{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "from alpha_vantage.timeseries import TimeSeries\n",
    "from alpha_vantage.techindicators import TechIndicators\n",
    "import requests \n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "from datetime import datetime\n",
    "from sklearn.metrics import classification_report\n",
    "import numpy as np\n",
    "from sklearn.metrics import classification_report"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'items': '0', 'sentiment_score_definition': 'x <= -0.35: Bearish; -0.35 < x <= -0.15: Somewhat-Bearish; -0.15 < x < 0.15: Neutral; 0.15 <= x < 0.35: Somewhat_Bullish; x >= 0.35: Bullish', 'relevance_score_definition': '0 < x <= 1, with a higher score indicating higher relevance.', 'feed': []}\n"
     ]
    }
   ],
   "source": [
    "# replace the \"demo\" apikey below with your own key from https://www.alphavantage.co/support/#api-key\n",
    "url = 'https://www.alphavantage.co/query?function=NEWS_SENTIMENT&tickers=AAPL&topics=economy_macro&apikey=HKHE3U0MF6OT06XT'\n",
    "r = requests.get(url)\n",
    "data = r.json()\n",
    "\n",
    "print(data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "API Response: {'items': '0', 'sentiment_score_definition': 'x <= -0.35: Bearish; -0.35 < x <= -0.15: Somewhat-Bearish; -0.15 < x < 0.15: Neutral; 0.15 <= x < 0.35: Somewhat_Bullish; x >= 0.35: Bullish', 'relevance_score_definition': '0 < x <= 1, with a higher score indicating higher relevance.', 'feed': []}\n",
      "  items                         sentiment_score_definition  \\\n",
      "0     0  x <= -0.35: Bearish; -0.35 < x <= -0.15: Somew...   \n",
      "\n",
      "                          relevance_score_definition  \n",
      "0  0 < x <= 1, with a higher score indicating hig...  \n"
     ]
    }
   ],
   "source": [
    "# Replace with your own API key\n",
    "api_key = 'HKHE3U0MF6OT06XT'\n",
    "base_url = 'https://www.alphavantage.co/query'\n",
    "\n",
    "# Function to extract economy macro news for the past 3 years\n",
    "def get_economy_macro_news(api_key):\n",
    "    url = f'{base_url}?function=NEWS_SENTIMENT&tickers=AAPL&topics=economy_macro&apikey={api_key}'\n",
    "    \n",
    "    # Send request to Alpha Vantage API\n",
    "    r = requests.get(url)\n",
    "    data = r.json()\n",
    "    \n",
    "    # Print full API response for reference\n",
    "    print(\"API Response:\", data)\n",
    "    \n",
    "    # Extract the feed data\n",
    "    news_data = data.get('feed', [])\n",
    "    \n",
    "    if not news_data:\n",
    "        # If the feed is empty, create a DataFrame with metadata fields\n",
    "        df = pd.DataFrame({\n",
    "            'items': [data.get('items', 'N/A')],\n",
    "            'sentiment_score_definition': [data.get('sentiment_score_definition', 'N/A')],\n",
    "            'relevance_score_definition': [data.get('relevance_score_definition', 'N/A')]\n",
    "        })\n",
    "    else:\n",
    "        # If there's data, create a DataFrame from the feed\n",
    "        df = pd.DataFrame(news_data)\n",
    "\n",
    "    return df\n",
    "\n",
    "# Call the function and create the DataFrame\n",
    "news_df = get_economy_macro_news(api_key)\n",
    "\n",
    "# Print the DataFrame\n",
    "print(news_df)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                     Metadata  \\\n",
      "0                       Items   \n",
      "1  Sentiment Score Definition   \n",
      "2  Relevance Score Definition   \n",
      "\n",
      "                                             Details  \n",
      "0                                                  0  \n",
      "1  x <= -0.35: Bearish; -0.35 < x <= -0.15: Somew...  \n",
      "2  0 < x <= 1, with a higher score indicating hig...  \n"
     ]
    }
   ],
   "source": [
    "# Function to extract economy macro news for multiple tickers\n",
    "def get_economy_macro_news(api_key, tickers):\n",
    "    url = f'{base_url}?function=NEWS_SENTIMENT&tickers=AAPL&topics=economy_macro&apikey={api_key}'\n",
    "    \n",
    "    # Send request to Alpha Vantage API\n",
    "    r = requests.get(url)\n",
    "    data = r.json()\n",
    "    \n",
    "    # Extract the feed data\n",
