4gOdxy+NcZK39qF77t9Bx/FokTiLOGNMWN4GrgPIJXPcB7XATuHIb8fCkBTPGPAX8CngAeLPKw9vravJpqcGfBtxTw8PfWmt7NdzRSEsWXJ5hLjAWN2psOZBqrd3amMclLZsxZis1X2LxXmttWq37t8TgFxGRmrXYNn4REamegl9ExGMU/CIiHqPgFxHxGAW/iIjHKPhFRDxGwS8i4jEKfhERj/n/KiBkk4LujDsAAAAASUVORK5CYII=\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "0c51e7c48f86466d9248ac26484791fb", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=0.0), HTML(value='')))" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "train: 70000 loss=-1.047 val_loss=-0.9568\n" + ] + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "62991eb60f1b4423ac8637e684f85ce1", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=0.0), HTML(value='')))" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "train: 80000 loss=-1.011 val_loss=-1.002\n" + ] + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "8fefec7ef3194eb6819214f91e525b14", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=0.0), HTML(value='')))" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "train: 90000 loss=-1.675 val_loss=-1.183\n" + ] + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n" + ] } ], "source": [ @@ -12138,10 +12514,122 @@ }, { "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [] + "execution_count": 22, + "metadata": { + "ExecuteTime": { + "end_time": "2020-02-16T01:18:20.214816Z", + "start_time": "2020-02-16T01:18:13.335447Z" + } + }, + "outputs": [ + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "7cc920548e074234b4314b187c81904d", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=0.0, max=300.0), HTML(value='')))" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "val loss (n=300) tensor(-1.2687, device='cuda:0', grad_fn=)\n" + ] + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "# Final validation\n", + "val_loss = evaluate(model, dataset_test, 300)\n", + "print('val loss (n=300)', loss)\n", + "writer.add_hparams(hparams, dict(val_loss=loss))\n", + "\n", + "for i in range(3):\n", + " set_seed(i)\n", + " test(model, dataset_test, writer, plot=True, global_step=-1, seed=i, title='test_final')\n", + " \n", + "for i in range(3):\n", + " test(model, dataset_train, writer, plot=True, global_step=-1, seed=i, title='train_final')" + ] }, { "cell_type": "code", diff --git a/docs/1.png b/docs/1.png deleted file mode 100644 index 65d60a6..0000000 Binary files a/docs/1.png and /dev/null differ diff --git a/docs/12.png b/docs/12.png deleted file mode 100644 index a345427..0000000 Binary files a/docs/12.png and /dev/null differ diff --git a/docs/17.png b/docs/17.png deleted file mode 100644 index 24d1260..0000000 Binary files a/docs/17.png and /dev/null differ diff --git a/docs/19.png b/docs/19.png deleted file mode 100644 index 4517990..0000000 Binary files a/docs/19.png and /dev/null differ diff --git a/docs/2.png b/docs/2.png deleted file mode 100644 index b94c175..0000000 Binary files a/docs/2.png and /dev/null differ diff --git a/docs/4.png b/docs/4.png deleted file mode 100644 index f70ad1a..0000000 Binary files a/docs/4.png and /dev/null differ diff --git a/docs/7.png b/docs/7.png deleted file mode 100644 index 30f7306..0000000 Binary files a/docs/7.png and /dev/null differ diff --git a/docs/anp-rnn_2.png b/docs/anp-rnn_2.png deleted file mode 100644 index e73ed9d..0000000 Binary files a/docs/anp-rnn_2.png and /dev/null differ diff --git a/docs/anp-rnn_3.png b/docs/anp-rnn_3.png deleted file mode 100644 index dc3d585..0000000 Binary files a/docs/anp-rnn_3.png and /dev/null differ diff --git a/docs/anp-rnn_4.png b/docs/anp-rnn_4.png index 66a817b..25977eb 100644 Binary files a/docs/anp-rnn_4.png and b/docs/anp-rnn_4.png differ diff --git a/docs/anp_4.png b/docs/anp_4.png new file mode 100644 index 0000000..e2f490f Binary files /dev/null and b/docs/anp_4.png differ diff --git a/docs/lstm_with_context.png b/docs/lstm_with_context.png deleted file mode 100644 index 3441299..0000000 Binary files a/docs/lstm_with_context.png and /dev/null differ diff --git a/docs/np_4.png b/docs/np_4.png new file mode 100644 index 0000000..f1506db Binary files /dev/null and b/docs/np_4.png differ diff --git a/readme.md b/readme.md index 5c3f2b1..cede842 100644 --- a/readme.md +++ b/readme.md @@ -1,29 +1,58 @@ -# Using recurrent attentive neural processes for forecasting power usage +# Neural Processes for sequential data -This repo implements ["Recurrent Attentive Neural Process for Sequential Data"](https://arxiv.org/abs/1910.09323) (ANP-RNN) and tests them on real data. +This repo implements ["Recurrent Attentive Neural Process for Sequential Data"](https://arxiv.org/abs/1910.09323) (ANP-RNN) on a toy regression problem. And also