{"cells":[{"cell_type":"code","execution_count":2,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":469,"status":"ok","timestamp":1744179906765,"user":{"displayName":"xin huang","userId":"07597279731200846682"},"user_tz":-480},"id":"BpBJGNbhpGpB","outputId":"9cc21f61-c6c3-46f0-c0fc-362ea439201b"},"outputs":[{"output_type":"stream","name":"stdout","text":["/content/drive/MyDrive/EM_sensitivity_0430\n"]}],"source":["import os\n","os.chdir('/content/drive/MyDrive/EM_sensitivity_0430')\n","!pwd"]},{"cell_type":"code","source":["!pip install TorchSisso -q\n","\n","# CUDA_VISIBLE_DEVICES=0\n","# !nvidia-smi"],"metadata":{"id":"1leFUHGCiDbd","executionInfo":{"status":"ok","timestamp":1744180029823,"user_tz":-480,"elapsed":121487,"user":{"displayName":"xin huang","userId":"07597279731200846682"}},"colab":{"base_uri":"https://localhost:8080/"},"outputId":"e9f02753-15b2-4f38-c00b-fd345f5f695d","collapsed":true},"execution_count":3,"outputs":[{"output_type":"stream","name":"stdout","text":["\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m363.4/363.4 MB\u001b[0m \u001b[31m4.7 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n","\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m13.8/13.8 MB\u001b[0m \u001b[31m85.9 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n","\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m24.6/24.6 MB\u001b[0m \u001b[31m65.3 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n","\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m883.7/883.7 kB\u001b[0m \u001b[31m40.0 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n","\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m664.8/664.8 MB\u001b[0m \u001b[31m2.8 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n","\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m211.5/211.5 MB\u001b[0m \u001b[31m5.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n","\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m56.3/56.3 MB\u001b[0m \u001b[31m11.3 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n","\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m127.9/127.9 MB\u001b[0m \u001b[31m6.6 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n","\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m207.5/207.5 MB\u001b[0m \u001b[31m5.9 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n","\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m21.1/21.1 MB\u001b[0m \u001b[31m53.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n","\u001b[?25h"]}]},{"cell_type":"code","source":["from TorchSisso import SissoModel\n","import numpy as np\n","import pandas as pd\n","import sympy\n","from sympy import symbols"],"metadata":{"id":"kKYzybOhjcC6","executionInfo":{"status":"ok","timestamp":1744180048118,"user_tz":-480,"elapsed":18301,"user":{"displayName":"xin huang","userId":"07597279731200846682"}}},"execution_count":4,"outputs":[]},{"cell_type":"code","source":["df_ = pd.read_csv('/content/drive/MyDrive/EM_sensitivity_0430/all.csv')\n","df_.head()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":206},"id":"2ixf2YmDjeal","executionInfo":{"status":"ok","timestamp":1744180048616,"user_tz":-480,"elapsed":501,"user":{"displayName":"xin huang","userId":"07597279731200846682"}},"outputId":"e2b9d565-2f44-45a0-d5dd-d2818717f75b"},"execution_count":5,"outputs":[{"output_type":"execute_result","data":{"text/plain":[" connectivity content distance filler interface K\n","0 0.0 0.02 0.005 100 10 1.30980\n","1 0.0 0.02 0.005 100 20 1.33020\n","2 0.0 0.02 0.005 100 30 1.35025\n","3 0.0 0.02 0.005 100 40 1.37000\n","4 0.0 0.02 0.005 100 50 1.38940"],"text/html":["\n","
\n","
\n","\n","\n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n","
connectivitycontentdistancefillerinterfaceK
00.00.020.005100101.30980
10.00.020.005100201.33020
20.00.020.005100301.35025
30.00.020.005100401.37000
40.00.020.005100501.38940
\n","
\n","
\n","\n","
\n"," \n","\n"," \n","\n"," \n","
\n","\n","\n","
\n"," \n","\n","\n","\n"," \n","
\n","\n","
\n","
\n"],"application/vnd.google.colaboratory.intrinsic+json":{"type":"dataframe","variable_name":"df_","summary":"{\n \"name\": \"df_\",\n \"rows\": 11000,\n \"fields\": [\n {\n \"column\": \"connectivity\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.3162421409862962,\n \"min\": 0.0,\n \"max\": 1.0,\n \"num_unique_values\": 11,\n \"samples\": [\n 0.5,\n 0.0,\n 0.9\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"content\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.02828555698381884,\n \"min\": 0.02,\n \"max\": 0.1,\n \"num_unique_values\": 5,\n \"samples\": [\n 0.04,\n 0.1,\n 0.06\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"distance\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.014362059452067229,\n \"min\": 0.005,\n \"max\": 0.05,\n \"num_unique_values\": 10,\n \"samples\": [\n 0.045,\n 0.01,\n 0.03\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"filler\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 111,\n \"min\": 100,\n \"max\": 400,\n \"num_unique_values\": 4,\n \"samples\": [\n 200,\n 400,\n 100\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"interface\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 14,\n \"min\": 10,\n \"max\": 50,\n \"num_unique_values\": 5,\n \"samples\": [\n 20,\n 50,\n 30\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"K\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 6.056763153282919,\n \"min\": 0.91085,\n \"max\": 33.1315,\n \"num_unique_values\": 9404,\n \"samples\": [\n 7.7015,\n 1.64,\n 16.3315\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}"}},"metadata":{},"execution_count":5}]},{"cell_type":"code","source":["df_.shape"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"3sxL-qG8-6lK","executionInfo":{"status":"ok","timestamp":1744092697344,"user_tz":-480,"elapsed":38,"user":{"displayName":"xin huang","userId":"07597279731200846682"}},"outputId":"ef344414-03fa-4e1e-d371-0852752bfaa3"},"execution_count":5,"outputs":[{"output_type":"execute_result","data":{"text/plain":["(11000, 6)"]},"metadata":{},"execution_count":5}]},{"cell_type":"code","source":["# 极低贯通率\n","X = df_.iloc[6001:8001, :-1].values # All rows, all columns except the last one\n","y = df_.iloc[6001:8001, -1].values # All rows, only the last column (target)\n","# Assign column names for the features\n","cols = [\"connectivity\",\"content\",\"distanc\",\"filler\",\"interface\"]\n","\n","# Insert the target variable into the dataframe at the beginning\n","df = pd.DataFrame(X, columns=cols)\n","df.insert(0, 'Target', y)\n","df.tail()\n","operators = ['+','-','*','/','exp','ln','sin','cos','pow(2)','pow(0.5)','sqrt','tanh']\n","rmse,equation,r2,_ = SissoModel(df,operators,n_expansion=3,n_term=3,k=20,use_gpu=False).fit()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"8GNJqSivf-vD","executionInfo":{"status":"ok","timestamp":1744167828608,"user_tz":-480,"elapsed":18997,"user":{"displayName":"xin huang","userId":"07597279731200846682"}},"outputId":"8242c61b-cf54-43ec-8596-415c9ddc3d98"},"execution_count":5,"outputs":[{"output_type":"stream","name":"stdout","text":["************************************************ Starting Feature Space Construction in cpu ************************************************ \n","\n","\n","\n","************************************************ Starting 1 level of feature expansion...************************************************ \n","\n","************************************************ 1 Feature Expansion Completed with feature space size::: 90 ************************************************ \n","\n","************************************************ Time taken to create the space is::: 0.22483372688293457 Seconds...************************************************ \n","\n","************************************************ Starting 2 level of feature expansion...************************************************ \n","\n","************************************************ 2 Feature Expansion Completed with feature space size::: 20703 ************************************************ \n","\n","************************************************ Time taken to create the space is::: 6.3963634967803955 Seconds...************************************************ \n","\n","\n","\n","************************************************ Starting sparse model building in cpu ************************************************ \n","\n","\n","\n","Equation: 0.906294*((content*filler)*(sin(connectivity)))+0.7833687088778927\n","\n","\n","RMSE: 0.2714812627874027\n","R2:: 0.9976974384478641\n","\n","\n","************************************************ Time taken to generate one dimensional equation: 0.09752154350280762 seconds ************************************************ \n","\n","\n","\n","1 term in the equation is 0.8897034027*((content*filler)*(sin(connectivity)))\n","\n","\n","2 term in the equation is 0.0602396186*((content*interface)+(content/distanc))\n","\n","\n","Equation: + 0.8897034027*((content*filler)*(sin(connectivity))) + 0.0602396186*((content*interface)+(content/distanc)) +0.613627263035654\n","\n","\n","RMSE: 0.11934555443950529\n","\n","\n","R2:: 0.9995550163847332\n","************************************************ Time taken for 2 dimension is: 0.2322230339050293 seconds ************************************************ \n","\n","1 term in the equation is 0.8872825927*((content*filler)*(sin(connectivity)))\n","\n","\n","2 term in the equation is 0.0524394267*((content*interface)+(content/distanc))\n","\n","\n","3 term in the equation is -0.0226772593*((distanc-interface)*(content)**0.5)\n","\n","\n","Equation: + 0.8872825927*((content*filler)*(sin(connectivity))) + 0.0524394267*((content*interface)+(content/distanc)) -0.0226772593*((distanc-interface)*(content)**0.5) +0.515884513525929\n","\n","\n","RMSE: 0.09016813849733112\n","\n","\n","R2:: 0.9997459976497983\n","************************************************ Time taken for 3 dimension is: 9.693444967269897 seconds ************************************************ \n","\n"]}]},{"cell_type":"code","source":["# 极低贯通率\n","X = df_.iloc[3001:5001, :-1].values # All rows, all columns except the last one\n","y = df_.iloc[3001:5001, -1].values # All rows, only the last column (target)\n","# Assign column names for the features\n","cols = [\"connectivity\",\"content\",\"distanc\",\"filler\",\"interface\"]\n","\n","# Insert the target variable into the dataframe at the beginning\n","df = pd.DataFrame(X, columns=cols)\n","df.insert(0, 'Target', y)\n","df.tail()\n","operators = ['+','-','*','/','exp','ln','sin','cos','pow(2)','pow(0.5)','sqrt','tanh']\n","rmse,equation,r2,_ = SissoModel(df,operators,n_expansion=3,n_term=3,k=20,use_gpu=False).fit()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"ql0wF5i6iKLI","executionInfo":{"status":"ok","timestamp":1744168436886,"user_tz":-480,"elapsed":18475,"user":{"displayName":"xin huang","userId":"07597279731200846682"}},"outputId":"a34c67de-567d-40e9-fbd0-32cc94ebbdf7"},"execution_count":7,"outputs":[{"output_type":"stream","name":"stdout","text":["************************************************ Starting Feature Space Construction in cpu ************************************************ \n","\n","\n","\n","************************************************ Starting 1 level of feature expansion...************************************************ \n","\n","************************************************ 1 Feature Expansion Completed with feature space size::: 90 ************************************************ \n","\n","************************************************ Time taken to create the space is::: 0.013163566589355469 Seconds...************************************************ \n","\n","************************************************ Starting 2 level of feature expansion...************************************************ \n","\n","************************************************ 2 Feature Expansion Completed with feature space size::: 20702 ************************************************ \n","\n","************************************************ Time taken to create the space is::: 4.490592002868652 Seconds...