    "    news_data = data.get('feed', [])\n",
    "    \n",
    "    if not news_data:\n",
    "        # If the feed is empty, create a structured DataFrame with metadata\n",
    "        structured_data = {\n",
    "            'Metadata': ['Items', 'Sentiment Score Definition', 'Relevance Score Definition'],\n",
    "            'Details': [\n",
    "                data.get('items', 'N/A'),\n",
    "                data.get('sentiment_score_definition', 'N/A'),\n",
    "                data.get('relevance_score_definition', 'N/A')\n",
    "            ]\n",
    "        }\n",
    "        df = pd.DataFrame(structured_data)\n",
    "    else:\n",
    "        # If there is data in the feed, create a DataFrame from the news data\n",
    "        df = pd.DataFrame(news_data)\n",
    "        \n",
    "        # Select specific columns and add a 'ticker' column to differentiate stocks\n",
    "        df = df[['title', 'summary', 'url', 'time_published', 'ticker']]\n",
    "\n",
    "    return df\n",
    "\n",
    "# List of multiple stocks you want to query\n",
    "tickers = 'AAPL,GOOGL,AMZN'\n",
    "\n",
    "# Call the function and create the DataFrame\n",
    "news_df = get_economy_macro_news(api_key, tickers)\n",
    "\n",
    "# Print the DataFrame in a structured table format\n",
    "print(news_df)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "ts = TimeSeries(api_key)\n",
    "ti = TechIndicators(api_key)\n",
    "# Get json object with the 30-min interval intraday data and another with  the call's metadata for January, 2014.\n",
    "data, meta_data = ts.get_intraday('GOOGL', month='2014-01', interval='30min')\n",
    "#Get json object with the 30-min interval simple moving average (SMA) values and another with  the call's metadata for January, 2014.\n",
    "data, meta_data = ti.get_sma('GOOGL', month='2014-01', interval='30min')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'1. Information': 'Daily Prices (open, high, low, close) and Volumes', '2. Symbol': 'MSFT', '3. Last Refreshed': '2024-12-09', '4. Output Size': 'Full size', '5. Time Zone': 'US/Eastern'}\n",
      "Total data points retrieved: 6317\n",
      "Total data points after filtering for 5 years: 1256\n",
      "            1. open  2. high    3. low  4. close   5. volume\n",
      "date                                                        \n",
      "2024-12-09   442.60   448.33  440.5000    446.02  19144388.0\n",
      "2024-12-06   442.30   446.10  441.7703    443.57  18821002.0\n",
      "2024-12-05   437.92   444.66  436.1710    442.62  21697775.0\n",
      "2024-12-04   433.03   439.67  432.6300    437.42  26009429.0\n",
      "2024-12-03   429.84   432.47  427.7400    431.20  18301987.0\n"
     ]
    }
   ],
   "source": [
    "# Initialize the TimeSeries object with your API key\n",
    "ts = TimeSeries(api_key, output_format='pandas')\n",
    "\n",
    "# Retrieve the daily adjusted time series for 'full' dataset\n",
    "data, meta_data = ts.get_daily(symbol='MSFT', outputsize='full')\n",
    "\n",
    "# Print metadata to check if there are any restrictions or errors\n",
    "print(meta_data)\n",
    "\n",
    "# Check the full dataset length\n",
    "print(f\"Total data points retrieved: {len(data)}\")\n",
    "\n",
    "# Convert the index to datetime\n",
    "data.index = pd.to_datetime(data.index)\n",
    "\n",
    "# Calculate the date from 5 years ago\n",
    "five_years_ago = datetime.now() - timedelta(days=5*365)\n",
    "\n",
    "# Filter the data for the past 5 years\n",
    "filtered_data = data[data.index >= five_years_ago]\n",
    "\n",
    "# Check the length of the filtered data\n",
    "print(f\"Total data points after filtering for 5 years: {len(filtered_data)}\")\n",
    "\n",
    "# Display the first two rows of the filtered data\n",
    "\n",
    "print(filtered_data.head(5))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    " # For the default date string index behavior\n",
    "ts = TimeSeries(key='YOUR_API_KEY',output_format='pandas', indexing_type='date')\n",
    "# For the default integer index behavior\n",