tests it on real smart meter data. + +![](docs/anp-rnn_4.png) + + +- [Neural Processes for sequential data](#neural-processes-for-sequential-data) + - [Models](#models) + - [Results](#results) + - [Example outputs](#example-outputs) + - [Example NP](#example-np) + - [Example ANP outputs (sequential)](#example-anp-outputs-sequential) + - [Example ANP-RNN outputs](#example-anp-rnn-outputs) + - [Replicating DeepMind's tensorflow ANP behaviour](#replicating-deepminds-tensorflow-anp-behaviour) + - [Usage](#usage) + - [Smartmeter Data](#smartmeter-data) + - [Code](#code) + - [See also:](#see-also) + +## Models + +- ANP-RNN ["Recurrent Attentive Neural Process for Sequential Data"](https://arxiv.org/abs/1910.09323) +- ANP: [Attentive Neural Processes](https://arxiv.org/abs/1901.05761) +- NP: [Neural Processes](https://arxiv.org/abs/1807.01622) + + +This implementation has lots of options so you can run it as a ANP-RNN, or ANP or NP. + +I've also made lots of tweaks for flexibility and stability and [replicated the DeepMind ANP results](anp_1d_regression.ipynb) in pytorch. The replication qualitatively seems like a better match than the other pytorch versions of ANP (as of 2019-11-01). You can see other code repositories in the see also section. ![](docs/np_lstm.jpeg) -This implementation has lots of options so you can run it as a [Attentive Neural Process](https://arxiv.org/abs/1901.05761) (ANP), or NP. -I've always made lots of tweaks for flexibility and stability and [replicated the DeepMind ANP results](anp_1d_regression.ipynb) in pytorch. The replication qualitatively seems like a better match than the other pytorch versions of ANP (as of 2019-11-01). +## Results +Results on [*Smartmeter* prediction](./smartmeters-ANP-RNN.ipynb) (lower is better) +|Model|val_np_loss|val_mse_loss| +|--|--|--| +|**ANP-RNN_imp**|**-1.38**|.00423 +|ANP-RNN|-1.27|0.0047| +|ANP|-1.3|0.0072| +|NP|-1.3|0.0040| +|LSTM| | | -## Usage +Results on [toy 1d regression](./anp-rnn_1d_regression.ipynb) (lower is better) -- clone this repository -- see requirements.txt for requirements and version -- Start and run the notebook [smartmeters.ipynb](https://github.com/wassname/attentive-neural-processes/blob/master/smartmeters.ipynb) - -## Data -- Some data is included, you can get more from https://www.kaggle.com/jeanmidev/smart-meters-in-london/version/11 -- Inputs are: - - Weather - - Time features: time of day, day of week, month of year, etc - - Bank holidays - - Position in sequence: days since start of window -- Target is: mean power usage on block 0 +|model|val_loss| +|-----|---------| +| **ANP-RNN(impr)**| **-1.3217**| +| ANP-RNN| -0.62| +| ANP| -0.4228| +| ANP(impr)| -0.3182| +| NP| -1.2687 | ## Example outputs @@ -32,58 +61,28 @@ Here the black dots are input data, the dotted line is the true data. The blue l I chose a difficult example below, it's a window in the test set that deviates from the previous pattern. Given 3 days inputs, it must predict the next day, and the next day has higher power usage than previously. The trained model manages to predict it based on the inputs. -### Example ANP-RNN outputs +### Example NP -![](docs/anp-rnn_2.png) +Here we see underfitting, since the curve doesn't match the data -![](docs/anp-rnn_3.png) +![](docs/np_4.png) -![](docs/anp-rnn_4.png) ### Example ANP outputs (sequential) -![](docs/1.png) +Here we see overfitting, but the uncertainty seems to small, and the fit could be improved -![](docs/4.png) +![](docs/anp_4.png) -![](docs/7.png) +### Example ANP-RNN outputs -![](docs/12.png)**** +This has a better calibrated uncertainty and a better fit -![](docs/19.png) - -### Example LSTM Baseline outputs - -Compare this to a quick LSTM baseline below, which didn't predict this divergance from the pattern. (Bear in mind that I didn't tweak this model as much). The uncertainty and prediction are also less smooth and the log probability is lower. - -An LSTM with an encoder style similar to ANP's: - -![](docs/lstm_with_context.png) - -and a normal LSTM: - -![](docs/lstm_baseline.png) - -## Code - -This is based on the code listed in the next section, with some changes. The most notable ones add stability, others are to make sure it can handle predicting into the future: - -Changes for a predictive use case: -- target points are always in the future, context is in the past -- context and targets are still sampled randomly during training +![](docs/anp-rnn_4.png) -Changes for stability: -- in eval mode, take mean of latent space, and mean of output isntead of sampling -- use log_variance where possible (there is a flag to try without this, and it