************************************************ \n","\n","\n","\n","************************************************ Starting sparse model building in cpu ************************************************ \n","\n","\n","\n","Equation: 0.963407*((connectivity-distanc)*(content*filler))+1.2141098314124008\n","\n","\n","RMSE: 0.3635365384550434\n","R2:: 0.9880890850626194\n","\n","\n","************************************************ Time taken to generate one dimensional equation: 0.029787540435791016 seconds ************************************************ \n","\n","\n","\n","1 term in the equation is 0.9051346625*((content*filler)*(tanh(connectivity)))\n","\n","\n","2 term in the equation is 0.0367523346*((content+distanc)/(connectivity*distanc))\n","\n","\n","Equation: + 0.9051346625*((content*filler)*(tanh(connectivity))) + 0.0367523346*((content+distanc)/(connectivity*distanc)) +0.833931331721931\n","\n","\n","RMSE: 0.17116354930891345\n","\n","\n","R2:: 0.9973595888679605\n","************************************************ Time taken for 2 dimension is: 0.17021703720092773 seconds ************************************************ \n","\n","1 term in the equation is 0.8305258934*((connectivity*filler)*(tanh(content)))\n","\n","\n","2 term in the equation is 0.5150716822*((content/connectivity)/(distanc)**0.5)\n","\n","\n","3 term in the equation is 0.0000037544*((filler*interface)/(distanc)**0.5)\n","\n","\n","Equation: + 0.8305258934*((connectivity*filler)*(tanh(content))) + 0.5150716822*((content/connectivity)/(distanc)**0.5) + 0.0000037544*((filler*interface)/(distanc)**0.5) +0.683162028777539\n","\n","\n","RMSE: 0.12505843668914202\n","\n","\n","R2:: 0.9985904678740326\n","************************************************ Time taken for 3 dimension is: 11.925750494003296 seconds ************************************************ \n","\n"]}]},{"cell_type":"code","source":["# 极低贯通率\n","X = df_.iloc[1001:3001, :-1].values # All rows, all columns except the last one\n","y = df_.iloc[1001:3001, -1].values # All rows, only the last column (target)\n","# Assign column names for the features\n","cols = [\"connectivity\",\"content\",\"distanc\",\"filler\",\"interface\"]\n","\n","# Insert the target variable into the dataframe at the beginning\n","df = pd.DataFrame(X, columns=cols)\n","df.insert(0, 'Target', y)\n","df.tail()\n","operators = ['+','-','*','/','exp','ln','sin','cos','pow(2)','pow(0.5)','sqrt','tanh']\n","rmse,equation,r2,_ = SissoModel(df,operators,n_expansion=3,n_term=3,k=20,use_gpu=False).fit()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"4H707F3-k-v_","executionInfo":{"status":"ok","timestamp":1744169189946,"user_tz":-480,"elapsed":36024,"user":{"displayName":"xin huang","userId":"07597279731200846682"}},"outputId":"0c9669f3-6b8c-4ad4-c4a1-ff86d6d27023"},"execution_count":8,"outputs":[{"output_type":"stream","name":"stdout","text":["************************************************ Starting Feature Space Construction in cpu ************************************************ \n","\n","\n","\n","************************************************ Starting 1 level of feature expansion...************************************************ \n","\n","************************************************ 1 Feature Expansion Completed with feature space size::: 90 ************************************************ \n","\n","************************************************ Time taken to create the space is::: 0.02023768424987793 Seconds...************************************************ \n","\n","************************************************ Starting 2 level of feature expansion...************************************************ \n","\n","************************************************ 2 Feature Expansion Completed with feature space size::: 20612 ************************************************ \n","\n","************************************************ Time taken to create the space is::: 9.204334497451782 Seconds...************************************************ \n","\n","\n","\n","************************************************ Starting sparse model building in cpu ************************************************ \n","\n","\n","\n","Equation: 1.123894*((connectivity-distanc)*(content*filler))+1.691975243292843\n","\n","\n","RMSE: 0.5608749366903957\n","R2:: 0.90834381451713\n","\n","\n","************************************************ Time taken to generate one dimensional equation: 0.11502838134765625 seconds ************************************************ \n","\n","\n","\n","1 term in the equation is 1.0705701357*((connectivity-distanc)*(content*filler))\n","\n","\n","2 term in the equation is 0.2089786294*((content/connectivity)/(distanc)**0.5)\n","\n","\n","Equation: + 1.0705701357*((connectivity-distanc)*(content*filler)) + 0.2089786294*((content/connectivity)/(distanc)**0.5) +1.12240728401573\n","\n","\n","RMSE: 0.25271482369872417\n","\n","\n","R2:: 0.9813923663597659\n","************************************************ Time taken for 2 dimension is: 0.33399033546447754 seconds ************************************************ \n","\n","1 term in the equation is 1.0614843059*((connectivity-distanc)*(content*filler))\n","\n","\n","2 term in the equation is 0.1928133200*((content/connectivity)/(distanc)**0.5)\n","\n","\n","3 term in the equation is -0.0001750580*((sin(filler))/(connectivity/filler))\n","\n","\n","Equation: + 1.0614843059*((connectivity-distanc)*(content*filler)) + 0.1928133200*((content/connectivity)/(distanc)**0.5) -0.0001750580*((sin(filler))/(connectivity/filler)) +0.90665627628275\n","\n","\n","RMSE: 0.1792275800587965\n","\n","\n","R2:: 0.9906407856501194\n","************************************************ Time taken for 3 dimension is: 23.578584909439087 seconds ************************************************ \n","\n"]}]},{"cell_type":"code","source":[],"metadata":{"id":"GvmKW2ELiLCv"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["# 极低贯通率0\n","X = df_.iloc[0:1000, :-1].values # All rows, all columns except the last one\n","y = df_.iloc[0:1000, -1].values # All rows, only the last column (target)\n","# Assign column names for the features\n","cols = [\"connectivity\",\"content\",\"distanc\",\"filler\",\"interface\"]\n","\n","# Insert the target variable into the dataframe at the beginning\n","df = pd.DataFrame(X, columns=cols)\n","df.insert(0, 'Target', y)\n","df.tail()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":206},"id":"wVY2vzqeCqtg","executionInfo":{"status":"ok","timestamp":1740117578991,"user_tz":-480,"elapsed":57,"user":{"displayName":"xin huang","userId":"07597279731200846682"}},"outputId":"ace3318b-bf2e-4e64-885f-dc3fe7d88e84"},"execution_count":null,"outputs":[{"output_type":"execute_result","data":{"text/plain":[" Target connectivity content distanc filler interface\n","995 1.70745 0.0 0.1 0.05 400.0 10.0\n","996 1.70870 0.0 0.1 0.05 400.0 20.0\n","997 1.70945 0.0 0.1 0.05 400.0 30.0\n","998 1.71010 0.0 0.1 0.05 400.0 40.0\n","999 1.71065 0.0 0.1 0.05 400.0 50.0"],"text/html":["\n","
\n","
\n","\n","\n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n","
Targetconnectivitycontentdistancfillerinterface
9951.707450.00.10.05400.010.0
9961.708700.00.10.05400.020.0
9971.709450.00.10.05400.030.0
9981.710100.00.10.05400.040.0
9991.710650.00.10.05400.050.0
\n","
\n","
\n","\n","
\n"," \n","\n"," \n","\n"," \n","
\n","\n","\n","
\n"," \n","\n","\n","\n"," \n","
\n","\n","
\n","
\n"],"application/vnd.google.colaboratory.intrinsic+json":{"type":"dataframe","summary":"{\n \"name\": \"df\",\n \"rows\": 5,\n \"fields\": [\n {\n \"column\": \"Target\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.0012512993247021373,\n \"min\": 1.70745,\n \"max\": 1.71065,\n \"num_unique_values\": 5,\n \"samples\": [\n 1.7087,\n 1.71065,\n 1.70945\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"connectivity\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.0,\n \"min\": 0.0,\n \"max\": 0.0,\n \"num_unique_values\": 1,\n \"samples\": [\n 0.0\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"content\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.0,\n \"min\": 0.1,\n \"max\": 0.1,\n \"num_unique_values\": 1,\n \"samples\": [\n 0.1\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"distanc\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.0,\n \"min\": 0.05,\n \"max\": 0.05,\n \"num_unique_values\": 1,\n \"samples\": [\n 0.05\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"filler\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.0,\n \"min\": 400.0,\n \"max\": 400.0,\n \"num_unique_values\": 1,\n \"samples\": [\n 400.0\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"interface\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 15.811388300841896,\n \"min\": 10.0,\n \"max\": 50.0,\n \"num_unique_values\": 5,\n \"samples\": [\n 20.0\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}"}},"metadata":{},"execution_count":15}]},{"cell_type":"code","source":["operators = ['+','-','*','/','exp','ln','sin','cos','pow(2)','pow(0.5)','sqrt','tanh']\n","rmse,equation,r2,_ = SissoModel(df,operators,n_expansion=3,n_term=3,k=20,use_gpu=False).fit()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"SbTlQG2eE4ee","executionInfo":{"status":"ok","timestamp":1740117615102,"user_tz":-480,"elapsed":7001,"user":{"displayName":"xin huang","userId":"07597279731200846682"}},"outputId":"ebf620d6-3a7c-4ca6-d27b-dd7ab2e2e409"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["************************************************ Starting Feature Space Construction in cpu ************************************************ \n","\n","\n","\n","************************************************ Starting 1 level of feature expansion...************************************************ \n","\n","************************************************ 1 Feature Expansion Completed with feature space size::: 62 ************************************************ \n","\n","************************************************ Time taken to create the space is::: 0.22970890998840332 Seconds...************************************************ \n","\n","************************************************ Starting 2 level of feature expansion...************************************************ \n","\n","************************************************ 2 Feature Expansion Completed with feature space size::: 9916 ************************************************ \n","\n","************************************************ Time taken to create the space is::: 1.8996198177337646 Seconds...************************************************ \n","\n","\n","\n","************************************************ Starting sparse model building in cpu ************************************************ \n","\n","\n","\n","Equation: 0.012406*((content/distanc)*(filler)**0.5)+1.4366692797211282\n","\n","\n","RMSE: 0.23267405183337445\n","R2:: 0.9173156045597003\n","\n","\n","************************************************ Time taken to generate one dimensional equation: 0.0052449703216552734 seconds ************************************************ \n","\n","\n","\n","1 term in the equation is 0.0108541957*((content/distanc)*(filler)**0.5)\n","\n","\n","2 term in the equation is -4.4509753886*((distanc/content)/(filler)**0.5)\n","\n","\n","Equation: + 0.0108541957*((content/distanc)*(filler)**0.5) -4.4509753886*((distanc/content)/(filler)**0.5) +1.715038624261009\n","\n","\n","RMSE: 0.1694360687512478\n","\n","\n","R2:: 0.9561530506702198\n","************************************************ Time taken for 2 dimension is: 0.2964358329772949 seconds ************************************************ \n","\n","1 term in the equation is 0.0100335292*((content/distanc)*(filler)**0.5)\n","\n","\n","2 term in the equation is -9.9795003442*((distanc)**0.5/(content*filler))\n","\n","\n","3 term in the equation is 467.0608582693*((distanc)**2/(ln(content)))\n","\n","\n","Equation: + 0.0100335292*((content/distanc)*(filler)**0.5) -9.9795003442*((distanc)**0.5/(content*filler)) + 467.0608582693*((distanc)**2/(ln(content))) +1.910875832961422\n","\n","\n","RMSE: 0.12325503472038966\n","\n","\n","R2:: 0.9767973821892026\n","************************************************ Time taken for 3 dimension is: 3.9036898612976074 seconds ************************************************ \n","\n"]}]},{"cell_type":"code","source":["# 极低贯通率\n","X = df_.iloc[1001:2001, :-1].values # All rows, all columns except the last one\n","y = df_.iloc[1001:2001, -1].values # All rows, only the last column (target)\n","# Assign column names for the features\n","cols = [\"connectivity\",\"content\",\"distanc\",\"filler\",\"interface\"]\n","\n","# Insert the target variable into the dataframe at the beginning\n","df = pd.DataFrame(X, columns=cols)\n","df.insert(0, 'Target', y)\n","df.tail()\n","operators = ['+','-','*','/','exp','ln','sin','cos','pow(2)','pow(0.5)','sqrt','tanh']\n","rmse,equation,r2,_ = SissoModel(df,operators,n_expansion=3,n_term=3,k=20,use_gpu=False).fit()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"eT6d_TYw_Llx","executionInfo":{"status":"ok","timestamp":1740117661037,"user_tz":-480,"elapsed":7993,"user":{"displayName":"xin huang","userId":"07597279731200846682"}},"outputId":"73944a1b-4cac-4871-fab2-e09acf72dfea"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["************************************************ Starting Feature Space Construction in cpu ************************************************ \n","\n","\n","\n","************************************************ Starting 1 level of feature expansion...************************************************ \n","\n","************************************************ 1 Feature Expansion Completed with feature space size::: 90 ************************************************ \n","\n","************************************************ Time taken to create the space is::: 0.0068209171295166016 Seconds...************************************************ \n","\n","************************************************ Starting 2 level of feature expansion...