    "ts = TimeSeries(key='YOUR_API_KEY',output_format='pandas', indexing_type='integer')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Pennant Pattern Examples - Bullish"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Available columns: Index(['1. open', '2. high', '3. low', '4. close', '5. volume'], dtype='object')\n",
      "            1. open  2. high   3. low  4. close   5. volume\n",
      "date                                                       \n",
      "2016-07-29    56.39    57.25  55.3800     57.10  10239664.0\n",
      "2016-07-28    56.19    56.60  56.0137     56.18   5547934.0\n",
      "2016-07-27    57.12    57.22  55.7200     56.06  11166886.0\n",
      "2016-07-26    56.17    56.92  56.0900     56.63  12862739.0\n",
      "2016-07-25    55.00    55.81  54.7600     55.68   7813211.0\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Initialize the TimeSeries object with your API key\n",
    "ts = TimeSeries(api_key, output_format='pandas')\n",
    "\n",
    "# Retrieve the daily time series data for LEH (Lehman Brothers)\n",
    "data, meta_data = ts.get_daily(symbol='NVDA', outputsize='full')\n",
    "\n",
    "# Print column names to inspect them\n",
    "print(\"Available columns:\", data.columns)\n",
    "\n",
    "# Convert the index to datetime for filtering\n",
    "data.index = pd.to_datetime(data.index)\n",
    "\n",
    "# Define the date range for 2007-2008\n",
    "start_date = '2016-01-01'\n",
    "end_date = '2016-07-31'\n",
    "\n",
    "# Filter the data for the time period between 2007 and 2008\n",
    "filtered_data = data[(data.index >= start_date) & (data.index <= end_date)]\n",
    "\n",
    "# Inspect the first few rows of the filtered data\n",
    "print(filtered_data.head())\n",
    "\n",
    "# Export as a csv file \n",
    "data.to_csv('NVDIA_data.csv', index=True)\n",
    "\n",
    "# Plot the 'close' price for the filtered data\n",
    "plt.figure(figsize=(10, 6))\n",
    "plt.plot(filtered_data.index, filtered_data['4. close'], label='Close Price')\n",
    "plt.title('NVIDIA Bullish Pennant Example (2016)')\n",
    "plt.xlabel('Date')\n",
    "plt.ylabel('Price (USD)')\n",
    "plt.grid(True)\n",
    "plt.legend()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Available columns: Index(['1. open', '2. high', '3. low', '4. close', '5. volume'], dtype='object')\n",
      "            1. open  2. high  3. low  4. close   5. volume\n",
      "date                                                      \n",
      "2020-05-29   319.25   321.15  316.47    317.94  38399532.0\n",
      "2020-05-28   316.77   323.44  315.63    318.25  33449103.0\n",
      "2020-05-27   316.14   318.71  313.09    318.11  28236274.0\n",
      "2020-05-26   323.50   324.24  316.50    316.73  31380454.0\n",
      "2020-05-22   315.77   319.23  315.35    318.89  20450754.0\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from datetime import datetime, timedelta\n",
    "# Initialize the TimeSeries object with your API key\n",
    "ts = TimeSeries(api_key, output_format='pandas')\n",
    "\n",
    "# Retrieve the daily time series data for LEH (Lehman Brothers)\n",
    "data, meta_data = ts.get_daily(symbol='AAPL', outputsize='full')\n",
    "\n",
    "# Print column names to inspect them\n",
    "print(\"Available columns:\", data.columns)\n",
    "\n",
    "# Convert the index to datetime for filtering\n",
    "data.index = pd.to_datetime(data.index)\n",
    "\n",
    "# Define the date range for 2007-2008\n",
    "start_date = '2020-03-01'\n",
    "end_date = '2020-05-31'\n",
    "\n",
    "# Filter the data for the time period between 2007 and 2008\n",
    "filtered_data = data[(data.index >= start_date) & (data.index <= end_date)]\n",
    "\n",
    "# Inspect the first few rows of the filtered data\n",
    "print(filtered_data.head())\n",
    "\n",
    "# Export as a csv file \n",
    "data.to_csv('AAPL_data.csv', index=True)\n",
    "\n",
    "# Plot the 'close' price for the filtered data\n",
    "plt.figure(figsize=(10, 6))\n",
    "plt.plot(filtered_data.index, filtered_data['4. close'], label='Close Price')\n",
    "plt.title('AAPL Bullish Pennant Example(2020)')\n",
    "plt.xlabel('Date')\n",
    "plt.ylabel('Price (USD)')\n",
    "plt.grid(True)\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Pennant Pattern Examples: Bearish "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Available columns: Index(['1. open', '2. high', '3. low', '4. close', '5. volume'], dtype='object')\n",