seems to help) - - and add a minimum bound to std (in log domain) to avoid mode collapse (one path using log_var one not) -- use log_prob loss (not mseloss or BCELoss) -- use pytorch attention (which has dropout) instead of custom attention -- use_deterministic option -- use batchnorm and dropout on channel dimensions -- check and skip nonfinite values because for extreme inputs we can still get nan's -## Replicating tensorflow ANP behaviour +## Replicating DeepMind's tensorflow ANP behaviour I put some work into replicating the behaviour shown in the [original deepmind tensorflow notebook](https://github.com/deepmind/neural-processes/blob/master/attentive_neural_process.ipynb). @@ -99,6 +98,43 @@ And a ANP-RNN It's just a qualitative comparison but we see the same kind of overfitting with uncertainty being tight where lots of data points exist, and wide where they do not. However this repo seems to miss points occasionally. + +## Usage + +- clone this repository +- see requirements.txt for requirements and version +- Start and run the notebook [smartmeters.ipynb](smartmeters-ANP-RNN.ipynb) +- To see a toy 1d regression problem, look at [anp-rnn_1d_regression.ipynb](anp-rnn_1d_regression.ipynb) + +## Smartmeter Data +- Some data is included, you can get more from https://www.kaggle.com/jeanmidev/smart-meters-in-london/version/11 +- Inputs are: + - Weather + - Time features: time of day, day of week, month of year, etc + - Bank holidays + - Position in sequence: days since start of window +- Target is: mean power usage on block 0 + + +## Code + +This is based on the code listed in the next section, with some changes. The most notable ones add stability, others are to make sure it can handle predicting into the future: + +Changes for a sequential/predictive use case: +- target points are always in the future, context is in the past +- context and targets are still sampled randomly during training + +Changes for stability: +- in eval mode, take mean of latent space, and mean of output isntead of sampling +- use log_variance where possible (there is a flag to try without this, and it seems to help) + - and add a minimum bound to std (in log domain) to avoid mode collapse (one path using log_var one not) +- use log_prob loss (not mseloss or BCELoss) +- use pytorch attention (which has dropout and is faster) instead of custom attention +- use_deterministic option, although it seems to do better with this off +- use batchnorm and dropout on channel dimensions +- check and skip nonfinite values because for extreme inputs we can still get nan's. Also gradient clipping +- use pytorch lightning for early stopping, hyperparam opt, and reduce learning rate on plateau + ## See also: A list of projects I used as reference or modified to make this one: diff --git a/smartmeters-ANP-RNN.ipynb b/smartmeters-ANP-RNN.ipynb index cc55cec..42a5926 100644 --- a/smartmeters-ANP-RNN.ipynb +++ b/smartmeters-ANP-RNN.ipynb @@ -28,7 +28,7 @@ "source": [ "Results on *Smartmeter* prediction\n", "\n", - "|Model|test_loss|\n", + "|Model|val_loss|\n", "|--|--| \n", "|ANP-RNN|-1.27|\n", "|ANP-RNN_imp|-1.38|\n", @@ -2180,6 +2180,928 @@ "text": [ "step 13169, {'val_loss': '-1.0961132049560547', 'val/kl': '0.0003247860004194081', 'val/mse': '0.006203540600836277', 'val/std': '0.05735539272427559'}\n" ] + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=0.0, description='Validating', layout=Layout(flex='2'), max=117.0, style=Pr…" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "step 15364, {'val_loss': '-0.8419629335403442', 'val/kl': '0.00034610298462212086', 'val/mse': '0.005819730460643768', 'val/std': '0.05368044599890709'}\n" + ] + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=0.0, description='Validating', layout=Layout(flex='2'), max=117.0, style=Pr…" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "step 17559, {'val_loss': '-1.3719918727874756', 'val/kl': '0.00039435975486412644', 'val/mse': '0.0042431410402059555', 'val/std': '0.059302836656570435'}\n", + "Epoch 7: reducing learning rate of group 0 to 5.8704e-04.