result_analyzer_2020.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "log_file = \"compare_with_ic3ref_hwmcc2020.csv\"\n",
    "\n",
    "# Set display options to avoid truncation\n",
    "pd.set_option(\"display.max_columns\", None)\n",
    "pd.set_option(\"display.expand_frame_repr\", False)\n",
    "pd.set_option(\"max_colwidth\", None)\n",
    "pd.set_option(\"display.max_rows\", None)\n",
    "\n",
    "# read csv table from log/compare_with_ic3ref.csv\n",
    "df = pd.read_csv(f\"log/{log_file}\")\n",
    "\n",
    "# Strip out the spaces from column names\n",
    "df.columns = df.columns.str.strip()\n",
    "\n",
    "# Filter and sort dataframe\n",
    "if \"ic3ref\" in log_file:\n",
    "    df = df[(df[\"NN-IC3-bg\"] == 1) & (df[\"IC3ref-bg\"] == 1) | (df[\"NN-IC3-bg\"] == 0) & (df[\"IC3ref-bg\"] == 0)]\n",
    "    df = df.sort_values(by=[\"case name\", \"NN-IC3 Time\"])\n",
    "    # Create a new dataframe that keeps the first duplicate\n",
    "    df_last = df.copy()\n",
    "    df_last.drop_duplicates(subset=\"case name\", keep=\"last\", inplace=True)\n",
    "    df_last = df_last.reset_index(drop=True)\n",
    "    #df.drop_duplicates(subset=\"case name\", keep=\"first\", inplace=True)\n",
    "    df.drop_duplicates(subset=\"case name\", keep=\"first\", inplace=True)\n",
    "    df = df.reset_index(drop=True)\n",
    "\n",
    "# Time scatter plot\n",
    "plt.scatter(df[\"NN-IC3 Time\"], df[\"IC3ref Time\"], color=\"blue\", label=\"Data Points\")\n",
    "plt.xlabel(\"NN-IC3 Time\")\n",
    "plt.ylabel(\"IC3ref Time\")\n",
    "plt.title(\"Scatter Plot of NN-IC3 Time and IC3ref Time\")\n",
    "\n",
    "# Apply the log scale\n",
    "plt.xscale(\"log\")\n",
    "plt.yscale(\"log\")\n",
    "\n",
    "# Set the axis limits\n",
    "plt.xlim(0.1, 1000)\n",
    "plt.ylim(0.1, 1000)\n",
    "\n",
    "# Diagonal line values\n",
    "diag_min = 0.1\n",
    "diag_max = 1000\n",
    "\n",
    "# Plot the diagonal line\n",
    "plt.plot([diag_min, diag_max], [diag_min, diag_max], color=\"red\", linestyle=\"--\", label=\"Central Line\")\n",
    "plt.legend()\n",
    "plt.show()\n",
    "\n",
    "# Frame scatter plot\n",
    "plt.scatter(df[\"NN-IC3 Frame\"], df[\"IC3ref Frame\"], color=\"blue\", label=\"Data Points\")\n",
    "plt.xlabel(\"NN-IC3 Frame\")\n",
    "plt.ylabel(\"IC3ref Frame\")\n",
    "plt.title(\"Scatter Plot of NN-IC3 Frame and IC3ref Frame\")\n",
    "\n",
    "# Apply the log scale\n",
    "plt.xscale(\"log\")\n",
    "plt.yscale(\"log\")\n",
    "\n",
    "# Set the axis limits\n",
    "plt.xlim(1, 1000)\n",
    "plt.ylim(1, 1000)\n",
    "\n",
    "# Plot the diagonal line\n",
    "plt.plot([diag_min, diag_max], [diag_min, diag_max], color=\"red\", linestyle=\"--\", label=\"Central Line\")\n",
    "plt.legend()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import os\n",
    "\n",
    "def read_clauses(file_path):\n",
    "    with open(file_path, 'r') as f:\n",
    "        lines = [line.strip() for line in f.readlines() if not line.startswith('unsat')]\n",
    "        clauses = [set(map(int, line.split())) for line in lines]\n",
    "    return clauses\n",
    "\n",
    "def coverage_ratio(inv_cti_clauses, predicted_clauses):\n",