      "            1. open  2. high  3. low  4. close   5. volume\n",
      "date                                                      \n",
      "2001-12-31    18.45    18.93   18.06     18.11  61550500.0\n",
      "2001-12-28    18.69    18.94   18.20     18.54  46191700.0\n",
      "2001-12-27    18.42    18.71   18.21     18.49  38393400.0\n",
      "2001-12-26    18.11    18.83   18.06     18.24  36264500.0\n",
      "2001-12-24    18.23    18.50   18.00     18.11  16740600.0\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from datetime import datetime, timedelta\n",
    "# Initialize the TimeSeries object with your API key\n",
    "ts = TimeSeries(api_key, output_format='pandas')\n",
    "\n",
    "# Retrieve the daily time series data for LEH (Lehman Brothers)\n",
    "data, meta_data = ts.get_daily(symbol='CSCO', outputsize='full')\n",
    "\n",
    "# Print column names to inspect them\n",
    "print(\"Available columns:\", data.columns)\n",
    "\n",
    "# Convert the index to datetime for filtering\n",
    "data.index = pd.to_datetime(data.index)\n",
    "\n",
    "# Define the date range for 2007-2008\n",
    "start_date = '2000-01-01'\n",
    "end_date = '2001-12-31'\n",
    "\n",
    "# Filter the data for the time period between 2007 and 2008\n",
    "filtered_data = data[(data.index >= start_date) & (data.index <= end_date)]\n",
    "\n",
    "# Inspect the first few rows of the filtered data\n",
    "print(filtered_data.head())\n",
    "\n",
    "# # Export as a csv file \n",
    "# data.to_csv('AAPL_data.csv', index=True)\n",
    "\n",
    "# Plot the 'close' price for the filtered data\n",
    "plt.figure(figsize=(10, 6))\n",
    "plt.plot(filtered_data.index, filtered_data['4. close'], label='Close Price')\n",
    "plt.title('CSCO Bearish Pennant Example')\n",
    "plt.xlabel('Date')\n",
    "plt.ylabel('Price (USD)')\n",
    "plt.grid(True)\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Available columns: Index(['1. open', '2. high', '3. low', '4. close', '5. volume'], dtype='object')\n",
      "            1. open  2. high  3. low  4. close   5. volume\n",
      "date                                                      \n",
      "2015-08-31    74.27    75.55   73.30     75.24  19567271.0\n",
      "2015-08-28    74.71    75.98   74.41     75.07  18991451.0\n",
      "2015-08-27    73.87    74.89   73.01     74.85  24564440.0\n",
      "2015-08-26    70.63    72.75   69.15     72.50  34431427.0\n",
      "2015-08-25    71.31    71.44   68.20     68.71  30405356.0\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from datetime import datetime, timedelta\n",
    "# Initialize the TimeSeries object with your API key\n",
    "ts = TimeSeries(api_key, output_format='pandas')\n",
    "\n",
    "# Retrieve the daily time series data for LEH (Lehman Brothers)\n",
    "data, meta_data = ts.get_daily(symbol='XOM', outputsize='full')\n",
    "\n",
    "# Print column names to inspect them\n",
    "print(\"Available columns:\", data.columns)\n",
    "\n",
    "# Convert the index to datetime for filtering\n",
    "data.index = pd.to_datetime(data.index)\n",
    "\n",
    "# Define the date range for 2007-2008\n",
    "start_date = '2015-01-01'\n",
    "end_date = '2015-08-31'\n",
    "\n",
    "# Filter the data for the time period between 2007 and 2008\n",
    "filtered_data = data[(data.index >= start_date) & (data.index <= end_date)]\n",
    "\n",
    "# Inspect the first few rows of the filtered data\n",
    "print(filtered_data.head())\n",
    "\n",
    "# # Export as a csv file \n",
    "# data.to_csv('AAPL_data.csv', index=True)\n",
    "\n",
    "# Plot the 'close' price for the filtered data\n",
    "plt.figure(figsize=(10, 6))\n",
    "plt.plot(filtered_data.index, filtered_data['4. close'], label='Close Price')\n",
    "plt.title('Exxon Bearish Pennant Example')\n",
    "plt.xlabel('Date')\n",
    "plt.ylabel('Price (USD)')\n",
    "plt.grid(True)\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Flag Pattern Examples"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Available columns: Index(['1. open', '2. high', '3. low', '4. close', '5. volume'], dtype='object')\n",