\n" + ] + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=0.0, description='Validating', layout=Layout(flex='2'), max=117.0, style=Pr…" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "step 19754, {'val_loss': '-1.5102424621582031', 'val/kl': '0.00029231738881208', 'val/mse': '0.0033912782091647387', 'val/std': '0.046705130487680435'}\n" + ] + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=0.0, description='Validating', layout=Layout(flex='2'), max=117.0, style=Pr…" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "step 21949, {'val_loss': '-1.423159122467041', 'val/kl': '0.00035497310454957187', 'val/mse': '0.0035869088023900986', 'val/std': '0.047349609434604645'}\n", + "\n", + "logger.metrics [{'val_loss': -1.423159122467041, 'val/kl': 0.00035497310454957187, 'val/mse': 0.0035869088023900986, 'val/std': 0.047349609434604645, 'epoch': 9}]\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "[I 2020-02-16 07:44:25,994] Finished trial#29 resulted in value: -1.423159122467041. Current best value is -1.5058155059814453 with parameters: {'attention_dropout': 0.2, 'attention_layers': 1, 'batch_size': 16, 'batchnorm': True, 'context_in_target': True, 'det_enc_cross_attn_type': 'ptmultihead', 'det_enc_self_attn_type': 'multihead', 'dropout': 0, 'grad_clip': 40, 'hidden_dim': 32, 'latent_dim': 64, 'latent_enc_self_attn_type': 'ptmultihead', 'learning_rate': 0.008663362578308754, 'max_nb_epochs': 10, 'min_std': 0.005, 'n_decoder_layers': 8, 'n_det_encoder_layers': 2, 'n_latent_encoder_layers': 2, 'num_context': 96, 'num_extra_target': 96, 'num_heads': 8, 'num_workers': 3, 'use_deterministic_path': False, 'use_lvar': True, 'use_rnn': False, 'use_self_attn': False, 'vis_i': 670, 'x_dim': 17, 'y_dim': 1}.\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "trial 30 params {'attention_dropout': 0.2, 'attention_layers': 2, 'batch_size': 16, 'batchnorm': True, 'context_in_target': True, 'det_enc_cross_attn_type': 'ptmultihead', 'det_enc_self_attn_type': 'multihead', 'dropout': 0, 'grad_clip': 40, 'hidden_dim': 32, 'latent_dim': 64, 'latent_enc_self_attn_type': 'ptmultihead', 'learning_rate': 0.0022419054847050398, 'max_nb_epochs': 10, 'min_std': 0.005, 'n_decoder_layers': 8, 'n_det_encoder_layers': 2, 'n_latent_encoder_layers': 2, 'num_context': 96, 'num_extra_target': 96, 'num_heads': 8, 'num_workers': 3, 'use_deterministic_path': False, 'use_lvar': False, 'use_rnn': False, 'use_self_attn': False, 'vis_i': 670, 'x_dim': 17, 'y_dim': 1}\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "INFO:root:gpu available: True, used: True\n", + "INFO:root:VISIBLE GPUS: 0\n", + "INFO:root:\n", + " Name Type Params\n", + "0 model LatentModel 94 K\n", + "1 model._latent_encoder LatentEncoder 8 K\n", + "2 model._latent_encoder._input_layer Linear 608 \n", + "3 model._latent_encoder._encoder ModuleList 2 K\n", + "4 model._latent_encoder._encoder.0 NPBlockRelu2d 1 K\n", + ".. ... ... ...\n", + "87 model._decoder._decoder.7.act ReLU 0 \n", + "88 model._decoder._decoder.7.dropout Dropout2d 0 \n", + "89 model._decoder._decoder.7.norm BatchNorm2d 192 \n", + "90 model._decoder._mean Linear 97 \n", + "91 model._decoder._std Linear 97 \n", + "\n", + "[92 rows x 3 columns]\n" + ] + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=0.0, description='Validation sanity check', layout=Layout(flex='2'), max=5.…" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "step 0, {'val_loss': '0.7910884022712708', 'val/kl': '0.00010800049494719133', 'val/mse': '0.2537727653980255', 'val/std': '0.654058575630188'}\n", + "\r" + ] + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "8d7f1df292af4d14ba291db5e80b3bec", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=1.0, bar_style='info', layout=Layout(flex='2'), max=1.0), HTML(value='')), …" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=0.0, description='Validating', layout=Layout(flex='2'), max=117.0, style=Pr…" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "step 2194, {'val_loss': '-1.1204354763031006', 'val/kl': '0.0014955311780795455', 'val/mse': '0.00617215083912015', 'val/std': '0.09348662197589874'}\n" + ] + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=0.0, description='Validating', layout=Layout(flex='2'), max=117.0, style=Pr…" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "step 4389, {'val_loss': '-1.3880571126937866', 'val/kl': '0.0010640228865668178', 'val/mse': '0.004546670243144035', 'val/std': '0.06830815225839615'}\n" + ] + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=0.0, description='Validating', layout=Layout(flex='2'), max=117.0, style=Pr…" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "step 6584, {'val_loss': '-1.3446733951568604', 'val/kl': '0.0006483710603788495', 'val/mse': '0.005048147868365049', 'val/std': '0.06271157413721085'}\n" + ] + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=0.0, description='Validating', layout=Layout(flex='2'), max=117.0, style=Pr…" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "step 8779, {'val_loss': '-1.5619605779647827', 'val/kl': '0.0007974352920427918', 'val/mse': '0.003406693460419774', 