************************************************ \n","\n","************************************************ 2 Feature Expansion Completed with feature space size::: 20612 ************************************************ \n","\n","************************************************ Time taken to create the space is::: 2.8835675716400146 Seconds...************************************************ \n","\n","\n","\n","************************************************ Starting sparse model building in cpu ************************************************ \n","\n","\n","\n","Equation: 0.11186*((content*filler)+(content/distanc))+1.153493160633769\n","\n","\n","RMSE: 0.28984050294813835\n","R2:: 0.9552618196323766\n","\n","\n","************************************************ Time taken to generate one dimensional equation: 0.09885334968566895 seconds ************************************************ \n","\n","\n","\n","1 term in the equation is 0.1058142758*((content*filler)+(content/distanc))\n","\n","\n","2 term in the equation is -0.0645712784*((distanc*filler)-(filler)**0.5)\n","\n","\n","Equation: + 0.1058142758*((content*filler)+(content/distanc)) -0.0645712784*((distanc*filler)-(filler)**0.5) +0.717181397663854\n","\n","\n","RMSE: 0.15000784118366664\n","\n","\n","R2:: 0.9880163783519437\n","************************************************ Time taken for 2 dimension is: 0.29915523529052734 seconds ************************************************ \n","\n","1 term in the equation is 0.0979009296*((content*filler)+(content/distanc))\n","\n","\n","2 term in the equation is -0.0718998584*((distanc*filler)-(filler)**0.5)\n","\n","\n","3 term in the equation is 0.3221175635*((ln(distanc))/(content*filler))\n","\n","\n","Equation: + 0.0979009296*((content*filler)+(content/distanc)) -0.0718998584*((distanc*filler)-(filler)**0.5) + 0.3221175635*((ln(distanc))/(content*filler)) +0.946574253967982\n","\n","\n","RMSE: 0.10842423358456328\n","\n","\n","R2:: 0.9937394442293562\n","************************************************ Time taken for 3 dimension is: 3.808899164199829 seconds ************************************************ \n","\n"]}]},{"cell_type":"code","source":["# 极低贯通率\n","X = df_.iloc[2001:3001, :-1].values # All rows, all columns except the last one\n","y = df_.iloc[2001:3001, -1].values # All rows, only the last column (target)\n","# Assign column names for the features\n","cols = [\"connectivity\",\"content\",\"distanc\",\"filler\",\"interface\"]\n","\n","# Insert the target variable into the dataframe at the beginning\n","df = pd.DataFrame(X, columns=cols)\n","df.insert(0, 'Target', y)\n","df.tail()\n","operators = ['+','-','*','/','exp','ln','sin','cos','pow(2)','pow(0.5)','sqrt','tanh']\n","rmse,equation,r2,_ = SissoModel(df,operators,n_expansion=3,n_term=3,k=20,use_gpu=False).fit()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"1hUhQo7u_Lxf","executionInfo":{"status":"ok","timestamp":1740117732110,"user_tz":-480,"elapsed":8096,"user":{"displayName":"xin huang","userId":"07597279731200846682"}},"outputId":"df7bc359-24a3-4a39-b884-fa5f656f99f2"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["************************************************ Starting Feature Space Construction in cpu ************************************************ \n","\n","\n","\n","************************************************ Starting 1 level of feature expansion...************************************************ \n","\n","************************************************ 1 Feature Expansion Completed with feature space size::: 90 ************************************************ \n","\n","************************************************ Time taken to create the space is::: 0.0076580047607421875 Seconds...************************************************ \n","\n","************************************************ Starting 2 level of feature expansion...************************************************ \n","\n","************************************************ 2 Feature Expansion Completed with feature space size::: 20702 ************************************************ \n","\n","************************************************ Time taken to create the space is::: 3.1171915531158447 Seconds...************************************************ \n","\n","\n","\n","************************************************ Starting sparse model building in cpu ************************************************ \n","\n","\n","\n","Equation: 0.174293*((content*filler)+(content/distanc))+1.0631506330750597\n","\n","\n","RMSE: 0.2530362236567606\n","R2:: 0.9855104526550739\n","\n","\n","************************************************ Time taken to generate one dimensional equation: 0.09811902046203613 seconds ************************************************ \n","\n","\n","\n","1 term in the equation is 0.1781831240*((content*filler)+(content/distanc))\n","\n","\n","2 term in the equation is -0.0417260516*((content/distanc)/(distanc*filler))\n","\n","\n","Equation: + 0.1781831240*((content*filler)+(content/distanc)) -0.0417260516*((content/distanc)/(distanc*filler)) +1.07196464374546\n","\n","\n","RMSE: 0.15977525197020528\n","\n","\n","R2:: 0.9942229191499277\n","************************************************ Time taken for 2 dimension is: 0.20273280143737793 seconds ************************************************ \n","\n","1 term in the equation is 0.1655595899*((content*filler)+(content/distanc))\n","\n","\n","2 term in the equation is 2.2800141812*(tanh((distanc*filler)))\n","\n","\n","3 term in the equation is -284.6160431316*((distanc/filler)/(content)**0.5)\n","\n","\n","Equation: + 0.1655595899*((content*filler)+(content/distanc)) + 2.2800141812*(tanh((distanc*filler))) -284.6160431316*((distanc/filler)/(content)**0.5) -0.792180941050710\n","\n","\n","RMSE: 0.11130114345668941\n","\n","\n","R2:: 0.9971965767198127\n","************************************************ Time taken for 3 dimension is: 3.817552328109741 seconds ************************************************ \n","\n"]}]},{"cell_type":"code","source":["# 极低贯通率\n","X = df_.iloc[3001:4001, :-1].values # All rows, all columns except the last one\n","y = df_.iloc[3001:4001, -1].values # All rows, only the last column (target)\n","# Assign column names for the features\n","cols = [\"connectivity\",\"content\",\"distanc\",\"filler\",\"interface\"]\n","\n","# Insert the target variable into the dataframe at the beginning\n","df = pd.DataFrame(X, columns=cols)\n","df.insert(0, 'Target', y)\n","df.tail()\n","operators = ['+','-','*','/','exp','ln','sin','cos','pow(2)','pow(0.5)','sqrt','tanh']\n","rmse,equation,r2,_ = SissoModel(df,operators,n_expansion=3,n_term=3,k=20,use_gpu=False).fit()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"GzxlnoDUGRPM","executionInfo":{"status":"ok","timestamp":1740117901420,"user_tz":-480,"elapsed":7363,"user":{"displayName":"xin huang","userId":"07597279731200846682"}},"outputId":"40785f1e-be83-4720-db5c-b694c235d447"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["************************************************ Starting Feature Space Construction in cpu ************************************************ \n","\n","\n","\n","************************************************ Starting 1 level of feature expansion...************************************************ \n","\n","************************************************ 1 Feature Expansion Completed with feature space size::: 90 ************************************************ \n","\n","************************************************ Time taken to create the space is::: 0.007186174392700195 Seconds...************************************************ \n","\n","************************************************ Starting 2 level of feature expansion...************************************************ \n","\n","************************************************ 2 Feature Expansion Completed with feature space size::: 20702 ************************************************ \n","\n","************************************************ Time taken to create the space is::: 2.483508586883545 Seconds...************************************************ \n","\n","\n","\n","************************************************ Starting sparse model building in cpu ************************************************ \n","\n","\n","\n","Equation: 1.019251*((connectivity-distanc)*(content*filler))+1.1824496249565453\n","\n","\n","RMSE: 0.3018702780872995\n","R2:: 0.9889683932421198\n","\n","\n","************************************************ Time taken to generate one dimensional equation: 0.09834814071655273 seconds ************************************************ \n","\n","\n","\n","1 term in the equation is 0.0864289368*((content*filler)/(connectivity+distanc))\n","\n","\n","2 term in the equation is 0.0720419986*((content*interface)+(content/distanc))\n","\n","\n","Equation: + 0.0864289368*((content*filler)/(connectivity+distanc)) + 0.0720419986*((content*interface)+(content/distanc)) +0.999880758389912\n","\n","\n","RMSE: 0.13535979474491697\n","\n","\n","R2:: 0.9977819188204917\n","************************************************ Time taken for 2 dimension is: 0.30084753036499023 seconds ************************************************ \n","\n","1 term in the equation is 0.0832150169*((content*filler)/(connectivity+distanc))\n","\n","\n","2 term in the equation is 0.0710790394*((content*interface)+(content/distanc))\n","\n","\n","3 term in the equation is -4.1647236293*((connectivity+content)/(content*filler))\n","\n","\n","Equation: + 0.0832150169*((content*filler)/(connectivity+distanc)) + 0.0710790394*((content*interface)+(content/distanc)) -4.1647236293*((connectivity+content)/(content*filler)) +1.322964622893490\n","\n","\n","RMSE: 0.08911581826231478\n","\n","\n","R2:: 0.9990385921567027\n","************************************************ Time taken for 3 dimension is: 3.7193357944488525 seconds ************************************************ \n","\n"]}]},{"cell_type":"code","source":["# 极低贯通率\n","X = df_.iloc[4001:5001, :-1].values # All rows, all columns except the last one\n","y = df_.iloc[4001:5001, -1].values # All rows, only the last column (target)\n","# Assign column names for the features\n","cols = [\"connectivity\",\"content\",\"distanc\",\"filler\",\"interface\"]\n","\n","# Insert the target variable into the dataframe at the beginning\n","df = pd.DataFrame(X, columns=cols)\n","df.insert(0, 'Target', y)\n","df.tail()\n","operators = ['+','-','*','/','exp','ln','sin','cos','pow(2)','pow(0.5)','sqrt','tanh']\n","rmse,equation,r2,_ = SissoModel(df,operators,n_expansion=3,n_term=3,k=20,use_gpu=False).fit()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"pCsF6vSnGRTG","executionInfo":{"status":"ok","timestamp":1740117920943,"user_tz":-480,"elapsed":7907,"user":{"displayName":"xin huang","userId":"07597279731200846682"}},"outputId":"8030d4fb-1bf5-42b5-e285-d44f92385ac3"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["************************************************ Starting Feature Space Construction in cpu ************************************************ \n","\n","\n","\n","************************************************ Starting 1 level of feature expansion...************************************************ \n","\n","************************************************ 1 Feature Expansion Completed with feature space size::: 90 ************************************************ \n","\n","************************************************ Time taken to create the space is::: 0.006832122802734375 Seconds...************************************************ \n","\n","************************************************ Starting 2 level of feature expansion...************************************************ \n","\n","************************************************ 2 Feature Expansion Completed with feature space size::: 20702 ************************************************ \n","\n","************************************************ Time taken to create the space is::: 2.7908663749694824 Seconds...