      "            1. open  2. high  3. low  4. close   5. volume\n",
      "date                                                      \n",
      "2020-03-31   501.25   542.96  497.00    524.00  17771485.0\n",
      "2020-03-30   510.26   516.65  491.23    502.13  11998067.0\n",
      "2020-03-27   505.00   525.80  494.03    514.36  14377408.0\n",
      "2020-03-26   547.39   560.00  512.25    528.16  17422082.0\n",
      "2020-03-25   545.25   557.00  511.11    539.25  21222745.0\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from datetime import datetime, timedelta\n",
    "# Initialize the TimeSeries object with your API key\n",
    "ts = TimeSeries(api_key, output_format='pandas')\n",
    "\n",
    "# Retrieve the daily time series data for LEH (Lehman Brothers)\n",
    "data, meta_data = ts.get_daily(symbol='TSLA', outputsize='full')\n",
    "\n",
    "# Print column names to inspect them\n",
    "print(\"Available columns:\", data.columns)\n",
    "\n",
    "# Convert the index to datetime for filtering\n",
    "data.index = pd.to_datetime(data.index)\n",
    "\n",
    "# Define the date range for 2007-2008\n",
    "start_date = '2020-02-01'\n",
    "end_date = '2020-03-31'\n",
    "\n",
    "# Filter the data for the time period between 2007 and 2008\n",
    "filtered_data = data[(data.index >= start_date) & (data.index <= end_date)]\n",
    "\n",
    "# Inspect the first few rows of the filtered data\n",
    "print(filtered_data.head())\n",
    "\n",
    "# Export as a csv file \n",
    "data.to_csv('TSLA_flag_data.csv', index=True)\n",
    "\n",
    "# Plot the 'close' price for the filtered data\n",
    "plt.figure(figsize=(10, 6))\n",
    "plt.plot(filtered_data.index, filtered_data['4. close'], label='Close Price')\n",
    "plt.title('Tesla Upward Flagpole Example')\n",
    "plt.xlabel('Date')\n",
    "plt.ylabel('Price (USD)')\n",
    "plt.grid(True)\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Available columns: Index(['1. open', '2. high', '3. low', '4. close', '5. volume'], dtype='object')\n",
      "            1. open  2. high  3. low  4. close   5. volume\n",
      "date                                                      \n",
      "2008-03-31    38.18    40.00   37.78     37.91  37031600.0\n",
      "2008-03-28    38.80    39.15   37.98     38.07  38462100.0\n",
      "2008-03-27    40.01    40.38   38.61     38.64  42795300.0\n",
      "2008-03-26    40.23    40.45   39.00     39.84  51047100.0\n",
      "2008-03-25    41.65    42.25   40.70     40.97  52389000.0\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from datetime import datetime, timedelta\n",
    "# Initialize the TimeSeries object with your API key\n",
    "ts = TimeSeries(api_key, output_format='pandas')\n",
    "\n",
    "# Retrieve the daily time series data for LEH (Lehman Brothers)\n",
    "data, meta_data = ts.get_daily(symbol='BAC', outputsize='full')\n",
    "\n",
    "# Print column names to inspect them\n",
    "print(\"Available columns:\", data.columns)\n",
    "\n",
    "# Convert the index to datetime for filtering\n",
    "data.index = pd.to_datetime(data.index)\n",
    "\n",
    "# Define the date range for 2007-2008\n",
    "start_date = '2007-10-01'\n",
    "end_date = '2008-03-31'\n",
    "\n",
    "# Filter the data for the time period between 2007 and 2008\n",
    "filtered_data = data[(data.index >= start_date) & (data.index <= end_date)]\n",
    "\n",
    "# Inspect the first few rows of the filtered data\n",
    "print(filtered_data.head())\n",
    "\n",
    "# Export as a csv file \n",
    "data.to_csv('BAC_flag_data.csv', index=True)\n",
    "\n",
    "# Plot the 'close' price for the filtered data\n",
    "plt.figure(figsize=(10, 6))\n",
    "plt.plot(filtered_data.index, filtered_data['4. close'], label='Close Price')\n",
    "plt.title('Bank of America Upward Flag')\n",
    "plt.xlabel('Date')\n",
    "plt.ylabel('Price (USD)')\n",
    "plt.grid(True)\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "labels = ['Bullish', 'Slightly Bullish', 'Neutral/Mixed', 'Bearish']\n",
    "counts = [10, 2, 0, 0]\n",
    "\n",
    "plt.figure(figsize=(10, 6))\n",