    "    # inv_cti_clauses: list of clauses that predicted before filtering\n",
    "    # predicted_clauses: list of clauses that predicted after filtering\n",
    "    \n",
    "    found_clauses = 0\n",
    "    total_clauses = len(inv_cti_clauses)\n",
    "\n",
    "    if total_clauses == 0:\n",
    "        return 0\n",
    "\n",
    "    for inv_cti_clause in inv_cti_clauses:\n",
    "        inv_cti_clause_sorted = sorted(inv_cti_clause)\n",
    "        for predicted_clause in predicted_clauses:\n",
    "            if inv_cti_clause_sorted == sorted(predicted_clause):\n",
    "                found_clauses += 1\n",
    "                break\n",
    "    \n",
    "    # equation: (amount of found clauses that converged) / (total amount of clauses)\n",
    "    return found_clauses / total_clauses\n",
    "\n",
    "base_path = \"/data/guangyuh/coding_env/AIG2INV/AIG2INV_main/case4comp/dataset_hwmcc2020_all_only_unsat_ic3ref_no_simplification_0-38_comp/\"\n",
    "cases = os.listdir(base_path)\n",
    "\n",
    "results = []\n",
    "\n",
    "for case in cases:\n",
    "    if not os.path.exists(os.path.join(base_path, case, f\"{case}_inv_CTI_predicted.cnf\")):\n",
    "        continue\n",
    "    inv_cti_path = os.path.join(base_path, case, f\"{case}_inv_CTI_predicted.cnf\")\n",
    "    predicted_path = os.path.join(base_path, case, f\"{case}_predicted_clauses_after_filtering.cnf\")\n",
    "\n",
    "    inv_cti_clauses = read_clauses(inv_cti_path)\n",
    "    predicted_clauses = read_clauses(predicted_path)\n",
    "\n",
    "    ratio = coverage_ratio(inv_cti_clauses, predicted_clauses)\n",
    "    results.append({\"case\": case, \"coverage_ratio\": ratio})\n",
    "\n",
    "results_df = pd.DataFrame(results)\n",
    "\n",
    "# Existing DataFrame\n",
    "existing_df = df\n",
    "\n",
    "# Merge DataFrames on the 'case' column\n",
    "merged_df = existing_df.merge(results_df, left_on='case name', right_on='case')\n",
    "\n",
    "# Drop the 'case' column as it's redundant\n",
    "merged_df.drop(columns=['case'], inplace=True)\n",
    "\n",
    "# Rename the 'coverage_ratio' column if necessary\n",
    "merged_df.rename(columns={'coverage_ratio': 'passing ratio'}, inplace=True)\n",
    "\n",
    "# Calculate the time speed up and add it to the DataFrame\n",
    "merged_df['time speed up'] = merged_df['IC3ref Time'] / merged_df['NN-IC3 Time']\n",
    "\n",
    "# Scatter plot\n",
    "plt.scatter(merged_df['passing ratio'], merged_df['time speed up'],alpha=0.5)\n",
    "plt.xlabel('Passing Ratio')\n",
    "plt.ylabel('Time Speed Up')\n",
    "plt.title('Scatter Plot of Time Speed Up vs. Passing Ratio')\n",
    "\n",
    "# Set y-axis range\n",
    "plt.ylim(0, 10)\n",
    "\n",
    "# No speed up line\n",
    "x = [0, max(merged_df['passing ratio'])]\n",
    "y = [1, 1]\n",
    "plt.plot(x, y, label='No Speed Up', linestyle='--', color='red')\n",
    "\n",
    "plt.legend()\n",
    "plt.show()\n",
    "\n",
    "plt.show()\n",
    "\n",
    "#print(merged_df)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import os\n",
    "\n",
    "def read_clauses(file_path):\n",
    "    with open(file_path, 'r') as f:\n",
    "        lines = [line.strip() for line in f.readlines() if not line.startswith('unsat')]\n",