************************************************ \n","\n","\n","\n","************************************************ Starting sparse model building in cpu ************************************************ \n","\n","\n","\n","Equation: 0.355326*((content*filler)-(ln(distanc)))-0.2750929825607864\n","\n","\n","RMSE: 0.24404495769778803\n","R2:: 0.9955479224253961\n","\n","\n","************************************************ Time taken to generate one dimensional equation: 0.09692144393920898 seconds ************************************************ \n","\n","\n","\n","1 term in the equation is 0.3512073999*((content*filler)/(exp(distanc)))\n","\n","\n","2 term in the equation is 0.0778333701*((content*interface)+(content/distanc))\n","\n","\n","Equation: + 0.3512073999*((content*filler)/(exp(distanc))) + 0.0778333701*((content*interface)+(content/distanc)) +0.861323549950275\n","\n","\n","RMSE: 0.11573387550406525\n","\n","\n","R2:: 0.9989987465219637\n","************************************************ Time taken for 2 dimension is: 0.298997163772583 seconds ************************************************ \n","\n","1 term in the equation is 0.3406500102*((content*filler)/(exp(distanc)))\n","\n","\n","2 term in the equation is 0.0799164747*((content*interface)+(content/distanc))\n","\n","\n","3 term in the equation is -8.6772250072*((content)**0.5/(content*filler))\n","\n","\n","Equation: + 0.3406500102*((content*filler)/(exp(distanc))) + 0.0799164747*((content*interface)+(content/distanc)) -8.6772250072*((content)**0.5/(content*filler)) +1.210881838159485\n","\n","\n","RMSE: 0.07247255463518511\n","\n","\n","R2:: 0.9996073820186763\n","************************************************ Time taken for 3 dimension is: 3.71016788482666 seconds ************************************************ \n","\n"]}]},{"cell_type":"code","source":["# 极低贯通率\n","X = df_.iloc[5001:6001, :-1].values # All rows, all columns except the last one\n","y = df_.iloc[5001:6001, -1].values # All rows, only the last column (target)\n","# Assign column names for the features\n","cols = [\"connectivity\",\"content\",\"distanc\",\"filler\",\"interface\"]\n","\n","# Insert the target variable into the dataframe at the beginning\n","df = pd.DataFrame(X, columns=cols)\n","df.insert(0, 'Target', y)\n","df.tail()\n","operators = ['+','-','*','/','exp','ln','sin','cos','pow(2)','pow(0.5)','sqrt','tanh']\n","rmse,equation,r2,_ = SissoModel(df,operators,n_expansion=3,n_term=3,k=20,use_gpu=False).fit()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"3EITmlyCGRXL","executionInfo":{"status":"ok","timestamp":1740117942213,"user_tz":-480,"elapsed":8864,"user":{"displayName":"xin huang","userId":"07597279731200846682"}},"outputId":"5e262b03-23e4-48c3-c693-807772c5def5"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["************************************************ Starting Feature Space Construction in cpu ************************************************ \n","\n","\n","\n","************************************************ Starting 1 level of feature expansion...************************************************ \n","\n","************************************************ 1 Feature Expansion Completed with feature space size::: 90 ************************************************ \n","\n","************************************************ Time taken to create the space is::: 0.0067138671875 Seconds...************************************************ \n","\n","************************************************ Starting 2 level of feature expansion...************************************************ \n","\n","************************************************ 2 Feature Expansion Completed with feature space size::: 20702 ************************************************ \n","\n","************************************************ Time taken to create the space is::: 2.8841304779052734 Seconds...************************************************ \n","\n","\n","\n","************************************************ Starting sparse model building in cpu ************************************************ \n","\n","\n","\n","Equation: 0.432127*((content*filler)-(ln(distanc)))-0.6728716071972265\n","\n","\n","RMSE: 0.22013915505091886\n","R2:: 0.9975457494019513\n","\n","\n","************************************************ Time taken to generate one dimensional equation: 0.09830689430236816 seconds ************************************************ \n","\n","\n","\n","1 term in the equation is 0.4237613866*((content*filler)-(ln(distanc)))\n","\n","\n","2 term in the equation is 0.8738222557*((content+distanc)*(interface)**0.5)\n","\n","\n","Equation: + 0.4237613866*((content*filler)-(ln(distanc))) + 0.8738222557*((content+distanc)*(interface)**0.5) -0.921038207394897\n","\n","\n","RMSE: 0.147370262596498\n","\n","\n","R2:: 0.9989001232294211\n","************************************************ Time taken for 2 dimension is: 0.40147852897644043 seconds ************************************************ \n","\n","1 term in the equation is 0.4166182754*((content*filler)-(distanc)**0.5)\n","\n","\n","2 term in the equation is 0.5529990680*((content*interface)/(ln(filler)))\n","\n","\n","3 term in the equation is -0.0899223959*((content/distanc)*(sin(filler)))\n","\n","\n","Equation: + 0.4166182754*((content*filler)-(distanc)**0.5) + 0.5529990680*((content*interface)/(ln(filler))) -0.0899223959*((content/distanc)*(sin(filler))) +0.821390997981874\n","\n","\n","RMSE: 0.09343956201747677\n","\n","\n","R2:: 0.9995578334314288\n","************************************************ Time taken for 3 dimension is: 4.608341932296753 seconds ************************************************ \n","\n"]}]},{"cell_type":"code","source":["# 极低贯通率\n","X = df_.iloc[6001:7001, :-1].values # All rows, all columns except the last one\n","y = df_.iloc[6001:7001, -1].values # All rows, only the last column (target)\n","# Assign column names for the features\n","cols = [\"connectivity\",\"content\",\"distanc\",\"filler\",\"interface\"]\n","\n","# Insert the target variable into the dataframe at the beginning\n","df = pd.DataFrame(X, columns=cols)\n","df.insert(0, 'Target', y)\n","df.tail()\n","operators = ['+','-','*','/','exp','ln','sin','cos','pow(2)','pow(0.5)','sqrt','tanh']\n","rmse,equation,r2,_ = SissoModel(df,operators,n_expansion=3,n_term=3,k=20,use_gpu=False).fit()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"Mv0bzAz7Gn1E","executionInfo":{"status":"ok","timestamp":1740117955793,"user_tz":-480,"elapsed":8446,"user":{"displayName":"xin huang","userId":"07597279731200846682"}},"outputId":"be019537-0592-470a-b78d-4977d6864818"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["************************************************ Starting Feature Space Construction in cpu ************************************************ \n","\n","\n","\n","************************************************ Starting 1 level of feature expansion...************************************************ \n","\n","************************************************ 1 Feature Expansion Completed with feature space size::: 90 ************************************************ \n","\n","************************************************ Time taken to create the space is::: 0.006833791732788086 Seconds...************************************************ \n","\n","************************************************ Starting 2 level of feature expansion...************************************************ \n","\n","************************************************ 2 Feature Expansion Completed with feature space size::: 20703 ************************************************ \n","\n","************************************************ Time taken to create the space is::: 2.88091778755188 Seconds...************************************************ \n","\n","\n","\n","************************************************ Starting sparse model building in cpu ************************************************ \n","\n","\n","\n","Equation: 0.319477*((content*filler)/(connectivity+distanc))+0.8436641441682768\n","\n","\n","RMSE: 0.2224377994674281\n","R2:: 0.9981903809720871\n","\n","\n","************************************************ Time taken to generate one dimensional equation: 0.09856057167053223 seconds ************************************************ \n","\n","\n","\n","1 term in the equation is 0.3156264803*((content*filler)/(connectivity+distanc))\n","\n","\n","2 term in the equation is 0.7584616079*((content*interface)/(ln(filler)))\n","\n","\n","Equation: + 0.3156264803*((content*filler)/(connectivity+distanc)) + 0.7584616079*((content*interface)/(ln(filler))) +0.680419248813873\n","\n","\n","RMSE: 0.14225533376679322\n","\n","\n","R2:: 0.999259871561108\n","************************************************ Time taken for 2 dimension is: 0.29936838150024414 seconds ************************************************ \n","\n","1 term in the equation is 0.5159352384*((content*filler)/(exp(distanc)))\n","\n","\n","2 term in the equation is 0.7067487434*((content*interface)/(ln(filler)))\n","\n","\n","3 term in the equation is 0.0001474497*((content+distanc)/(distanc)**2)\n","\n","\n","Equation: + 0.5159352384*((content*filler)/(exp(distanc))) + 0.7067487434*((content*interface)/(ln(filler))) + 0.0001474497*((content+distanc)/(distanc)**2) +0.653130181421398\n","\n","\n","RMSE: 0.09032457750174748\n","\n","\n","R2:: 0.9997016116577232\n","************************************************ Time taken for 3 dimension is: 3.8972878456115723 seconds ************************************************ \n","\n"]}]},{"cell_type":"code","source":["# 极低贯通率\n","X = df_.iloc[7001:8001, :-1].values # All rows, all columns except the last one\n","y = df_.iloc[7001:8001, -1].values # All rows, only the last column (target)\n","# Assign column names for the features\n","cols = [\"connectivity\",\"content\",\"distanc\",\"filler\",\"interface\"]\n","\n","# Insert the target variable into the dataframe at the beginning\n","df = pd.DataFrame(X, columns=cols)\n","df.insert(0, 'Target', y)\n","df.tail()\n","operators = ['+','-','*','/','exp','ln','sin','cos','pow(2)','pow(0.5)','sqrt','tanh']\n","rmse,equation,r2,_ = SissoModel(df,operators,n_expansion=3,n_term=3,k=20,use_gpu=False).fit()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"t4DeUExbGn6T","executionInfo":{"status":"ok","timestamp":1740117976292,"user_tz":-480,"elapsed":7829,"user":{"displayName":"xin huang","userId":"07597279731200846682"}},"outputId":"36a31b3e-951f-409c-bbce-7f1513b149c7"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["************************************************ Starting Feature Space Construction in cpu ************************************************ \n","\n","\n","\n","************************************************ Starting 1 level of feature expansion...************************************************ \n","\n","************************************************ 1 Feature Expansion Completed with feature space size::: 90 ************************************************ \n","\n","************************************************ Time taken to create the space is::: 0.0066699981689453125 Seconds...************************************************ \n","\n","************************************************ Starting 2 level of feature expansion...************************************************ \n","\n","************************************************ 2 Feature Expansion Completed with feature space size::: 20703 ************************************************ \n","\n","************************************************ Time taken to create the space is::: 2.6682116985321045 Seconds...************************************************ \n","\n","\n","\n","************************************************ Starting sparse model building in cpu ************************************************ \n","\n","\n","\n","Equation: 0.602422*((content*filler)/(exp(distanc)))+0.7242138305815047\n","\n","\n","RMSE: 0.2138576543303438\n","R2:: 0.9987346092102664\n","\n","\n","************************************************ Time taken to generate one dimensional equation: 0.09758257865905762 seconds ************************************************ \n","\n","\n","\n","1 term in the equation is 0.5956181501*((content*filler)/(exp(distanc)))\n","\n","\n","2 term in the equation is 0.8169569342*((content*interface)/(ln(filler)))\n","\n","\n","Equation: + 0.5956181501*((content*filler)/(exp(distanc))) + 0.8169569342*((content*interface)/(ln(filler))) +0.548477177918423\n","\n","\n","RMSE: 0.10868611253953847\n","\n","\n","R2:: 0.9996731690459995\n","************************************************ Time taken for 2 dimension is: 0.302278995513916 seconds ************************************************ \n","\n","1 term in the equation is 0.5782182815*((content*filler)-(distanc)**0.5)\n","\n","\n","2 term in the equation is 0.7623202590*((content*interface)/(ln(filler)))\n","\n","\n","3 term in the equation is 0.0001532183*((content+distanc)/(distanc)**2)\n","\n","\n","Equation: + 0.5782182815*((content*filler)-(distanc)**0.5) + 0.7623202590*((content*interface)/(ln(filler))) + 0.0001532183*((content+distanc)/(distanc)**2) +0.612320248390004\n","\n","\n","RMSE: 0.07425725063337259\n","\n","\n","R2:: 0.9998474357417012\n","************************************************ Time taken for 3 dimension is: 3.8115487098693848 seconds ************************************************ \n","\n"]}]},{"cell_type":"code","source":["# 极低贯通率\n","X = df_.iloc[8001:9001, :-1].values # All rows, all columns except the last one\n","y = df_.iloc[8001:9001, -1].values # All rows, only the last column (target)\n","# Assign column names for the features\n","cols = [\"connectivity\",\"content\",\"distanc\",\"filler\",\"interface\"]\n","\n","# Insert the target variable into the dataframe at the beginning\n","df = pd.DataFrame(X, columns=cols)\n","df.insert(0, 'Target', y)\n","df.tail()\n","operators = ['+','-','*','/','exp','ln','sin','cos','pow(2)','pow(0.5)','sqrt','tanh']\n","rmse,equation,r2,_ = SissoModel(df,operators,n_expansion=3,n_term=3,k=20,use_gpu=False).fit()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"bfu2n8OjGn9w","executionInfo":{"status":"ok","timestamp":1740117996068,"user_tz":-480,"elapsed":8177,"user":{"displayName":"xin huang","userId":"07597279731200846682"}},"outputId":"5ac3508b-51e3-4737-af4b-9490617bbf3d"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["************************************************ Starting Feature Space Construction in cpu ************************************************ \n","\n","\n","\n","************************************************ Starting 1 level of feature expansion...