    "plt.bar(labels, counts, color=['green', 'lightgreen', 'gray', 'red'])\n",
    "plt.title('Distribution of Ground Truth Labels')\n",
    "plt.xlabel('Labels')\n",
    "plt.ylabel('Number of Stocks')\n",
    "\n",
    "for i, v in enumerate(counts):\n",
    "    plt.text(i, v + 0.1, str(v), ha='center')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Available columns: Index(['1. open', '2. high', '3. low', '4. close', '5. volume'], dtype='object')\n",
      "            1. open  2. high  3. low  4. close   5. volume\n",
      "date                                                      \n",
      "2019-03-29   118.07   118.32  116.96    117.94  25399752.0\n",
      "2019-03-28   117.44   117.58  116.13    116.93  18334755.0\n",
      "2019-03-27   117.88   118.21  115.52    116.77  22733427.0\n",
      "2019-03-26   118.62   118.71  116.85    117.91  26097665.0\n",
      "2019-03-25   116.56   118.01  116.32    117.66  27067117.0\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from datetime import datetime, timedelta\n",
    "# Initialize the TimeSeries object with your API key\n",
    "ts = TimeSeries(api_key, output_format='pandas')\n",
    "\n",
    "# Retrieve the daily time series data for LEH (Lehman Brothers)\n",
    "data, meta_data = ts.get_daily(symbol='MSFT', outputsize='full')\n",
    "\n",
    "# Print column names to inspect them\n",
    "print(\"Available columns:\", data.columns)\n",
    "\n",
    "# Convert the index to datetime for filtering\n",
    "data.index = pd.to_datetime(data.index)\n",
    "\n",
    "# Define the date range for 2007-2008\n",
    "start_date = '2019-01-01'\n",
    "end_date = '2019-03-31'\n",
    "\n",
    "# Filter the data for the time period between 2007 and 2008\n",
    "filtered_data = data[(data.index >= start_date) & (data.index <= end_date)]\n",
    "\n",
    "# Inspect the first few rows of the filtered data\n",
    "print(filtered_data.head())\n",
    "\n",
    "# Export as a csv file \n",
    "data.to_csv('MSFT_downwardflag.csv', index=True)\n",
    "\n",
    "# Plot the 'close' price for the filtered data\n",
    "plt.figure(figsize=(10, 6))\n",
    "plt.plot(filtered_data.index, filtered_data['4. close'], label='Close Price')\n",
    "plt.title('Microsoft Downward Flag Pattern')\n",
    "plt.xlabel('Date')\n",
    "plt.ylabel('Price (USD)')\n",
    "plt.grid(True)\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Shape of combined data: (1258, 20)\n",
      "Number of NaN values: 0\n",
      "Shape of cleaned data: (1257, 44)\n",
      "Final shape of data: (1256, 45)\n",
      "Training set shape: (753, 23)\n",
      "Validation set shape: (251, 23)\n",
      "Testing set shape: (252, 23)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/var/folders/09/73wvt1t102l9g7hdmc9gkxzm0000gn/T/ipykernel_70358/1094137802.py:50: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  combined_data_cleaned['Target'] = combined_data_cleaned[[f'{stock}_Returns' for stock in stocks]].mean(axis=1).shift(-1)\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Script completed.\n"
     ]
    }
   ],
   "source": [
    "import yfinance as yf\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "from sklearn.model_selection import train_test_split\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# List of 20 S&P 500 stocks from various sectors\n",
    "stocks = ['AAPL', 'MSFT', 'AMZN', 'GOOGL', 'META', 'TSLA', 'JPM', 'JNJ', 'V', 'PG', \n",
    "          'UNH', 'HD', 'MA', 'DIS', 'NVDA', 'BAC', 'ADBE', 'CMCSA', 'XOM', 'NFLX']\n",
    "\n",
    "# Function to fetch stock data\n",
    "def get_stock_data(ticker, start_date, end_date):\n",
    "    stock = yf.Ticker(ticker)\n",
    "    data = stock.history(start=start_date, end=end_date)\n",
    "    return data['Close']\n",
    "\n",
    "# Fetch data for different time periods\n",
    "data_list = []\n",
    "for stock in stocks:\n",
    "    # Fetch data for a 5-year period\n",
    "    data = get_stock_data(stock, \"2019-01-01\", \"2024-01-01\")\n",
    "    data = data.rename(stock)\n",
    "    data_list.append(data)\n",
    "\n",
    "# Combine all stock data\n",
    "combined_data = pd.concat(data_list, axis=1)\n",
    "\n",
    "print(f\"Shape of combined data: {combined_data.shape}\")\n",