'val/std': '0.05695287138223648'}\n" + ] + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=0.0, description='Validating', layout=Layout(flex='2'), max=117.0, style=Pr…" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "step 10974, {'val_loss': '-1.567822813987732', 'val/kl': '0.0007052362198010087', 'val/mse': '0.0034684455022215843', 'val/std': '0.0511687695980072'}\n" + ] + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=0.0, description='Validating', layout=Layout(flex='2'), max=117.0, style=Pr…" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "step 13169, {'val_loss': '-1.469671368598938', 'val/kl': '0.0006657325429841876', 'val/mse': '0.0036809651646763086', 'val/std': '0.04665883630514145'}\n" + ] + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=0.0, description='Validating', layout=Layout(flex='2'), max=117.0, style=Pr…" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "step 15364, {'val_loss': '-1.4495060443878174', 'val/kl': '0.0009186150855384767', 'val/mse': '0.003881386946886778', 'val/std': '0.04494263231754303'}\n" + ] + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=0.0, description='Validating', layout=Layout(flex='2'), max=117.0, style=Pr…" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "step 17559, {'val_loss': '-1.4232021570205688', 'val/kl': '0.0007268059998750687', 'val/mse': '0.0037214024923741817', 'val/std': '0.04474468529224396'}\n", + "Epoch 7: reducing learning rate of group 0 to 2.2419e-04.\n" + ] + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=0.0, description='Validating', layout=Layout(flex='2'), max=117.0, style=Pr…" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "step 19754, {'val_loss': '-1.4388623237609863', 'val/kl': '0.0007217807578854263', 'val/mse': '0.003411452053114772', 'val/std': '0.03921413794159889'}\n" + ] + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=0.0, description='Validating', layout=Layout(flex='2'), max=117.0, style=Pr…" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "step 21949, {'val_loss': '-1.4176470041275024', 'val/kl': '0.0006476619746536016', 'val/mse': '0.003419178072363138', 'val/std': '0.03866790235042572'}\n", + "\n", + "logger.metrics [{'val_loss': -1.4176470041275024, 'val/kl': 0.0006476619746536016, 'val/mse': 0.003419178072363138, 'val/std': 0.03866790235042572, 'epoch': 9}]\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "[I 2020-02-16 08:18:44,472] Finished trial#30 resulted in value: -1.4176470041275024. Current best value is -1.5058155059814453 with parameters: {'attention_dropout': 0.2, 'attention_layers': 1, 'batch_size': 16, 'batchnorm': True, 'context_in_target': True, 'det_enc_cross_attn_type': 'ptmultihead', 'det_enc_self_attn_type': 'multihead', 'dropout': 0, 'grad_clip': 40, 'hidden_dim': 32, 'latent_dim': 64, 'latent_enc_self_attn_type': 'ptmultihead', 'learning_rate': 0.008663362578308754, 'max_nb_epochs': 10, 'min_std': 0.005, 'n_decoder_layers': 8, 'n_det_encoder_layers': 2, 'n_latent_encoder_layers': 2, 'num_context': 96, 'num_extra_target': 96, 'num_heads': 8, 'num_workers': 3, 'use_deterministic_path': False, 'use_lvar': True, 'use_rnn': False, 'use_self_attn': False, 'vis_i': 670, 'x_dim': 17, 'y_dim': 1}.\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "trial 31 params {'attention_dropout': 0, 'attention_layers': 3, 'batch_size': 16, 'batchnorm': True, 'context_in_target': True, 'det_enc_cross_attn_type': 'multihead', 'det_enc_self_attn_type': 'multihead', 'dropout': 0, 'grad_clip': 40, 'hidden_dim': 32, 'latent_dim': 64, 'latent_enc_self_attn_type': 'ptmultihead', 'learning_rate': 0.0065892056326513826, 'max_nb_epochs': 10, 'min_std': 0.005, 'n_decoder_layers': 8, 'n_det_encoder_layers': 2, 'n_latent_encoder_layers': 2, 'num_context': 96, 'num_extra_target': 96, 'num_heads': 8, 'num_workers': 3, 'use_deterministic_path': False, 'use_lvar': True, 'use_rnn': False, 'use_self_attn': False, 'vis_i': 670, 'x_dim': 17, 'y_dim': 1}\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "INFO:root:gpu available: True, used: True\n", + "INFO:root:VISIBLE GPUS: 0\n", + "INFO:root:\n", + " Name Type Params\n", + "0 model LatentModel 124 K\n", + "1 model._latent_encoder LatentEncoder 8 K\n", + "2 model._latent_encoder._input_layer Linear 608 \n", + "3 model._latent_encoder._encoder ModuleList 2 K\n", + "4 model._latent_encoder._encoder.0 NPBlockRelu2d 1 K\n", + ".. ... ... ...\n", + "146 model._decoder._decoder.7.act ReLU 0 \n", + "147 model._decoder._decoder.7.dropout Dropout2d 0 \n", + "148 model._decoder._decoder.7.norm BatchNorm2d 192 \n", + "149 model._decoder._mean Linear 97 \n", + "150 model._decoder._std Linear 97 \n", + "\n", + "[151 rows x 3 columns]\n" + ] + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=0.0, description='Validation sanity check', layout=Layout(flex='2'), max=5.…" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "step 0, {'val_loss': '1.5510637760162354', 'val/kl': '0.504607617855072', 'val/mse': '0.33672571182250977', 'val/std': '0.9382523894309998'}\n", + "\r" + ] + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "e68b6e1b39af43209e02eec071204c46", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=1.0, bar_style='info', layout=Layout(flex='2'), max=1.0), HTML(value='')), …" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=0.0, description='Validating', layout=Layout(flex='2'), max=117.0, style=Pr…" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "step 2194, {'val_loss': '-0.9993453621864319', 'val/kl': '0.00492581631988287', 'val/mse': '0.0076696309261024', 'val/std': '0.10437598824501038'}\n" + ] + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=0.0, description='Validating', layout=Layout(flex='2'), max=117.0, style=Pr…" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "step 4389, {'val_loss': '-1.3951125144958496', 'val/kl': '0.0013331789523363113', 'val/mse': '0.004176048096269369', 'val/std': '0.05822036787867546'}\n" + ] + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=0.0, description='Validating', layout=Layout(flex='2'), max=117.0, style=Pr…" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "step 6584, {'val_loss': '-1.4472817182540894', 'val/kl': '0.001394702005200088', 'val/mse': '0.0040388815104961395', 'val/std': '0.05575721710920334'}\n" + ] + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=0.0, description='Validating', layout=Layout(flex='2'), max=117.0, style=Pr…" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "step 8779, {'val_loss': '-1.4240511655807495', 'val/kl': '0.0011677203001454473', 'val/mse': '0.003808689536526799', 'val/std': '0.05027220398187637'}\n" + ] + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=0.0, description='Validating', layout=Layout(flex='2'), max=117.0, style=Pr…" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "step 10974, {'val_loss': '-1.51090669631958', 'val/kl': '0.001719470601528883', 'val/mse': '0.0037067916709929705', 'val/std': '0.05691952630877495'}\n" + ] + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=0.0, description='Validating', layout=Layout(flex='2'), max=117.0, style=Pr…" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "step 13169, {'val_loss': '-1.556814193725586', 'val/kl': '0.0011921778786927462', 'val/mse': '0.003143388545140624', 'val/std': '0.04716252535581589'}\n" + ] + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=0.0, description='Validating', layout=Layout(flex='2'), max=117.0, style=Pr…" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "step 15364, {'val_loss': '-1.5254632234573364', 'val/kl': '0.0011767667019739747', 'val/mse': '0.003485812107101083', 'val/std': '0.048899002373218536'}\n" + ] + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=0.0, description='Validating', layout=Layout(flex='2'), max=117.0, style=Pr…" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "step 17559, {'val_loss': '-1.4824448823928833', 'val/kl': '0.0012150286929681897', 'val/mse': '0.0035667528863996267', 'val/std': '0.0452319011092186'}\n" + ] + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=0.0, description='Validating', layout=Layout(flex='2'), max=117.0, style=Pr…" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "step 19754, {'val_loss': '-1.4969969987869263', 'val/kl': '0.0013163189869374037', 'val/mse': '0.00328625226393342', 'val/std': '0.04428704082965851'}\n", + "Epoch 8: reducing learning rate of group 0 to 6.5892e-04.\n" + ] + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=0.0, description='Validating', layout=Layout(flex='2'), max=117.0, style=Pr…" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "step 21949, {'val_loss': '-1.4131932258605957', 'val/kl': '0.0012195135932415724', 'val/mse': '0.0033923806622624397', 'val/std': '0.0400848314166069'}\n", + "\n", + "logger.metrics [{'val_loss': -1.4131932258605957, 'val/kl': 0.0012195135932415724, 'val/mse': 0.0033923806622624397, 'val/std': 0.0400848314166069, 'epoch': 9}]\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "[I 2020-02-16 08:47:42,528] Finished trial#31 resulted in value: -1.4131932258605957. Current best value is -1.5058155059814453 with parameters: {'attention_dropout': 0.2, 'attention_layers': 1, 'batch_size': 16, 'batchnorm': True, 'context_in_target': True, 'det_enc_cross_attn_type': 'ptmultihead', 'det_enc_self_attn_type': 'multihead', 'dropout': 0, 'grad_clip': 40, 'hidden_dim': 32, 'latent_dim': 64, 'latent_enc_self_attn_type': 'ptmultihead', 'learning_rate': 0.008663362578308754, 'max_nb_epochs': 10, 'min_std': 0.005, 'n_decoder_layers': 8, 'n_det_encoder_layers': 2, 'n_latent_encoder_layers': 2, 'num_context': 96, 'num_extra_target': 96, 'num_heads': 8, 'num_workers': 3, 'use_deterministic_path': False, 'use_lvar': True, 'use_rnn': False, 'use_self_attn': False, 'vis_i': 670, 'x_dim': 17, 'y_dim': 1}.