************************************************ \n","\n","************************************************ 1 Feature Expansion Completed with feature space size::: 90 ************************************************ \n","\n","************************************************ Time taken to create the space is::: 0.006484270095825195 Seconds...************************************************ \n","\n","************************************************ Starting 2 level of feature expansion...************************************************ \n","\n","************************************************ 2 Feature Expansion Completed with feature space size::: 20703 ************************************************ \n","\n","************************************************ Time taken to create the space is::: 2.5675947666168213 Seconds...************************************************ \n","\n","\n","\n","************************************************ Starting sparse model building in cpu ************************************************ \n","\n","\n","\n","Equation: 0.66596*((content*filler)+(interface/filler))+0.44805286380090337\n","\n","\n","RMSE: 0.18879929176766982\n","R2:: 0.9992272496540688\n","\n","\n","************************************************ Time taken to generate one dimensional equation: 0.004456996917724609 seconds ************************************************ \n","\n","\n","\n","1 term in the equation is 0.6510545822*((content*filler)+(content)**0.5)\n","\n","\n","2 term in the equation is 0.1524052561*((content*interface)-(ln(distanc)))\n","\n","\n","Equation: + 0.6510545822*((content*filler)+(content)**0.5) + 0.1524052561*((content*interface)-(ln(distanc))) -0.2302872567431837\n","\n","\n","RMSE: 0.05916130236332384\n","\n","\n","R2:: 0.9999241223384057\n","************************************************ Time taken for 2 dimension is: 0.4957435131072998 seconds ************************************************ \n","\n","1 term in the equation is 0.6461997323*((content*filler)+(content)**0.5)\n","\n","\n","2 term in the equation is 0.1598073684*((content*interface)-(ln(distanc)))\n","\n","\n","3 term in the equation is -1.4779577347*((connectivity/filler)/(content+distanc))\n","\n","\n","Equation: + 0.6461997323*((content*filler)+(content)**0.5) + 0.1598073684*((content*interface)-(ln(distanc))) -1.4779577347*((connectivity/filler)/(content+distanc)) -0.1137664697805380\n","\n","\n","RMSE: 0.033226684219148234\n","\n","\n","R2:: 0.9999760661538748\n","************************************************ Time taken for 3 dimension is: 4.010974645614624 seconds ************************************************ \n","\n"]}]},{"cell_type":"code","source":["# 极低贯通率\n","X = df_.iloc[9001:10001, :-1].values # All rows, all columns except the last one\n","y = df_.iloc[9001:10001, -1].values # All rows, only the last column (target)\n","# Assign column names for the features\n","cols = [\"connectivity\",\"content\",\"distanc\",\"filler\",\"interface\"]\n","\n","# Insert the target variable into the dataframe at the beginning\n","df = pd.DataFrame(X, columns=cols)\n","df.insert(0, 'Target', y)\n","df.tail()\n","operators = ['+','-','*','/','exp','ln','sin','cos','pow(2)','pow(0.5)','sqrt','tanh']\n","rmse,equation,r2,_ = SissoModel(df,operators,n_expansion=3,n_term=3,k=20,use_gpu=False).fit()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"WMhlA6QCGoBR","executionInfo":{"status":"ok","timestamp":1740118036794,"user_tz":-480,"elapsed":7889,"user":{"displayName":"xin huang","userId":"07597279731200846682"}},"outputId":"feb9191a-2377-45d9-e191-29498031a4f5"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["************************************************ Starting Feature Space Construction in cpu ************************************************ \n","\n","\n","\n","************************************************ Starting 1 level of feature expansion...************************************************ \n","\n","************************************************ 1 Feature Expansion Completed with feature space size::: 90 ************************************************ \n","\n","************************************************ Time taken to create the space is::: 0.006674289703369141 Seconds...************************************************ \n","\n","************************************************ Starting 2 level of feature expansion...************************************************ \n","\n","************************************************ 2 Feature Expansion Completed with feature space size::: 20614 ************************************************ \n","\n","************************************************ Time taken to create the space is::: 2.6173486709594727 Seconds...************************************************ \n","\n","\n","\n","************************************************ Starting sparse model building in cpu ************************************************ \n","\n","\n","\n","Equation: 0.7422*((content*filler)+(interface/filler))+0.30863911150605716\n","\n","\n","RMSE: 0.16978335694895935\n","R2:: 0.9994967304867768\n","\n","\n","************************************************ Time taken to generate one dimensional equation: 0.09837627410888672 seconds ************************************************ \n","\n","\n","\n","1 term in the equation is 0.7269977560*((content+distanc)+(content*filler))\n","\n","\n","2 term in the equation is -0.0457367232*((content*interface)*(ln(distanc)))\n","\n","\n","Equation: + 0.7269977560*((content+distanc)+(content*filler)) -0.0457367232*((content*interface)*(ln(distanc))) +0.2771844318469227\n","\n","\n","RMSE: 0.04332375857701267\n","\n","\n","R2:: 0.9999672310410384\n","************************************************ Time taken for 2 dimension is: 0.304640531539917 seconds ************************************************ \n","\n","1 term in the equation is 0.7288037346*((connectivity+filler)*(tanh(content)))\n","\n","\n","2 term in the equation is -0.0421384950*((content*interface)*(ln(distanc)))\n","\n","\n","3 term in the equation is -0.9296704484*((connectivity-interface)*(content*distanc))\n","\n","\n","Equation: + 0.7288037346*((connectivity+filler)*(tanh(content))) -0.0421384950*((content*interface)*(ln(distanc))) -0.9296704484*((connectivity-interface)*(content*distanc)) +0.276119457057367\n","\n","\n","RMSE: 0.027687176020641698\n","\n","\n","R2:: 0.9999866165613068\n","************************************************ Time taken for 3 dimension is: 3.9022727012634277 seconds ************************************************ \n","\n"]}]},{"cell_type":"code","source":["# 极低贯通率\n","X = df_.iloc[10001:11001, :-1].values # All rows, all columns except the last one\n","y = df_.iloc[10001:11001, -1].values # All rows, only the last column (target)\n","# Assign column names for the features\n","cols = [\"connectivity\",\"content\",\"distanc\",\"filler\",\"interface\"]\n","\n","# Insert the target variable into the dataframe at the beginning\n","df = pd.DataFrame(X, columns=cols)\n","df.insert(0, 'Target', y)\n","df.tail()\n","operators = ['+','-','*','/','exp','ln','sin','cos','pow(2)','pow(0.5)','sqrt','tanh']\n","rmse,equation,r2,_ = SissoModel(df,operators,n_expansion=3,n_term=3,k=20,use_gpu=False).fit()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"hDwesQx0GoEv","executionInfo":{"status":"ok","timestamp":1740118054176,"user_tz":-480,"elapsed":5913,"user":{"displayName":"xin huang","userId":"07597279731200846682"}},"outputId":"d3f57b9e-6169-4915-e9b1-9c7c272f819f"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["************************************************ Starting Feature Space Construction in cpu ************************************************ \n","\n","\n","\n","************************************************ Starting 1 level of feature expansion...************************************************ \n","\n","************************************************ 1 Feature Expansion Completed with feature space size::: 62 ************************************************ \n","\n","************************************************ Time taken to create the space is::: 0.005533933639526367 Seconds...************************************************ \n","\n","************************************************ Starting 2 level of feature expansion...************************************************ \n","\n","************************************************ 2 Feature Expansion Completed with feature space size::: 9916 ************************************************ \n","\n","************************************************ Time taken to create the space is::: 1.4641339778900146 Seconds...************************************************ \n","\n","\n","\n","************************************************ Starting sparse model building in cpu ************************************************ \n","\n","\n","\n","Equation: 0.815454*((content*filler)+(interface/filler))+0.1380568012436303\n","\n","\n","RMSE: 0.16505497076753864\n","R2:: 0.99960569121659\n","\n","\n","************************************************ Time taken to generate one dimensional equation: 0.003037691116333008 seconds ************************************************ \n","\n","\n","\n","1 term in the equation is 0.8002683552*(content*(content+filler))\n","\n","\n","2 term in the equation is 0.2018594842*((content*interface)-(content)**0.5)\n","\n","\n","Equation: + 0.8002683552*(content*(content+filler)) + 0.2018594842*((content*interface)-(content)**0.5) +0.1741871828496410\n","\n","\n","RMSE: 0.0038664592814519963\n","\n","\n","R2:: 0.9999997836254801\n","************************************************ Time taken for 2 dimension is: 0.39575791358947754 seconds ************************************************ \n","\n","1 term in the equation is 0.8000027001*((content*filler)-(cos(content)))\n","\n","\n","2 term in the equation is -0.0137245102*((content*interface)-(ln(content)))\n","\n","\n","3 term in the equation is 0.2137518081*((content*interface)-(content)**0.5)\n","\n","\n","Equation: + 0.8000027001*((content*filler)-(cos(content))) -0.0137245102*((content*interface)-(ln(content))) + 0.2137518081*((content*interface)-(content)**0.5) +1.02659977540764\n","\n","\n","RMSE: 0.0002473112549341224\n","\n","\n","R2:: 0.9999999991147478\n","************************************************ Time taken for 3 dimension is: 3.7059905529022217 seconds ************************************************ \n","\n"]}]},{"cell_type":"code","source":["df.tail()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":206},"id":"alhHPZoWH-Y-","executionInfo":{"status":"ok","timestamp":1740118170300,"user_tz":-480,"elapsed":58,"user":{"displayName":"xin huang","userId":"07597279731200846682"}},"outputId":"05316003-76c6-48ec-eb16-4500fde15682"},"execution_count":null,"outputs":[{"output_type":"execute_result","data":{"text/plain":[" Target connectivity content distanc filler interface\n","994 32.3315 1.0 0.1 0.05 400.0 10.0\n","995 32.5315 1.0 0.1 0.05 400.0 20.0\n","996 32.7315 1.0 0.1 0.05 400.0 30.0\n","997 32.9315 1.0 0.1 0.05 400.0 40.0\n","998 33.1315 1.0 0.1 0.05 400.0 50.0"],"text/html":["\n","
\n","
\n","\n","\n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n","
Targetconnectivitycontentdistancfillerinterface
99432.33151.00.10.05400.010.0
99532.53151.00.10.05400.020.0
99632.73151.00.10.05400.030.0
99732.93151.00.10.05400.040.0
99833.13151.00.10.05400.050.0
\n","
\n","
\n","\n","
\n"," \n","\n"," \n","\n"," \n","
\n","\n","\n","
\n"," \n","\n","\n","\n"," \n","
\n","\n","
\n","