    "        clauses = [set(map(int, line.split())) for line in lines]\n",
    "    return clauses\n",
    "\n",
    "def coverage_ratio(inv_cti_clauses, ground_truth_clauses):\n",
    "    # inv_cti_clauses: list of clauses that predicted before filtering\n",
    "    # ground_truth_clauses: list of clauses that predicted after filtering\n",
    "    \n",
    "    found_clauses = 0\n",
    "    total_clauses = len(inv_cti_clauses)\n",
    "\n",
    "    if total_clauses == 0:\n",
    "        return 0\n",
    "\n",
    "    for inv_cti_clause in inv_cti_clauses:\n",
    "        inv_cti_clause_sorted = sorted(inv_cti_clause)\n",
    "        for gt_clause in ground_truth_clauses:\n",
    "            if inv_cti_clause_sorted == sorted(gt_clause):\n",
    "                found_clauses += 1\n",
    "                break\n",
    "\n",
    "    # equation: (amount of found clauses that converged) / (total amount of clauses)\n",
    "    return found_clauses / total_clauses\n",
    "\n",
    "\n",
    "base_path = \"/data/guangyuh/coding_env/AIG2INV/AIG2INV_main/case4comp/dataset_hwmcc2020_all_only_unsat_ic3ref_no_simplification_0-38_comp/\"\n",
    "ground_truth_base_path = \"/data/guangyuh/coding_env/AIG2INV/AIG2INV_main/ground_truth/hwmcc20-ic3ref_7200-result\"\n",
    "cases = os.listdir(base_path)\n",
    "\n",
    "results = []\n",
    "\n",
    "for case in cases:\n",
    "    if not os.path.exists(os.path.join(base_path, case, f\"{case}_predicted_clauses_after_filtering.cnf\")):\n",
    "        continue\n",
    "    ground_truth_path  = os.path.join(ground_truth_base_path, case, \"inv.cnf\")\n",
    "    inv_cti_path  = os.path.join(base_path, case, f\"{case}_predicted_clauses_after_filtering.cnf\")\n",
    "\n",
    "    inv_cti_clauses = read_clauses(inv_cti_path)\n",
    "    ground_truth_clauses = read_clauses(ground_truth_path)\n",
    "\n",
    "    ratio = coverage_ratio(inv_cti_clauses, ground_truth_clauses)\n",
    "    results.append({\"case\": case, \"coverage_ratio\": ratio})\n",
    "\n",
    "results_df = pd.DataFrame(results)\n",
    "\n",
    "# Existing DataFrame\n",
    "existing_df = df\n",
    "\n",
    "# Merge DataFrames on the 'case' column\n",
    "merged_df = existing_df.merge(results_df, left_on='case name', right_on='case')\n",
    "\n",
    "# Drop the 'case' column as it's redundant\n",
    "merged_df.drop(columns=['case'], inplace=True)\n",
    "\n",
    "# Rename the 'coverage_ratio' column if necessary\n",
    "merged_df.rename(columns={'coverage_ratio': 'prefect prediction ratio'}, inplace=True)\n",
    "\n",
    "# Calculate the time speed up and add it to the DataFrame\n",
    "merged_df['time speed up'] = merged_df['IC3ref Time'] / merged_df['NN-IC3 Time']\n",
    "\n",
    "# Scatter plot\n",
    "plt.scatter(merged_df['prefect prediction ratio'], merged_df['time speed up'],alpha=0.5)\n",
    "plt.xlabel('Prefect Prediction Ratio')\n",
    "plt.ylabel('Speed Up')\n",
    "plt.title('Scatter Plot of Time Speed Up vs. Prefect Prediction Ratio')\n",
    "\n",
    "# Set y-axis range\n",
    "plt.ylim(0, 10)\n",
    "\n",
    "# No speed up line\n",
    "\n",
    "x = [0, max(merged_df['prefect prediction ratio'])]\n",