\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "trial 32 params {'attention_dropout': 0.2, 'attention_layers': 2, 'batch_size': 16, 'batchnorm': True, 'context_in_target': True, 'det_enc_cross_attn_type': 'ptmultihead', 'det_enc_self_attn_type': 'multihead', 'dropout': 0, 'grad_clip': 40, 'hidden_dim': 32, 'latent_dim': 64, 'latent_enc_self_attn_type': 'multihead', 'learning_rate': 0.004388551085821375, 'max_nb_epochs': 10, 'min_std': 0.005, 'n_decoder_layers': 8, 'n_det_encoder_layers': 4, 'n_latent_encoder_layers': 8, 'num_context': 96, 'num_extra_target': 96, 'num_heads': 8, 'num_workers': 3, 'use_deterministic_path': False, 'use_lvar': True, 'use_rnn': False, 'use_self_attn': False, 'vis_i': 670, 'x_dim': 17, 'y_dim': 1}\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "INFO:root:gpu available: True, used: True\n", + "INFO:root:VISIBLE GPUS: 0\n", + "INFO:root:\n", + " Name Type Params\n", + "0 model LatentModel 102 K\n", + "1 model._latent_encoder LatentEncoder 14 K\n", + "2 model._latent_encoder._input_layer Linear 608 \n", + "3 model._latent_encoder._encoder ModuleList 8 K\n", + "4 model._latent_encoder._encoder.0 NPBlockRelu2d 1 K\n", + ".. ... ... ...\n", + "127 model._decoder._decoder.7.act ReLU 0 \n", + "128 model._decoder._decoder.7.dropout Dropout2d 0 \n", + "129 model._decoder._decoder.7.norm BatchNorm2d 192 \n", + "130 model._decoder._mean Linear 97 \n", + "131 model._decoder._std Linear 97 \n", + "\n", + "[132 rows x 3 columns]\n" + ] + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=0.0, description='Validation sanity check', layout=Layout(flex='2'), max=5.…" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "step 0, {'val_loss': '1.6433210372924805', 'val/kl': '0.502083420753479', 'val/mse': '0.3020164668560028', 'val/std': '1.1032137870788574'}\n", + "\r" + ] + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "5e06748306f340e3809faaac521f4844", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=1.0, bar_style='info', layout=Layout(flex='2'), max=1.0), HTML(value='')), …" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=0.0, description='Validating', layout=Layout(flex='2'), max=117.0, style=Pr…" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "step 2194, {'val_loss': '-0.5562781095504761', 'val/kl': '0.004524527583271265', 'val/mse': '0.01594124175608158', 'val/std': '0.20407840609550476'}\n" + ] + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=0.0, description='Validating', layout=Layout(flex='2'), max=117.0, style=Pr…" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "step 4389, {'val_loss': '-1.2800618410110474', 'val/kl': '0.001454153680242598', 'val/mse': '0.004913512151688337', 'val/std': '0.07995617389678955'}\n" + ] + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=0.0, description='Validating', layout=Layout(flex='2'), max=117.0, style=Pr…" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "step 6584, {'val_loss': '-1.4909727573394775', 'val/kl': '0.0005825856351293623', 'val/mse': '0.0038258214481174946', 'val/std': '0.05889587849378586'}\n" + ] + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=0.0, description='Validating', layout=Layout(flex='2'), max=117.0, style=Pr…" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "step 8779, {'val_loss': '-1.4236963987350464', 'val/kl': '0.0012530366657301784', 'val/mse': '0.0038965647108852863', 'val/std': '0.051243484020233154'}\n" + ] + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=0.0, description='Validating', layout=Layout(flex='2'), max=117.0, style=Pr…" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "step 10974, {'val_loss': '-1.4802192449569702', 'val/kl': '0.0008309829281643033', 'val/mse': '0.0034029236994683743', 'val/std': '0.05015803128480911'}\n" + ] + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=0.0, description='Validating', layout=Layout(flex='2'), max=117.0, style=Pr…" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "step 13169, {'val_loss': '1679.2408447265625', 'val/kl': '1679.53564453125', 'val/mse': '606.5656127929688', 'val/std': '752.9714965820312'}\n", + "Epoch 5: reducing learning rate of group 0 to 4.3886e-04.