\n"],"application/vnd.google.colaboratory.intrinsic+json":{"type":"dataframe","summary":"{\n \"name\": \"df\",\n \"rows\": 5,\n \"fields\": [\n {\n \"column\": \"Target\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.31622776601683905,\n \"min\": 32.3315,\n \"max\": 33.1315,\n \"num_unique_values\": 5,\n \"samples\": [\n 32.5315,\n 33.1315,\n 32.7315\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"connectivity\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.0,\n \"min\": 1.0,\n \"max\": 1.0,\n \"num_unique_values\": 1,\n \"samples\": [\n 1.0\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"content\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.0,\n \"min\": 0.1,\n \"max\": 0.1,\n \"num_unique_values\": 1,\n \"samples\": [\n 0.1\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"distanc\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.0,\n \"min\": 0.05,\n \"max\": 0.05,\n \"num_unique_values\": 1,\n \"samples\": [\n 0.05\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"filler\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.0,\n \"min\": 400.0,\n \"max\": 400.0,\n \"num_unique_values\": 1,\n \"samples\": [\n 400.0\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"interface\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 15.811388300841896,\n \"min\": 10.0,\n \"max\": 50.0,\n \"num_unique_values\": 5,\n \"samples\": [\n 20.0\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}"}},"metadata":{},"execution_count":27}]},{"cell_type":"code","source":["df.head()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":206},"id":"SVioL7IjGRbr","executionInfo":{"status":"ok","timestamp":1740118183161,"user_tz":-480,"elapsed":23,"user":{"displayName":"xin huang","userId":"07597279731200846682"}},"outputId":"ea0c73ad-1815-4d64-ec00-0f7ffa1e7f6b"},"execution_count":null,"outputs":[{"output_type":"execute_result","data":{"text/plain":[" Target connectivity content distanc filler interface\n","0 1.8231 1.0 0.02 0.005 100.0 20.0\n","1 1.8631 1.0 0.02 0.005 100.0 30.0\n","2 1.9031 1.0 0.02 0.005 100.0 40.0\n","3 1.9431 1.0 0.02 0.005 100.0 50.0\n","4 3.3831 1.0 0.02 0.005 200.0 10.0"],"text/html":["\n","
\n","
\n","\n","\n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n","
Targetconnectivitycontentdistancfillerinterface
01.82311.00.020.005100.020.0
11.86311.00.020.005100.030.0
21.90311.00.020.005100.040.0
31.94311.00.020.005100.050.0
43.38311.00.020.005200.010.0
\n","
\n","
\n","\n","
\n"," \n","\n"," \n","\n"," \n","
\n","\n","\n","
\n"," \n","\n","\n","\n"," \n","
\n","\n","
\n","
\n"],"application/vnd.google.colaboratory.intrinsic+json":{"type":"dataframe","variable_name":"df","summary":"{\n \"name\": \"df\",\n \"rows\": 999,\n \"fields\": [\n {\n \"column\": \"Target\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 8.316256115618938,\n \"min\": 1.7831,\n \"max\": 33.1315,\n \"num_unique_values\": 100,\n \"samples\": [\n 8.9315,\n 15.017,\n 19.4945\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"connectivity\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.0,\n \"min\": 1.0,\n \"max\": 1.0,\n \"num_unique_values\": 1,\n \"samples\": [\n 1.0\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"content\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.0282842428781252,\n \"min\": 0.02,\n \"max\": 0.1,\n \"num_unique_values\": 5,\n \"samples\": [\n 0.04\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"distanc\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.01435811806442429,\n \"min\": 0.005,\n \"max\": 0.05,\n \"num_unique_values\": 10,\n \"samples\": [\n 0.045\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"filler\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 111.81450014359588,\n \"min\": 100.0,\n \"max\": 400.0,\n \"num_unique_values\": 4,\n \"samples\": [\n 200.0\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"interface\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 14.142121439062599,\n \"min\": 10.0,\n \"max\": 50.0,\n \"num_unique_values\": 5,\n \"samples\": [\n 30.0\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}"}},"metadata":{},"execution_count":28}]},{"cell_type":"code","source":["# 极低贯通率\n","X = df_.iloc[8001:, :-1].values # All rows, all columns except the last one\n","y = df_.iloc[8001:, -1].values # All rows, only the last column (target)\n","# Assign column names for the features\n","cols = [\"connectivity\",\"content\",\"distanc\",\"filler\",\"interface\"]\n","\n","# Insert the target variable into the dataframe at the beginning\n","df = pd.DataFrame(X, columns=cols)\n","df.insert(0, 'Target', y)\n","df.tail()\n","operators = ['+','-','*','/','exp','ln','sin','cos','pow(2)','pow(0.5)','sqrt','tanh']\n","rmse,equation,r2,_ = SissoModel(df,operators,n_expansion=3,n_term=3,k=20,use_gpu=False).fit()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"HuxpdkLgOJ16","executionInfo":{"status":"ok","timestamp":1744180087543,"user_tz":-480,"elapsed":38907,"user":{"displayName":"xin huang","userId":"07597279731200846682"}},"outputId":"87a8bf33-0ebe-4d1d-e327-158caea9cc1e"},"execution_count":6,"outputs":[{"output_type":"stream","name":"stdout","text":["************************************************ Starting Feature Space Construction in cpu ************************************************ \n","\n","\n","\n","************************************************ Starting 1 level of feature expansion...************************************************ \n","\n","************************************************ 1 Feature Expansion Completed with feature space size::: 90 ************************************************ \n","\n","************************************************ Time taken to create the space is::: 0.2474353313446045 Seconds...************************************************ \n","\n","************************************************ Starting 2 level of feature expansion...************************************************ \n","\n","************************************************ 2 Feature Expansion Completed with feature space size::: 20614 ************************************************ \n","\n","************************************************ Time taken to create the space is::: 10.542184591293335 Seconds...************************************************ \n","\n","\n","\n","************************************************ Starting sparse model building in cpu ************************************************ \n","\n","\n","\n","Equation: 0.298299*((content*filler)*(exp(connectivity)))+0.48995997658614066\n","\n","\n","RMSE: 0.2493955141641328\n","R2:: 0.9989301949728626\n","\n","\n","************************************************ Time taken to generate one dimensional equation: 0.17671895027160645 seconds ************************************************ \n","\n","\n","\n","1 term in the equation is 0.2939970150*((content*filler)*(exp(connectivity)))\n","\n","\n","2 term in the equation is 0.1978718347*((content*interface)-(exp(connectivity)))\n","\n","\n","Equation: + 0.2939970150*((content*filler)*(exp(connectivity))) + 0.1978718347*((content*interface)-(exp(connectivity))) +0.781294705596192\n","\n","\n","RMSE: 0.10023813322136116\n","\n","\n","R2:: 0.9998271802792231\n","************************************************ Time taken for 2 dimension is: 0.6432037353515625 seconds ************************************************ \n","\n","1 term in the equation is 0.2943885645*((content*filler)*(exp(connectivity)))\n","\n","\n","2 term in the equation is 0.1889939690*((content*interface)-(exp(connectivity)))\n","\n","\n","3 term in the equation is -0.0103095993*((connectivity/distanc)*(ln(connectivity)))\n","\n","\n","Equation: + 0.2943885645*((content*filler)*(exp(connectivity))) + 0.1889939690*((content*interface)-(exp(connectivity))) -0.0103095993*((connectivity/distanc)*(ln(connectivity))) +0.705950188864630\n","\n","\n","RMSE: 0.06529238778889496\n","\n","\n","R2:: 0.9999266749341272\n","************************************************ Time taken for 3 dimension is: 23.69806981086731 seconds ************************************************ \n","\n"]}]},{"cell_type":"code","source":["#全部数据测试\n","X = df_.iloc[:, :-1].values # All rows, all columns except the last one\n","y = df_.iloc[:, -1].values # All rows, only the last column (target)\n","\n","# Assign column names for the features\n","cols = [\"connectivity\",\"content\",\"distanc\",\"filler\",\"interface\"]\n","\n","# Insert the target variable into the dataframe at the beginning\n","df = pd.DataFrame(X, columns=cols)\n","df.insert(0, 'Target', y)\n","\n","# Display the dataframe (optional)\n","print(df.head())\n","operators = ['+','-','*','/','exp','ln','sin','cos','pow(2)','pow(0.5)','sqrt','tanh']\n","rmse,equation,r2,_ = SissoModel(df,operators,n_expansion=3,n_term=3,k=20,use_gpu=False).fit()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"z0nJWsd9NbSO","executionInfo":{"status":"ok","timestamp":1740112064553,"user_tz":-480,"elapsed":57854,"user":{"displayName":"xin huang","userId":"07597279731200846682"}},"outputId":"6b2fcea8-cbca-48ea-fba4-0a7396eea955"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":[" Target connectivity content distanc filler interface\n","0 1.30980 0.0 0.02 0.005 100.0 10.0\n","1 1.33020 0.0 0.02 0.005 100.0 20.0\n","2 1.35025 0.0 0.02 0.005 100.0 30.0\n","3 1.37000 0.0 0.02 0.005 100.0 40.0\n","4 1.38940 0.0 0.02 0.005 100.0 50.0\n","************************************************ Starting Feature Space Construction in cpu ************************************************ \n","\n","\n","\n","************************************************ Starting 1 level of feature expansion...************************************************ \n","\n","************************************************ 1 Feature Expansion Completed with feature space size::: 85 ************************************************ \n","\n","************************************************ Time taken to create the space is::: 0.39151501655578613 Seconds...************************************************ \n","\n","************************************************ Starting 2 level of feature expansion...************************************************ \n","\n","************************************************ 2 Feature Expansion Completed with feature space size::: 17297 ************************************************ \n","\n","************************************************ Time taken to create the space is::: 17.086864709854126 Seconds...************************************************ \n","\n","\n","\n","************************************************ Starting sparse model building in cpu ************************************************ \n","\n","\n","\n","Equation: 0.716818*((connectivity+content)*(content*filler))+1.2365915052269774\n","\n","\n","RMSE: 0.5059197065077421\n","R2:: 0.9930221521831663\n","\n","\n","************************************************ Time taken to generate one dimensional equation: 0.10386276245117188 seconds ************************************************ \n","\n","\n","\n","1 term in the equation is 0.7986623963*((connectivity+distanc)*(content*filler))\n","\n","\n","2 term in the equation is -0.8259146599*((ln(distanc))*(cos(connectivity)))\n","\n","\n","Equation: + 0.7986623963*((connectivity+distanc)*(content*filler)) -0.8259146599*((ln(distanc))*(cos(connectivity))) -1.528419658794546\n","\n","\n","RMSE: 0.3655101879372584\n","\n","\n","R2:: 0.9963578546450037\n","************************************************ Time taken for 2 dimension is: 1.1184749603271484 seconds ************************************************ \n","\n","1 term in the equation is 0.7858940228*((connectivity+distanc)*(content*filler))\n","\n","\n","2 term in the equation is -0.9734521064*((ln(distanc))*(cos(connectivity)))\n","\n","\n","3 term in the equation is -0.0493158929*((cos(connectivity))/(content+distanc))\n","\n","\n","Equation: + 0.7858940228*((connectivity+distanc)*(content*filler)) -0.9734521064*((ln(distanc))*(cos(connectivity))) -0.0493158929*((cos(connectivity))/(content+distanc)) -1.33341232277519\n","\n","\n","RMSE: 0.2769606153366895\n","\n","\n","R2:: 0.9979088059636949\n","************************************************ Time taken for 3 dimension is: 33.181583881378174 seconds ************************************************ \n","\n"]}]},{"cell_type":"code","source":[],"metadata":{"id":"eT_dp0NZBGlI"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":[],"metadata":{"id":"xs84OO44BGoa"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":[],"metadata":{"id":"cnp_PKVlBGrX"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":[],"metadata":{"id":"VKu9l9-iBGuW"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":[],"metadata":{"id":"b7s57ZuIBGxu"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["df_1 = pd.read_csv('/content/drive/MyDrive/EM_sensitivity_0430/summary_linear_K.csv')\n","df_1.head()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":206},"id":"L06u5qDnVpXX","executionInfo":{"status":"ok","timestamp":1740112130883,"user_tz":-480,"elapsed":741,"user":{"displayName":"xin huang","userId":"07597279731200846682"}},"outputId":"a36197d7-8aed-47bd-aa72-bde796a3df9c"},"execution_count":null,"outputs":[{"output_type":"execute_result","data":{"text/plain":[" connectivity content distance filler interface K\n","0 10 0.02 0.005 100 10 1.36155\n","1 10 0.02 0.005 100 20 1.38415\n","2 10 0.02 0.005 100 30 1.40650\n","3 10 0.02 0.005 100 40 1.42835\n","4 10 0.02 0.005 100 50 1.45010"],"text/html":["\n","
\n","
\n","\n","\n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n","
connectivitycontentdistancefillerinterfaceK
0100.020.005100101.36155
1100.020.005100201.38415
2100.020.005100301.40650
3100.020.005100401.42835
4100.020.005100501.45010
\n","
\n","
\n","\n","
\n"," \n","\n"," \n","\n"," \n","
\n","\n","\n","
\n"," \n","\n","\n","\n"," \n","
\n","\n","
\n","