    "y = [1, 1]\n",
    "plt.plot(x, y, label='No Speed Up', linestyle='--', color='red')\n",
    "\n",
    "plt.legend()\n",
    "plt.show()\n",
    "\n",
    "plt.show()\n",
    "\n",
    "#print(merged_df)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10.819270625233406\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import numpy as np\n",
    "\n",
    "def calculate_speedup_factor(df, x):\n",
    "    # Filter cases where IC3ref Time is greater than x seconds\n",
    "    #df_filtered = df[ (df[\"IC3ref Time\"] > x) & (df[\"IC3ref Time\"] < 10000)]\n",
    "    df_filtered = df[df[\"IC3ref Time\"] > x]\n",
    "    # Calculate the accumulated time for IC3ref and NN-IC3\n",
    "    accumulated_ic3ref_time = df_filtered[\"IC3ref Time\"].sum()\n",
    "    accumulated_nn_ic3_time = df_filtered[\"NN-IC3 Time\"].sum()\n",
    "\n",
    "    # Calculate the speedup factor for the accumulated times\n",
    "    #print(f\"accumulated_ic3ref_time is {accumulated_ic3ref_time}, accumulated_nn_ic3_time is {accumulated_nn_ic3_time}\")\n",
    "    speedup_factor = accumulated_ic3ref_time / accumulated_nn_ic3_time\n",
    "\n",
    "    return speedup_factor\n",
    "\n",
    "# Define the range of x values (thresholds)\n",
    "x_values = np.arange(0, 1000, 10)\n",
    "\n",
    "# Calculate the speedup factor for each x value\n",
    "speedup_factors = [calculate_speedup_factor(df, x) for x in x_values]\n",
    "print(speedup_factors[-1])\n",
    "\n",
    "# Create the plot\n",
    "plt.plot(x_values, speedup_factors, color=\"blue\", label=\"NN-IC3 Integraton\")\n",
    "plt.xlabel(\"IC3ref Time Threshold (seconds)\")\n",
    "plt.ylabel(\"Speed up\")\n",
    "#plt.title(\"Speedup Factor by IC3ref Time Threshold\")\n",
    "plt.axhline(y=1, color=\"red\", linestyle=\"--\", label=\"No Speedup Line\")\n",
    "plt.legend()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import numpy as np\n",
    "\n",
    "def calculate_reduce_factor(df, x):\n",
    "    # Filter cases where IC3ref Time is greater than x seconds\n",
    "    #df_filtered = df[ (df[\"IC3ref Time\"] > x) & (df[\"ABC Time\"] < 20000) ]\n",
    "    df_filtered = df[ df[\"IC3ref Time\"]> x ]\n",
    "    # Calculate the accumulated time for IC3ref and NN-IC3\n",
    "    accumulated_ic3ref_frame = df_filtered[\"IC3ref Frame\"].sum()\n",
    "    accumulated_nn_ic3_frame = df_filtered[\"NN-IC3 Frame\"].sum()\n",
    "\n",
    "    # Calculate the speedup factor for the accumulated times\n",
    "    reduce_factor = accumulated_ic3ref_frame / accumulated_nn_ic3_frame\n",
    "\n",
    "    return reduce_factor\n",
    "\n",
    "# Define the range of x values (thresholds)\n",
    "#x_values = np.arange(0, 230, 10)\n",
    "x_values = np.arange(0, 1000, 50)\n",
    "#x_values = np.arange(0, 3500, 10)\n",
    "\n",
    "# Calculate the speedup factor for each x value\n",
    "reduction_factors = [calculate_reduce_factor(df, x) for x in x_values]\n",
    "\n",
    "# Create the plot\n",
    "plt.plot(x_values, reduction_factors, color=\"blue\", label=\"NN-IC3 Integraton\")\n",
    "plt.xlabel(\"IC3ref Time Threshold (seconds)\")\n",
    "plt.ylabel(\"Frame Reduction\")\n",
    "#plt.title(\"Speedup Factor by ABC Time Threshold\")\n",