    "print(f\"Number of NaN values: {combined_data.isna().sum().sum()}\")\n",
    "\n",
    "# Create features\n",
    "combined_data['Date'] = combined_data.index\n",
    "combined_data['Day'] = combined_data['Date'].dt.dayofweek\n",
    "combined_data['Month'] = combined_data['Date'].dt.month\n",
    "combined_data['Year'] = combined_data['Date'].dt.year\n",
    "\n",
    "# Calculate returns for each stock\n",
    "for stock in stocks:\n",
    "    combined_data[f'{stock}_Returns'] = combined_data[stock].pct_change()\n",
    "\n",
    "# Drop rows with NaN values\n",
    "combined_data_cleaned = combined_data.dropna()\n",
    "\n",
    "print(f\"Shape of cleaned data: {combined_data_cleaned.shape}\")\n",
    "\n",
    "if combined_data_cleaned.empty:\n",
    "    print(\"All data was dropped due to NaN values. Check your data source and date range.\")\n",
    "else:\n",
    "    # Create target variable (next day's average return across all stocks)\n",
    "    combined_data_cleaned['Target'] = combined_data_cleaned[[f'{stock}_Returns' for stock in stocks]].mean(axis=1).shift(-1)\n",
    "\n",
    "    # Drop the last row (which will have NaN in the Target column)\n",
    "    combined_data_cleaned = combined_data_cleaned.dropna()\n",
    "\n",
    "    print(f\"Final shape of data: {combined_data_cleaned.shape}\")\n",
    "\n",
    "    if combined_data_cleaned.empty:\n",
    "        print(\"All data was dropped. Check your data processing steps.\")\n",
    "    else:\n",
    "        # Select features and target\n",
    "        features = ['Day', 'Month', 'Year'] + [f'{stock}_Returns' for stock in stocks]\n",
    "        X = combined_data_cleaned[features]\n",
    "        y = combined_data_cleaned['Target']\n",
    "\n",
    "        # Split the data into training (60%), validation (20%), and testing (20%) sets\n",
    "        X_train_val, X_test, y_train_val, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n",
    "        X_train, X_val, y_train, y_val = train_test_split(X_train_val, y_train_val, test_size=0.25, random_state=42)\n",
    "\n",
    "        # Print the shapes of the datasets\n",
    "        print(\"Training set shape:\", X_train.shape)\n",
    "        print(\"Validation set shape:\", X_val.shape)\n",
    "        print(\"Testing set shape:\", X_test.shape)\n",
    "\n",
    "        # Visualize the average stock price trend\n",
    "        plt.figure(figsize=(12, 6))\n",
    "        combined_data_cleaned[stocks].mean(axis=1).plot()\n",
    "        plt.title('Average Stock Price Trend (20 S&P 500 Stocks)')\n",
    "        plt.xlabel('Date')\n",
    "        plt.ylabel('Average Price')\n",
    "        plt.show()\n",
    "\n",
    "        # Save the datasets\n",
    "        X_train.to_csv('X_train.csv', index=False)\n",
    "        y_train.to_csv('y_train.csv', index=False)\n",
    "        X_val.to_csv('X_val.csv', index=False)\n",
    "        y_val.to_csv('y_val.csv', index=False)\n",
    "        X_test.to_csv('X_test.csv', index=False)\n",
    "        y_test.to_csv('y_test.csv', index=False)\n",
    "\n",
    "print(\"Script completed.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "import zipfile\n",
    "import os\n",
    "\n",
    "# Name of your CSV file\n",
    "csv_filename = 'X_train.csv'\n",
    "csv_filename2 = \"y_train.csv\"\n",
    "\n",
    "# Name for the zip file (you can change this)\n",
    "X_train = 'X_train.zip'\n",
    "\n",
    "# Create a ZipFile object\n",
    "with zipfile.ZipFile(X_train, 'w') as zipf:\n",
    "    # Add the CSV file to the zip\n",
    "    zipf.write(csv_filename, os.path.basename(csv_filename))\n",
    "\n",
    "\n",
    "# # Save the datasets\n",
    "#         X_train.to_csv('X_train.csv', index=False)\n",
    "#         y_train.to_csv('y_train.csv', index=False)\n",
    "#         X_val.to_csv('X_val.csv', index=False)\n",
    "#         y_val.to_csv('y_val.csv', index=False)\n",
    "#         X_test.to_csv('X_test.csv', index=False)\n",
    "#         y_test.to_csv('y_test.csv', index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "            Trend  Slope  Trend_Numeric  Slope_Normalized\n",
      "JBL       Bullish    2.5              1               0.5\n",