\n"],"application/vnd.google.colaboratory.intrinsic+json":{"type":"dataframe","variable_name":"df_1","summary":"{\n \"name\": \"df_1\",\n \"rows\": 5000,\n \"fields\": [\n {\n \"column\": \"connectivity\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 14,\n \"min\": 0,\n \"max\": 40,\n \"num_unique_values\": 5,\n \"samples\": [\n 0,\n 40,\n 20\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"content\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.02828710009892144,\n \"min\": 0.02,\n \"max\": 0.1,\n \"num_unique_values\": 5,\n \"samples\": [\n 0.04,\n 0.1,\n 0.06\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"distance\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.014362842972463716,\n \"min\": 0.005,\n \"max\": 0.05,\n \"num_unique_values\": 10,\n \"samples\": [\n 0.045,\n 0.01,\n 0.03\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"filler\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 111,\n \"min\": 100,\n \"max\": 400,\n \"num_unique_values\": 4,\n \"samples\": [\n 200,\n 400,\n 100\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"interface\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 14,\n \"min\": 10,\n \"max\": 50,\n \"num_unique_values\": 5,\n \"samples\": [\n 20,\n 50,\n 30\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"K\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 2.831096843825513,\n \"min\": 0.91085,\n \"max\": 16.65,\n \"num_unique_values\": 4828,\n \"samples\": [\n 3.0051,\n 5.509,\n 2.55795\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}"}},"metadata":{},"execution_count":8}]},{"cell_type":"code","source":["X = df_1.iloc[:, :-1].values # All rows, all columns except the last one\n","y = df_1.iloc[:, -1].values # All rows, only the last column (target)\n","\n","# Assign column names for the features\n","cols = [\"connectivity\",\"content\",\"distanc\",\"filler\",\"interface\"]\n","# Insert the target variable into the dataframe at the beginning\n","df = pd.DataFrame(X, columns=cols)\n","df.insert(0, 'Target', y)\n","\n","# Display the dataframe (optional)\n","print(df.head())\n","operators = ['+','-','*','/','exp','ln','sin','cos','pow(2)','pow(0.5)','sqrt','tanh']\n","rmse, equation,r2,_ = SissoModel(df,operators,n_expansion=3,n_term=3,k=20,use_gpu=False).fit()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"u4UDZ3yTNbVe","executionInfo":{"status":"ok","timestamp":1740112181382,"user_tz":-480,"elapsed":27437,"user":{"displayName":"xin huang","userId":"07597279731200846682"}},"outputId":"d935dd4c-2350-4968-af6d-aa28f288ab5e"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":[" Target connectivity content distanc filler interface\n","0 1.36155 10.0 0.02 0.005 100.0 10.0\n","1 1.38415 10.0 0.02 0.005 100.0 20.0\n","2 1.40650 10.0 0.02 0.005 100.0 30.0\n","3 1.42835 10.0 0.02 0.005 100.0 40.0\n","4 1.45010 10.0 0.02 0.005 100.0 50.0\n","************************************************ Starting Feature Space Construction in cpu ************************************************ \n","\n","\n","\n","************************************************ Starting 1 level of feature expansion...************************************************ \n","\n","************************************************ 1 Feature Expansion Completed with feature space size::: 85 ************************************************ \n","\n","************************************************ Time taken to create the space is::: 0.1598498821258545 Seconds...************************************************ \n","\n","************************************************ Starting 2 level of feature expansion...************************************************ \n","\n","************************************************ 2 Feature Expansion Completed with feature space size::: 17288 ************************************************ \n","\n","************************************************ Time taken to create the space is::: 8.938556909561157 Seconds...************************************************ \n","\n","\n","\n","************************************************ Starting sparse model building in cpu ************************************************ \n","\n","\n","\n","Equation: -0.009299*((connectivity-filler)*(connectivity*content))+1.8191983502717939\n","\n","\n","RMSE: 0.6738821982146321\n","R2:: 0.9433310215163878\n","\n","\n","************************************************ Time taken to generate one dimensional equation: 0.10163044929504395 seconds ************************************************ \n","\n","\n","\n","1 term in the equation is -0.0086947843*((connectivity-filler)*(connectivity*content))\n","\n","\n","2 term in the equation is 2.1135275175*((content-distanc)/(distanc)**0.5)\n","\n","\n","Equation: -0.0086947843*((connectivity-filler)*(connectivity*content)) + 2.1135275175*((content-distanc)/(distanc)**0.5) +1.41405871772330\n","\n","\n","RMSE: 0.25831837675137836\n","\n","\n","R2:: 0.9916730103994779\n","************************************************ Time taken for 2 dimension is: 0.812962532043457 seconds ************************************************ \n","\n","1 term in the equation is -0.0074129406*((content-filler)*(connectivity*content))\n","\n","\n","2 term in the equation is 2.2844517287*((content-distanc)/(distanc)**0.5)\n","\n","\n","3 term in the equation is -0.0740392018*((cos(filler))/(distanc)**0.5)\n","\n","\n","Equation: -0.0074129406*((content-filler)*(connectivity*content)) + 2.2844517287*((content-distanc)/(distanc)**0.5) -0.0740392018*((cos(filler))/(distanc)**0.5) +1.54601067980591\n","\n","\n","RMSE: 0.19020379882705093\n","\n","\n","R2:: 0.9954854362552431\n","************************************************ Time taken for 3 dimension is: 14.323631525039673 seconds ************************************************ \n","\n"]}]},{"cell_type":"code","source":[],"metadata":{"id":"qfQUxLvLNbYD"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":[],"metadata":{"id":"z0l4iogqNbbK"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":[],"metadata":{"id":"ZCM0l0t8NbeD"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["import torch"],"metadata":{"id":"uTgCKvizlple"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["# create SISSO model object with relevant user-defined inputs\n","sm = SissoModel(df = df, #Takes the dataframe as input with first variable as target variable\n"," operators = operators, #Takes the user-defined operators to perform the feature engineering\n"," n_expansion = 3, #Defines the number of feature expansions need to be considered\n"," n_term = 1, #Defines the number of terms in the final equation\n"," k = 20, #Defines the number of SIS features to be screened for $L_0$ regularization\n"," initial_screening = \"spearman\", #Defines the feature screening option for high dimensional and 1-quantile_value defines\n"," #the features within this quantile range should be kept for feature expansion.\n"," use_gpu = False, #Defines the flag whether to consider GPU or not (For efficient computation we consider using GPU only for $L_0$ Regularization.\n"," dimensionality = ['u1','u2','u3'], #Defines the units of the feature variables in string representation which later converted into sympy format to do the meaningful feature construction.\n"," relational_units = [(symbols('u1')*symbols('u2'),symbols('u3'))], #Defines the list of tuples where each tuple represents the relational transformation.\n"," output_dim = (symbols('u1')*symbols('u1')), #Defines the units of the target variable which helps in narrowing down the space for Regularization.\n"," custom_unary_functions = [('cos', '-sin', 'exp', 'pow(2)'), ('tanh', 'pow(2)','sin')], #list of tuples (Note only torch supported functions can be customized)\n"," custom_binary_functions = [\n"," {\n"," 'functions1': ['pow(2)'], # Function to apply on var1 (X1)\n"," 'functions2': ['sin'], # No function to apply on var2 (X2)\n"," 'operation': torch.add, # Operation to apply between var1 and var2\n"," 'final_functions': [] # Final function to apply to the result of addition\n"," }\n"," ]\n","#List of dicts where transformations of variables and operation is provided.. (For example the given code above gives sin(x2) + X^2)\n","\n"," )\n","\n"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":219},"id":"CFXeyzoXjekp","executionInfo":{"status":"error","timestamp":1739773623909,"user_tz":-480,"elapsed":367,"user":{"displayName":"xin huang","userId":"07597279731200846682"}},"outputId":"221b6eb2-83b4-4f86-8413-5fd78e3f12b6"},"execution_count":null,"outputs":[{"output_type":"error","ename":"TypeError","evalue":"SissoModel.__init__() got an unexpected keyword argument 'df'","traceback":["\u001b[0;31m---------------------------------------------------------------------------\u001b[0m","\u001b[0;31mTypeError\u001b[0m Traceback (most recent call last)","\u001b[0;32m\u001b[0m in \u001b[0;36m\u001b[0;34m()\u001b[0m\n\u001b[1;32m 1\u001b[0m \u001b[0;31m# create SISSO model object with relevant user-defined inputs\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 2\u001b[0;31m sm = SissoModel(df = df, #Takes the dataframe as input with first variable as target variable\n\u001b[0m\u001b[1;32m 3\u001b[0m \u001b[0moperators\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0moperators\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;31m#Takes the user-defined operators to perform the feature engineering\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 4\u001b[0m \u001b[0mn_expansion\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;36m3\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;31m#Defines the number of feature expansions need to be considered\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 5\u001b[0m \u001b[0mn_term\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;31m#Defines the number of terms in the final equation\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n","\u001b[0;31mTypeError\u001b[0m: SissoModel.__init__() got an unexpected keyword argument 'df'"]}]},{"cell_type":"code","source":["# Run the SISSO algorithm to get the interpretable model with the highest accuracy\n","rmse, equation, r2,_ = sm.fit()"],"metadata":{"id":"eqDGYx7tjeny"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["from sklearn.model_selection import train_test_split\n","from sklearn.metrics import mean_squared_error\n","from TorchSisso import FeatureSpaceConstruction\n","import torch\n","import numpy as np\n","import pandas as pd\n","import time\n","import matplotlib.pyplot as plt"],"metadata":{"id":"aME8XTE1iQ7t"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["df = pd.read_csv('/content/drive/MyDrive/EM_sensitivity_0430/summary_linear_K.csv')\n","df.head()\n","X = df.drop('K', axis=1)\n","y = df['K']"],"metadata":{"id":"nrdRFPNRhqTU"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":[],"metadata":{"id":"92A2lyW5hqhE"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":[],"metadata":{"id":"WeAXtqFIhqpV"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":[],"metadata":{"id":"insj910xhqu6"},"execution_count":null,"outputs":[]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"background_save":true,"base_uri":"https://localhost:8080/"},"collapsed":true,"id":"cc-3APm7DNc1","outputId":"4f929fd9-5aec-48d2-b7e7-89eb180bfbad"},"outputs":[{"name":"stderr","output_type":"stream","text":["Command line\n","python /usr/local/lib/python3.10/dist-packages/colab_kernel_launcher.py -f /root/.local/share/jupyter/runtime/kernel-da1faf97-4067-43e7-8680-2fb638e2579f.json\n","Args\n","{'activation': 'ReLU',\n"," 'adding_bond_types': True,\n"," 'adding_h': False,\n"," 'aggregation': 'mean',\n"," 'aggregation_norm': 100,\n"," 'atom_constraints': [],\n"," 'atom_descriptor_scaling': True,\n"," 'atom_descriptors': None,\n"," 'atom_descriptors_path': None,\n"," 'atom_descriptors_size': 0,\n"," 'atom_features_size': 0,\n"," 'atom_messages': False,\n"," 'atom_targets': [],\n"," 'batch_size': 50,\n"," 'bias': False,\n"," 'bias_solvent': False,\n"," 'bond_constraints': [],\n"," 'bond_descriptor_scaling': True,\n"," 'bond_descriptors': None,\n"," 'bond_descriptors_path': None,\n"," 'bond_descriptors_size': 0,\n"," 'bond_features_size': 0,\n"," 'bond_targets': [],\n"," 'cache_cutoff': 10000,\n"," 'checkpoint_dir': None,\n"," 'checkpoint_frzn': None,\n"," 'checkpoint_path': None,\n"," 'checkpoint_paths': None,\n"," 'class_balance': False,\n"," 'config_path': None,\n"," 'constraints_path': None,\n"," 'crossval_index_dir': None,\n"," 'crossval_index_file': None,\n"," 'crossval_index_sets': None,\n"," 'cuda': True,\n"," 'data_path': '/content/drive/MyDrive/EM_sensitivity_0430/reg_IS_DV_V2.csv',\n"," 'data_weights_path': None,\n"," 'dataset_type': 'regression',\n"," 'depth': 3,\n"," 'depth_solvent': 3,\n"," 'device': device(type='cuda'),\n"," 'dropout': 0.0,\n"," 'empty_cache': False,\n"," 'ensemble_size': 1,\n"," 'epochs': 100,\n"," 'evidential_regularization': 0,\n"," 'explicit_h': False,\n"," 'extra_metrics': ['mse', 'rmse', 'mae'],\n"," 'features_generator': None,\n"," 'features_only': False,\n"," 'features_path': ['/content/drive/MyDrive/EM_sensitivity_0430/descriptor_08.csv'],\n"," 