    "plt.axhline(y=1, color=\"red\", linestyle=\"--\", label=\"No Reduction Line\")\n",
    "plt.legend()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import fileinput\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "log_file = \"compare_with_abc_hwmcc2020.csv\"\n",
    "\n",
    "# Set display options to avoid truncation\n",
    "pd.set_option(\"display.max_columns\", None)\n",
    "pd.set_option(\"display.expand_frame_repr\", False)\n",
    "pd.set_option(\"max_colwidth\", None)\n",
    "pd.set_option(\"display.max_rows\", None)\n",
    "\n",
    "# read csv table from log/compare_with_ic3ref.csv\n",
    "df = pd.read_csv(f\"log/{log_file}\")\n",
    "\n",
    "# Strip out the spaces from column names\n",
    "df.columns = df.columns.str.strip()\n",
    "\n",
    "# Filter and sort dataframe\n",
    "if \"abc\" in log_file:\n",
    "    #df = df[(df[\"NN-IC3-bg\"] == 1) & (df[\"IC3ref-bg\"] == 1) | (df[\"NN-IC3-bg\"] == 0) & (df[\"IC3ref-bg\"] == 0)]\n",
    "    df = df.sort_values(by=[\"case name\", \"NN-ABC Time\"])\n",
    "    # Create a new dataframe that keeps the first duplicate\n",
    "    df_last = df.copy()\n",
    "    df_last.drop_duplicates(subset=\"case name\", keep=\"last\", inplace=True)\n",
    "    df_last = df_last.reset_index(drop=True)\n",
    "    #df.drop_duplicates(subset=\"case name\", keep=\"first\", inplace=True)\n",
    "    df.drop_duplicates(subset=\"case name\", keep=\"first\", inplace=True)\n",
    "    df = df.reset_index(drop=True)\n",
    "\n",
    "# Create scatter plot for time\n",
    "plt.scatter(df[\"NN-ABC Time\"], df[\"ABC Time\"], color=\"blue\", label=\"Data Points\")\n",
    "plt.xlabel(\"NN-ABC Time\")\n",
    "plt.ylabel(\"ABC Time\")\n",
    "plt.title(\"Scatter Plot of NN-ABC Time and ABC Time\")\n",
    "plt.plot(df[\"ABC Time\"], df[\"ABC Time\"], color=\"red\", linestyle=\"--\", label=\"Central Line\")\n",
    "plt.xscale(\"log\")\n",
    "plt.yscale(\"log\")\n",
    "plt.legend()\n",
    "plt.show()\n",
    "\n",
    "# Create scatter plot for frames\n",
    "plt.scatter(df[\"NN-ABC Frame\"], df[\"ABC Frame\"], color=\"blue\", label=\"Data Points\")\n",
    "plt.xlabel(\"NN-ABC Frame\")\n",
    "plt.ylabel(\"ABC Frame\")\n",
    "plt.title(\"Scatter Plot of NN-ABC Frame and ABC Frame\")\n",
    "plt.plot(df[\"ABC Frame\"], df[\"ABC Frame\"], color=\"red\", linestyle=\"--\", label=\"Central Line\")\n",
    "plt.xscale(\"log\")\n",
    "plt.yscale(\"log\")\n",
    "plt.legend()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import os\n",
    "\n",
    "def read_clauses(file_path):\n",
    "    with open(file_path, 'r') as f:\n",
    "        lines = [line.strip() for line in f.readlines() if not line.startswith('unsat')]\n",
    "        clauses = [set(map(int, line.split())) for line in lines]\n",
    "    return clauses\n",
    "\n",
    "def coverage_ratio(inv_cti_clauses, predicted_clauses):\n",
    "    # inv_cti_clauses: list of clauses that predicted before filtering\n",
    "    # predicted_clauses: list of clauses that predicted after filtering\n",
    "    \n",
    "    found_clauses = 0\n",
    "    total_clauses = len(inv_cti_clauses)\n",
    "\n",
    "    if total_clauses == 0:\n",