\n" + ] + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=0.0, description='Validating', layout=Layout(flex='2'), max=117.0, style=Pr…" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "step 15364, {'val_loss': '-1.515012502670288', 'val/kl': '0.0006384316366165876', 'val/mse': '0.0032132412306964397', 'val/std': '0.04366536810994148'}\n" + ] + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=0.0, description='Validating', layout=Layout(flex='2'), max=117.0, style=Pr…" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "step 17559, {'val_loss': '23501292.0', 'val/kl': '23501292.0', 'val/mse': '8748267.0', 'val/std': '6837.646484375'}\n" + ] + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "HBox(children=(FloatProgress(value=0.0, description='Validating', layout=Layout(flex='2'), max=117.0, style=Pr…" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "step 19754, {'val_loss': '-1.4798544645309448', 'val/kl': '0.0004828128730878234', 'val/mse': '0.0033622710034251213', 'val/std': '0.04285237193107605'}\n" + ] } ], "source": [ diff --git a/src/models/lstm.py b/src/models/lstm.py index 0346744..05c5211 100644 --- a/src/models/lstm.py +++ b/src/models/lstm.py @@ -33,7 +33,7 @@ class SequenceDfDataSet(torch.utils.data.Dataset): self.transforms = transforms def __len__(self): - return len(self.data) - +self.hparams.window_length - self.hparams.target_length + return len(self.data) - self.hparams.window_length - self.hparams.target_length - 1 def iloc(self, idx): k = idx + self.hparams.window_length + self.hparams.target_length @@ -41,8 +41,14 @@ class SequenceDfDataSet(torch.utils.data.Dataset): i = j - self.hparams.window_length assert i >= 0 assert idx <= len(self.data) + x_rows = self.data.iloc[i:j].copy() - y_rows = self.data.iloc[k].to_frame().T.copy() + # x_rows = x_rows.drop(columns=self.label_names) + # Note the NP models do have access to the previous labels for the context, we will allow the LSTM to do the same. Although it will likely just return an autoregressive solution for the first half... + x_rows.loc[x_rows.index[self.hparams.window_length:], self.label_names] = 0 + assert (x_rows.loc[x_rows.index[self.hparams.window_length:], self.label_names]==0).all().all() + + y_rows = self.data[self.label_names].iloc[i+1:j+1].copy() # print(i,j,k) # add seconds since start of window index @@ -56,11 +62,11 @@ class SequenceDfDataSet(torch.utils.data.Dataset): def __getitem__(self, idx): x_rows, y_rows = self.iloc(idx) - y = y_rows[self.label_names].astype(np.float32).values x = x_rows.astype(np.float32).values + y = y_rows[self.label_names].astype(np.float32).values return ( self.transforms(x).squeeze(0).float(), - self.transforms(y[:, None,])[:, 0, 0].float(), + self.transforms(y).squeeze(0).squeeze(-1).float(), ) @@ -79,15 +85,15 @@ class LSTMNet(nn.Module): ) self.hidden_out_size = ( self.hparams.hidden_size - * self.hparams.lstm_layers * (self.hparams.bidirectional + 1) ) self.linear = nn.Linear(self.hidden_out_size, 1) def forward(self, x): outputs, (h_out, _) = self.lstm1(x) - h_out = h_out.permute((1, 0, 2)).reshape((-1, self.hidden_out_size)) - return self.linear(h_out) + # outputs: [B, T, num_direction * H] + y = self.linear(outputs).squeeze(2) + return y class LSTM_PL(pl.LightningModule): @@ -122,9 +128,13 @@ class LSTM_PL(pl.LightningModule): def validation_end(self, outputs): # TODO send an image to tensroboard, like in the lighting_anp.py file - if self.hparams["vis_i"] > 0: + if int(self.hparams["vis_i"]) > 0: loader = self.val_dataloader()[0] - vis_i = min(self.hparams["vis_i"], len(loader.dataset)) + vis_i = min(int(self.hparams["vis_i"]), len(loader.dataset)) + if isinstance(self.hparams["vis_i"], str): + image = plot_from_loader(loader, self, vis_i=vis_i) + plt.show() + else: image = plot_from_loader_to_tensor(loader, self, vis_i=vis_i) self.logger.experiment.add_image( "val/image", image, self.trainer.global_step @@ -227,7 +237,7 @@ class LSTM_PL(pl.LightningModule): return parser -def plot_from_loader(loader, model, vis_i=670, n=100): +def plot_from_loader(loader, model, vis_i=670, n=1): dset_test = loader.dataset label_names = dset_test.label_names y_trues = [] @@ -241,7 +251,7 @@ def plot_from_loader(loader, model, vis_i=670, n=100): model.eval() with torch.no_grad(): y_hat = model.forward(x) - y_hat = y_hat.cpu().numpy() + y_hat = y_hat.cpu().squeeze(0).numpy() dt = y_rows.iloc[0].name diff --git a/src/utils.py b/src/utils.py index ceb4f83..b6f646f 100644 --- a/src/utils.py +++ b/src/utils.py @@ -17,4 +17,4 @@ class ObjectDict(dict): @property def __dict__(self): - return self + return dict(self)