'features_scaling': True,\n"," 'features_size': None,\n"," 'ffn_hidden_size': 300,\n"," 'ffn_num_layers': 2,\n"," 'final_lr': 0.0001,\n"," 'folds_file': None,\n"," 'freeze_first_only': False,\n"," 'frzn_ffn_layers': 0,\n"," 'gpu': None,\n"," 'grad_clip': None,\n"," 'hidden_size': 300,\n"," 'hidden_size_solvent': 300,\n"," 'ignore_columns': None,\n"," 'init_lr': 0.0001,\n"," 'is_atom_bond_targets': False,\n"," 'keeping_atom_map': False,\n"," 'log_frequency': 10,\n"," 'loss_function': 'mse',\n"," 'max_data_size': None,\n"," 'max_lr': 0.001,\n"," 'metric': 'r2',\n"," 'metrics': ['r2', 'mse', 'rmse', 'mae'],\n"," 'minimize_score': False,\n"," 'mpn_shared': False,\n"," 'multiclass_num_classes': 3,\n"," 'no_adding_bond_types': False,\n"," 'no_atom_descriptor_scaling': False,\n"," 'no_bond_descriptor_scaling': False,\n"," 'no_cache_mol': False,\n"," 'no_cuda': False,\n"," 'no_features_scaling': False,\n"," 'no_shared_atom_bond_ffn': False,\n"," 'num_folds': 1,\n"," 'num_lrs': 1,\n"," 'num_tasks': 2,\n"," 'num_workers': 8,\n"," 'number_of_molecules': 1,\n"," 'overwrite_default_atom_features': False,\n"," 'overwrite_default_bond_features': False,\n"," 'phase_features_path': None,\n"," 'pytorch_seed': 0,\n"," 'quiet': False,\n"," 'reaction': False,\n"," 'reaction_mode': 'reac_diff',\n"," 'reaction_solvent': False,\n"," 'resume_experiment': False,\n"," 'save_dir': '/content/drive/MyDrive/EM_sensitivity_0430/reg_IS_DV_100_epoch',\n"," 'save_preds': False,\n"," 'save_smiles_splits': True,\n"," 'seed': 0,\n"," 'separate_test_atom_descriptors_path': None,\n"," 'separate_test_bond_descriptors_path': None,\n"," 'separate_test_constraints_path': None,\n"," 'separate_test_features_path': None,\n"," 'separate_test_path': None,\n"," 'separate_test_phase_features_path': None,\n"," 'separate_val_atom_descriptors_path': None,\n"," 'separate_val_bond_descriptors_path': None,\n"," 'separate_val_constraints_path': None,\n"," 'separate_val_features_path': None,\n"," 'separate_val_path': None,\n"," 'separate_val_phase_features_path': None,\n"," 'shared_atom_bond_ffn': True,\n"," 'show_individual_scores': False,\n"," 'smiles_columns': ['SMILES'],\n"," 'spectra_activation': 'exp',\n"," 'spectra_phase_mask_path': None,\n"," 'spectra_target_floor': 1e-08,\n"," 'split_key_molecule': 0,\n"," 'split_sizes': [0.8, 0.1, 0.1],\n"," 'split_type': 'random',\n"," 'target_columns': None,\n"," 'target_weights': None,\n"," 'task_names': ['IS', 'DV'],\n"," 'test': False,\n"," 'test_fold_index': None,\n"," 'train_data_size': None,\n"," 'undirected': False,\n"," 'use_input_features': True,\n"," 'val_fold_index': None,\n"," 'warmup_epochs': 2.0,\n"," 'weights_ffn_num_layers': 2}\n","Setting molecule featurization parameters to default.\n","Loading data\n","1525it [00:00, 146479.35it/s]\n","100%|██████████| 1525/1525 [00:00<00:00, 83455.94it/s]\n","100%|██████████| 1525/1525 [00:00<00:00, 106793.90it/s]\n","Number of tasks = 2\n","Fold 0\n","Splitting data with seed 0\n","1525it [00:00, 231197.63it/s]\n","Total size = 1,525 | train size = 1,220 | val size = 152 | test size = 153\n","Fitting scaler\n","Building model 0\n","MoleculeModel(\n"," (encoder): MPN(\n"," (encoder): ModuleList(\n"," (0): MPNEncoder(\n"," (dropout): Dropout(p=0.0, inplace=False)\n"," (act_func): ReLU()\n"," (W_i): Linear(in_features=147, out_features=300, bias=False)\n"," (W_h): Linear(in_features=300, out_features=300, bias=False)\n"," (W_o): Linear(in_features=433, out_features=300, bias=True)\n"," )\n"," )\n"," )\n"," (readout): Sequential(\n"," (0): Dropout(p=0.0, inplace=False)\n"," (1): Linear(in_features=323, out_features=300, bias=True)\n"," (2): ReLU()\n"," (3): Dropout(p=0.0, inplace=False)\n"," (4): Linear(in_features=300, out_features=2, bias=True)\n"," )\n",")\n","Number of parameters = 362,402\n","Moving model to cuda\n"," 0%| | 0/100 [00:00 test r2 = 0.705395\n","\tSeed 0 ==> test mse = 1.088389\n","\tSeed 0 ==> test rmse = 1.043259\n","\tSeed 0 ==> test mae = 0.775628\n","Overall test r2 = 0.705395 +/- 0.000000\n","Overall test mse = 1.088389 +/- 0.000000\n","Overall test rmse = 1.043259 +/- 0.000000\n","Overall test mae = 0.775628 +/- 0.000000\n","Elapsed time = 0:03:32\n"]}],"source":["# 回归任务，标签为撞击感度和爆速。100epoch\n","\n","arguments = [\n"," '--data_path', '/content/drive/MyDrive/EM_sensitivity_0430/reg_IS_DV.csv',\n"," '--dataset_type', 'regression',\n"," '--metric', 'r2',\n"," '--extra_metrics', 'mse','rmse','mae',\n"," '--features_path', '/content/drive/MyDrive/EM_sensitivity_0430/descriptor_08.csv',\n"," '--save_dir', '/content/drive/MyDrive/EM_sensitivity_0430/reg_IS_DV_100_epoch',\n"," '--epochs', '100',\n"," '--save_smiles_splits'\n","]\n","\n","args = chemprop.args.TrainArgs().parse_args(arguments)\n","mean_score, std_score = chemprop.train.cross_validate(args=args, train_func=chemprop.train.run_training)"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"oIJWC6M-qKt5"},"outputs":[],"source":["# 分类任务，标签为撞击感度和爆速。100epoch\n","\n","arguments = [\n"," '--data_path', '/content/drive/MyDrive/EM_sensitivity_0430/reg_IS_DV_V2.csv',\n"," '--dataset_type', 'classification',\n"," '--metric', 'accuracy',\n"," '--extra_metrics', 'auc','f1','binary_cross_entropy','prc-auc','mcc',\n"," # '--num_folds','1',\n"," '--features_path', '/content/drive/MyDrive/EM_sensitivity_0430/descriptor_08.csv',\n"," '--save_dir', '/content/drive/MyDrive/EM_sensitivity_0430/reg_IS_DV_100_epoch',\n"," '--epochs', '100',\n"," # '--save_smiles_splits'\n","]\n","args = chemprop.args.TrainArgs().parse_args(arguments)"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"collapsed":true,"id":"6rQ6_1jBqKlZ","outputId":"2bf635a5-252e-41b8-fda6-b33f7ee94b76"},"outputs":[{"name":"stderr","output_type":"stream","text":["Command line\n","python /usr/local/lib/python3.10/dist-packages/colab_kernel_launcher.py -f /root/.local/share/jupyter/runtime/kernel-db6b6b1b-f9d9-49a5-9226-8a81eceac7eb.json\n","Args\n","{'activation': 'ReLU',\n"," 'adding_bond_types': True,\n"," 'adding_h': False,\n"," 'aggregation': 'mean',\n"," 'aggregation_norm': 100,\n"," 'atom_constraints': [],\n"," 'atom_descriptor_scaling': True,\n"," 'atom_descriptors': None,\n"," 'atom_descriptors_path': None,\n"," 'atom_descriptors_size': 0,\n"," 'atom_features_size': 0,\n"," 'atom_messages': False,\n"," 'atom_targets': [],\n"," 'batch_size': 50,\n"," 'bias': False,\n"," 'bias_solvent': False,\n"," 'bond_constraints': [],\n"," 'bond_descriptor_scaling': True,\n"," 'bond_descriptors': None,\n"," 'bond_descriptors_path': None,\n"," 'bond_descriptors_size': 0,\n"," 'bond_features_size': 0,\n"," 'bond_targets': [],\n"," 'cache_cutoff': 10000,\n"," 'checkpoint_dir': None,\n"," 'checkpoint_frzn': None,\n"," 'checkpoint_path': None,\n"," 'checkpoint_paths': None,\n"," 'class_balance': False,\n"," 'config_path': None,\n"," 'constraints_path': None,\n"," 'crossval_index_dir': None,\n"," 'crossval_index_file': None,\n"," 'crossval_index_sets': None,\n"," 'cuda': True,\n"," 'data_path': '/content/drive/MyDrive/EM_sensitivity_0430/cla_IS_density_V2.csv',\n"," 'data_weights_path': None,\n"," 'dataset_type': 'classification',\n"," 'depth': 3,\n"," 'depth_solvent': 3,\n"," 'device': device(type='cuda'),\n"," 'dropout': 0.0,\n"," 'empty_cache': False,\n"," 'ensemble_size': 1,\n"," 'epochs': 100,\n"," 'evidential_regularization': 0,\n"," 'explicit_h': False,\n"," 'extra_metrics': ['auc', 'f1', 'binary_cross_entropy', 'prc-auc', 'mcc'],\n"," 'features_generator': None,\n"," 'features_only': False,\n"," 'features_path': ['/content/drive/MyDrive/EM_sensitivity_0430/descriptor_08.csv'],\n"," 'features_scaling': True,\n"," 'features_size': None,\n"," 'ffn_hidden_size': 300,\n"," 'ffn_num_layers': 2,\n"," 'final_lr': 0.0001,\n"," 'folds_file': None,\n"," 'freeze_first_only': False,\n"," 'frzn_ffn_layers': 0,\n"," 'gpu': None,\n"," 'grad_clip': None,\n"," 'hidden_size': 300,\n"," 'hidden_size_solvent': 300,\n"," 'ignore_columns': None,\n"," 'init_lr': 0.0001,\n"," 'is_atom_bond_targets': False,\n"," 'keeping_atom_map': False,\n"," 'log_frequency': 10,\n"," 'loss_function': 'binary_cross_entropy',\n"," 'max_data_size': None,\n"," 'max_lr': 0.001,\n"," 'metric': 'accuracy',\n"," 'metrics': ['accuracy', 'auc', 'f1', 'binary_cross_entropy', 'prc-auc', 'mcc'],\n"," 'minimize_score': False,\n"," 'mpn_shared': False,\n"," 'multiclass_num_classes': 3,\n"," 'no_adding_bond_types': False,\n"," 'no_atom_descriptor_scaling': False,\n"," 'no_bond_descriptor_scaling': False,\n"," 'no_cache_mol': False,\n"," 'no_cuda': False,\n"," 'no_features_scaling': False,\n"," 'no_shared_atom_bond_ffn': False,\n"," 'num_folds': 1,\n"," 'num_lrs': 1,\n"," 'num_tasks': 2,\n"," 'num_workers': 8,\n"," 'number_of_molecules': 1,\n"," 'overwrite_default_atom_features': False,\n"," 'overwrite_default_bond_features': False,\n"," 'phase_features_path': None,\n"," 'pytorch_seed': 0,\n"," 'quiet': False,\n"," 'reaction': False,\n"," 'reaction_mode': 'reac_diff',\n"," 'reaction_solvent': False,\n"," 'resume_experiment': False,\n"," 'save_dir': '/content/drive/MyDrive/EM_sensitivity_0430/cla_IS_density_100_epoch',\n"," 'save_preds': False,\n"," 'save_smiles_splits': False,\n"," 'seed': 0,\n"," 'separate_test_atom_descriptors_path': None,\n"," 'separate_test_bond_descriptors_path': None,\n"," 'separate_test_constraints_path': None,\n"," 'separate_test_features_path': None,\n"," 'separate_test_path': None,\n"," 'separate_test_phase_features_path': None,\n"," 'separate_val_atom_descriptors_path': None,\n"," 'separate_val_bond_descriptors_path': None,\n"," 'separate_val_constraints_path': None,\n"," 'separate_val_features_path': None,\n"," 'separate_val_path': None,\n"," 'separate_val_phase_features_path': None,\n"," 'shared_atom_bond_ffn': True,\n"," 'show_individual_scores': False,\n"," 'smiles_columns': ['SMILES'],\n"," 'spectra_activation': 'exp',\n"," 'spectra_phase_mask_path': None,\n"," 'spectra_target_floor': 1e-08,\n"," 'split_key_molecule': 0,\n"," 'split_sizes': [0.8, 0.1, 0.1],\n"," 'split_type': 'random',\n"," 'target_columns': None,\n"," 'target_weights': None,\n"," 'task_names': ['IS', 'density'],\n"," 'test': False,\n"," 'test_fold_index': None,\n"," 'train_data_size': None,\n"," 'undirected': False,\n"," 'use_input_features': True,\n"," 'val_fold_index': None,\n"," 'warmup_epochs': 2.0,\n"," 'weights_ffn_num_layers': 2}\n","Setting molecule featurization parameters to default.\n","Loading data\n","1525it [00:00, 99564.37it/s]\n","100%|██████████| 1525/1525 [00:00<00:00, 52739.18it/s]\n","100%|██████████| 1525/1525 [00:00<00:00, 2738.39it/s]\n","Number of tasks = 2\n","Fold 0\n","Splitting data with seed 0\n","Class sizes\n","IS 0: 32.20%, 1: 67.80%\n","density 0: 48.00%, 1: 52.00%\n","Total size = 1,525 | train size = 1,220 | val size = 152 | test size = 153\n","Building model 0\n","MoleculeModel(\n"," (sigmoid): Sigmoid()\n"," (encoder): MPN(\n"," (encoder): ModuleList(\n"," (0): MPNEncoder(\n"," (dropout): Dropout(p=0.0, inplace=False)\n"," (act_func): ReLU()\n"," (W_i): Linear(in_features=147, out_features=300, bias=False)\n"," (W_h): Linear(in_features=300, out_features=300, bias=False)\n"," (W_o): Linear(in_features=433, out_features=300, bias=True)\n"," )\n"," )\n"," )\n"," (readout): Sequential(\n"," (0): Dropout(p=0.0, inplace=False)\n"," (1): Linear(in_features=323, out_features=300, bias=True)\n"," (2): ReLU()\n"," (3): Dropout(p=0.0, inplace=False)\n"," (4): Linear(in_features=300, out_features=2, bias=True)\n"," )\n",")\n","Number of parameters = 362,402\n","Moving model to cuda\n"," 0%| | 0/100 [00:00 test accuracy = 0.947712\n","\tSeed 0 ==> test auc = 0.989908\n","\tSeed 0 ==> test f1 = 0.949367\n","\tSeed 0 ==> test binary_cross_entropy = 0.124764\n","\tSeed 0 ==> test prc-auc = 0.991765\n","\tSeed 0 ==> test mcc = 0.895313\n","Overall test accuracy = 0.947712 +/- 0.000000\n","Overall test auc = 0.989908 +/- 0.000000\n","Overall test f1 = 0.949367 +/- 0.000000\n","Overall test binary_cross_entropy = 0.124764 +/- 0.000000\n","Overall test prc-auc = 0.991765 +/- 0.000000\n","Overall test mcc = 0.895313 +/- 0.000000\n","Elapsed time = 0:03:28\n"]}],"source":["# 分类任务，标签为密度。100epoch\n","\n","arguments = [\n"," '--data_path', '/content/drive/MyDrive/EM_sensitivity_0430/cla_density_V2.csv',\n"," '--dataset_type', 'classification',\n"," '--metric', 'accuracy',\n"," '--extra_metrics', 'auc','f1','binary_cross_entropy','prc-auc','mcc',\n"," # '--num_folds','1',\n"," '--features_path', '/content/drive/MyDrive/EM_sensitivity_0430/descriptor_08.csv',\n"," '--save_dir', '/content/drive/MyDrive/EM_sensitivity_0430/cla_density_100_epoch',\n"," '--epochs', '100',\n"," # '--save_smiles_splits'\n","]\n","args = chemprop.args.TrainArgs().parse_args(arguments)\n","mean_score, std_score = chemprop.train.cross_validate(args=args, train_func=chemprop.train.run_training)"]}],"metadata":{"colab":{"provenance":[],"gpuType":"V28","mount_file_id":"1j0Uky2Sf6b9H7OyDomHjnSL6eOFCf329","authorship_tag":"ABX9TyNi+npsZD2E8NY3iCbMvzGI"},"kernelspec":{"display_name":"Python 3","name":"python3"},"language_info":{"name":"python"},"accelerator":"TPU"},"nbformat":4,"nbformat_minor":0}