    "        return 0\n",
    "\n",
    "    for inv_cti_clause in inv_cti_clauses:\n",
    "        inv_cti_clause_sorted = sorted(inv_cti_clause)\n",
    "        for predicted_clause in predicted_clauses:\n",
    "            if inv_cti_clause_sorted == sorted(predicted_clause):\n",
    "                found_clauses += 1\n",
    "                break\n",
    "    \n",
    "    # equation: (amount of found clauses that converged) / (total amount of clauses)\n",
    "    return found_clauses / total_clauses\n",
    "\n",
    "base_path = \"/data/guangyuh/coding_env/AIG2INV/AIG2INV_main/case4comp/dataset_hwmcc2020_all_only_unsat_abc_no_simplification_0-38_comp/\"\n",
    "cases = os.listdir(base_path)\n",
    "\n",
    "results = []\n",
    "\n",
    "for case in cases:\n",
    "    if not os.path.exists(os.path.join(base_path, case, f\"{case}_inv_CTI_predicted.cnf\")):\n",
    "        continue\n",
    "    inv_cti_path = os.path.join(base_path, case, f\"{case}_inv_CTI_predicted.cnf\")\n",
    "    predicted_path = os.path.join(base_path, case, f\"{case}_predicted_clauses_after_filtering.cnf\")\n",
    "\n",
    "    inv_cti_clauses = read_clauses(inv_cti_path)\n",
    "    predicted_clauses = read_clauses(predicted_path)\n",
    "\n",
    "    ratio = coverage_ratio(inv_cti_clauses, predicted_clauses)\n",
    "    results.append({\"case\": case, \"coverage_ratio\": ratio})\n",
    "\n",
    "results_df = pd.DataFrame(results)\n",
    "\n",
    "# Existing DataFrame\n",
    "existing_df = df\n",
    "\n",
    "# Merge DataFrames on the 'case' column\n",
    "merged_df = existing_df.merge(results_df, left_on='case name', right_on='case')\n",
    "\n",
    "# Drop the 'case' column as it's redundant\n",
    "merged_df.drop(columns=['case'], inplace=True)\n",
    "\n",
    "# Rename the 'coverage_ratio' column if necessary\n",
    "merged_df.rename(columns={'coverage_ratio': 'passing ratio'}, inplace=True)\n",
    "\n",
    "# Calculate the time speed up and add it to the DataFrame\n",
    "merged_df['time speed up'] = merged_df['ABC Time'] / merged_df['NN-ABC Time']\n",
    "\n",
    "# Scatter plot\n",
    "plt.scatter(merged_df['passing ratio'], merged_df['time speed up'])\n",
    "plt.xlabel('Passing Ratio')\n",
    "plt.ylabel('Time Speed Up')\n",
    "plt.title('Scatter Plot of Time Speed Up vs. Passing Ratio')\n",
    "# No speed up line\n",
    "x = [0, max(merged_df['passing ratio'])]\n",
    "y = [1, 1]\n",
    "plt.plot(x, y, label='No Speed Up', linestyle='--', color='red')\n",
    "\n",
    "plt.legend()\n",
    "plt.show()\n",
    "\n",
    "plt.show()\n",
    "\n",
    "#print(merged_df)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import os\n",
    "\n",
    "def read_clauses(file_path):\n",
    "    with open(file_path, 'r') as f:\n",
    "        lines = [line.strip() for line in f.readlines() if not line.startswith('unsat')]\n",
    "        clauses = [set(map(int, line.split())) for line in lines]\n",
    "    return clauses\n",
    "\n",
    "def coverage_ratio(inv_cti_clauses, ground_truth_clauses):\n",
    "    # inv_cti_clauses: list of clauses that predicted before filtering\n",
    "    # ground_truth_clauses: list of clauses that predicted after filtering\n",
    "    \n",
    "    found_clauses = 0\n",
    "    total_clauses = len(inv_cti_clauses)\n",
    "\n",
    "    if total_clauses == 0:\n",
    "        return 0\n",
    "\n",
    "    for inv_cti_clause in inv_cti_clauses:\n",
    "        inv_cti_clause_sorted = sorted(inv_cti_clause)\n",
    "        for gt_clause in ground_truth_clauses:\n",
    "            if inv_cti_clause_sorted == sorted(gt_clause):\n",
    "                found_clauses += 1\n",
    "                break\n",
    "\n",
    "    # equation: (amount of found clauses that converged) / (total amount of clauses)\n",
    "    return found_clauses / total_clauses\n",
    "\n",
    "\n",
    "base_path = \"/data/guangyuh/coding_env/AIG2INV/AIG2INV_main/case4comp/dataset_hwmcc2020_all_only_unsat_abc_no_simplification_0-38_comp/\"\n",
    "ground_truth_base_path = \"/data/guangyuh/coding_env/AIG2INV/AIG2INV_main/ground_truth/hwmcc20_abc_7200_result\"\n",
    "cases = os.listdir(base_path)\n",
    "\n",
    "results = []\n",
    "\n",
    "for case in cases:\n",
    "    # if file not exist, skip\n",
    "    if not os.path.exists(os.path.join(ground_truth_base_path, case, \"inv.cnf\")):\n",
    "        continue\n",
    "    ground_truth_path  = os.path.join(ground_truth_base_path, case, \"inv.cnf\")\n",
    "    if not os.path.exists(os.path.join(base_path, case, f\"{case}_predicted_clauses_after_filtering.cnf\")):\n",
    "        continue\n",
    "    inv_cti_path  = os.path.join(base_path, case, f\"{case}_predicted_clauses_after_filtering.cnf\")\n",
    "\n",
    "    inv_cti_clauses = read_clauses(inv_cti_path)\n",
    "    ground_truth_clauses = read_clauses(ground_truth_path)\n",
    "\n",
    "    ratio = coverage_ratio(inv_cti_clauses, ground_truth_clauses)\n",
    "    results.append({\"case\": case, \"coverage_ratio\": ratio})\n",
    "\n",
    "results_df = pd.DataFrame(results)\n",
    "\n",
    "# Existing DataFrame\n",
    "existing_df = df\n",
    "\n",
    "# Merge DataFrames on the 'case' column\n",
    "merged_df = existing_df.merge(results_df, left_on='case name', right_on='case')\n",
    "\n",
    "# Drop the 'case' column as it's redundant\n",
    "merged_df.drop(columns=['case'], inplace=True)\n",
    "\n",
    "# Rename the 'coverage_ratio' column if necessary\n",
    "merged_df.rename(columns={'coverage_ratio': 'prefect prediction ratio'}, inplace=True)\n",
    "\n",
    "# Calculate the time speed up and add it to the DataFrame\n",
    "merged_df['time speed up'] = merged_df['ABC Time'] / merged_df['NN-ABC Time']\n",
    "\n",
    "# Scatter plot\n",
    "plt.scatter(merged_df['prefect prediction ratio'], merged_df['time speed up'],alpha=0.5)\n",
    "plt.xlabel('Prefect Prediction Ratio')\n",
    "plt.ylabel('Speed Up')\n",
    "plt.title('Scatter Plot of Time Speed Up vs. Prefect Prediction Ratio')\n",
    "\n",
    "# Set y-axis to log scale\n",
    "#plt.yscale('log')\n",
    "\n",
    "# Set y-axis range\n",
    "plt.ylim(0, 10)\n",
    "\n",
    "# No speed up line\n",
    "x = [0, max(merged_df['prefect prediction ratio'])]\n",
    "y = [1, 1]\n",
    "plt.plot(x, y, label='No Speed Up', linestyle='--', color='red')\n",
    "\n",
    "plt.legend()\n",
    "\n",
    "plt.show()\n",
    "\n",
    "#print(merged_df)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import numpy as np\n",
    "\n",
    "def calculate_speedup_factor(df, x):\n",
    "    # Filter cases where IC3ref Time is greater than x seconds\n",
    "    df_filtered = df[ (df[\"ABC Time\"] > x) & (df[\"ABC Time\"] < 30000) ]\n",
    "    \n",
    "    #df_filtered = df[ df[\"ABC Time\"]> x ]\n",
    "    \n",
    "    # Calculate the accumulated time for IC3ref and NN-IC3\n",
    "    accumulated_ic3_abc_time = df_filtered[\"ABC Time\"].sum()\n",
    "    accumulated_nn_ic3_time = df_filtered[\"NN-ABC Time\"].sum()\n",
    "\n",
    "    # Calculate the speedup factor for the accumulated times\n",
    "    speedup_factor = accumulated_ic3_abc_time / accumulated_nn_ic3_time\n",
    "\n",
    "    return speedup_factor\n",
    "\n",
    "# Define the range of x values (thresholds)\n",
    "#x_values = np.arange(0, 230, 10)\n",
    "x_values = np.arange(0, 1000, 10)\n",
    "#x_values = np.arange(0, 50000, 10)\n",
    "\n",
    "# Calculate the speedup factor for each x value\n",
    "speedup_factors = [calculate_speedup_factor(df, x) for x in x_values]\n",
    "\n",
    "# Create the plot\n",
    "plt.plot(x_values, speedup_factors, color=\"blue\", label=\"NN-IC3 Integraton\")\n",
    "plt.xlabel(\"ABC Time Threshold (seconds)\")\n",
    "plt.ylabel(\"Speed up\")\n",
    "#plt.title(\"Speedup Factor by ABC Time Threshold\")\n",
    "plt.axhline(y=1, color=\"red\", linestyle=\"--\", label=\"No Speedup Line\")\n",
    "plt.legend()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import numpy as np\n",
    "\n",
    "def calculate_reduce_factor(df, x):\n",
    "    # Filter cases where IC3ref Time is greater than x seconds\n",
    "    df_filtered = df[ (df[\"ABC Time\"] > x) & (df[\"ABC Time\"] < 30000) ]\n",
    "    #df_filtered = df[ df[\"ABC Time\"]> x ]\n",
    "    # Calculate the accumulated time for IC3ref and NN-IC3\n",
    "    accumulated_ic3_abc_frame = df_filtered[\"ABC Frame\"].sum()\n",
    "    accumulated_nn_ic3_frame = df_filtered[\"NN-ABC Frame\"].sum()\n",
    "\n",
    "    # Calculate the speedup factor for the accumulated times\n",
    "    reduce_factor = accumulated_ic3_abc_frame / accumulated_nn_ic3_frame\n",
    "\n",
    "    return reduce_factor\n",
    "\n",
    "# Define the range of x values (thresholds)\n",
    "#x_values = np.arange(0, 230, 10)\n",
    "x_values = np.arange(0, 1000, 10)\n",
    "#x_values = np.arange(0, 3500, 10)\n",
    "\n",
    "# Calculate the speedup factor for each x value\n",
    "reduction_factors = [calculate_reduce_factor(df, x) for x in x_values]\n",
    "\n",
    "# Create the plot\n",
    "plt.plot(x_values, reduction_factors, color=\"blue\", label=\"NN-IC3 Integraton\")\n",
    "plt.xlabel(\"ABC Time Threshold (seconds)\")\n",
    "plt.ylabel(\"Frame Reduction\")\n",
    "#plt.title(\"Speedup Factor by ABC Time Threshold\")\n",
    "plt.axhline(y=1, color=\"red\", linestyle=\"--\", label=\"No Reduction Line\")\n",
    "plt.legend()\n",
    "plt.show()\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "pytorch-gpu-aig2inv",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.7"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}