      "PANW      Bullish    4.0              1               0.8\n",
      "CRWD      Bullish    3.5              1               0.7\n",
      "BLDR      Bullish    4.0              1               0.8\n",
      "RCL       Bullish    4.5              1               0.9\n",
      "AMD       Bullish    3.5              1               0.7\n",
      "DELL      Bullish    4.0              1               0.8\n",
      "GE        Bullish    3.5              1               0.7\n",
      "LLY       Bullish    4.0              1               0.8\n",
      "MSFT      Bullish    2.5              1               0.5\n",
      "Palantir  Bullish    4.0              1               0.8\n",
      "NVDA      Bullish    5.0              1               1.0\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "\n",
    "# Create a dictionary to store the labels\n",
    "stock_labels = {\n",
    "    'JBL': {'Trend': 'Bullish', 'Slope': 2.5},\n",
    "    'PANW': {'Trend': 'Bullish', 'Slope': 4.0},\n",
    "    'CRWD': {'Trend': 'Bullish', 'Slope': 3.5},\n",
    "    'BLDR': {'Trend': 'Bullish', 'Slope': 4.0},\n",
    "    'RCL': {'Trend': 'Bullish', 'Slope': 4.5},\n",
    "    'AMD': {'Trend': 'Bullish', 'Slope': 3.5},\n",
    "    'DELL': {'Trend': 'Bullish', 'Slope': 4.0},\n",
    "    'GE': {'Trend': 'Bullish', 'Slope': 3.5},\n",
    "    'LLY': {'Trend': 'Bullish', 'Slope': 4.0},\n",
    "    'MSFT': {'Trend': 'Bullish', 'Slope': 2.5},\n",
    "    'Palantir': {'Trend': 'Bullish', 'Slope': 4.0},\n",
    "    'NVDA': {'Trend': 'Bullish', 'Slope': 5.0}\n",
    "}\n",
    "\n",
    "# Create a DataFrame from the dictionary\n",
    "df = pd.DataFrame.from_dict(stock_labels, orient='index')\n",
    "\n",
    "# Add a numeric trend column (1 for Bullish, 0 for Bearish)\n",
    "df['Trend_Numeric'] = df['Trend'].map({'Bullish': 1, 'Bearish': 0})\n",
    "\n",
    "# Normalize the slope to be between 0 and 1\n",
    "df['Slope_Normalized'] = df['Slope'] / 5.0\n",
    "\n",
    "# Print the results\n",
    "print(df)\n",
    "\n",
    "# Save to CSV\n",
    "df.to_csv('stock_testing_labels.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Processed AAPL: Bullish (Slope: 0.0987)\n",
      "Processed MSFT: Bullish (Slope: 0.2722)\n",
      "Processed AMZN: Bullish (Slope: 0.0920)\n",
      "Processed GOOGL: Bullish (Slope: 0.0807)\n",
      "Processed NVDA: Bullish (Slope: 0.1647)\n",
      "Processed JPM: Bullish (Slope: 0.1470)\n",
      "Processed V: Bullish (Slope: 0.1347)\n",
      "Processed UNH: Bullish (Slope: 0.1045)\n",
      "Processed HD: Bullish (Slope: 0.1182)\n",
      "Processed DIS: Bearish (Slope: -0.0416)\n",
      "Processed PYPL: Bearish (Slope: -0.0781)\n",
      "Processed NFLX: Bullish (Slope: 0.6332)\n",
      "Processed INTC: Bearish (Slope: -0.0155)\n",
      "Processed VZ: Bearish (Slope: -0.0023)\n",
      "Processed T: Bullish (Slope: 0.0050)\n",
      "Processed BA: Bullish (Slope: 0.0076)\n",
      "Processed GE: Bullish (Slope: 0.1980)\n",
      "Processed XOM: Bullish (Slope: 0.0615)\n",
      "Processed WBA: Bearish (Slope: -0.0439)\n",
      "Processed CSCO: Bullish (Slope: 0.0069)\n",
      "Processed DIS: Bearish (Slope: -0.0416)\n",
      "Processed META: Bullish (Slope: 0.5648)\n",
      "Processed JNJ: Bearish (Slope: -0.0107)\n",
      "Processed V: Bullish (Slope: 0.1347)\n",
      "Processed JPM: Bullish (Slope: 0.1470)\n",
      "Processed WMT: Bullish (Slope: 0.0476)\n",
      "Processed PG: Bullish (Slope: 0.0433)\n",
      "Processed MA: Bullish (Slope: 0.2442)\n",
      "Processed BAC: Bullish (Slope: 0.0040)\n",
      "Processed CMCSA: Bullish (Slope: 0.0026)\n",
      "Processed KO: Bullish (Slope: 0.0119)\n",
      "Processed PEP: Bullish (Slope: 0.0159)\n",
      "Processed NKE: Bearish (Slope: -0.0645)\n",
      "Processed ORCL: Bullish (Slope: 0.1264)\n",
      "Processed CVX: Bullish (Slope: 0.0208)\n",
      "Processed WFC: Bullish (Slope: 0.0256)\n",
      "Processed ABBV: Bullish (Slope: 0.0759)\n",
      "Processed ABT: Bearish (Slope: -0.0017)\n",
      "Processed MCD: Bullish (Slope: 0.0749)\n",
      "Processed HON: Bullish (Slope: 0.0373)\n",
      "Processed MDT: Bearish (Slope: -0.0125)\n",
      "Processed RTX: Bullish (Slope: 0.0331)\n",
      "Processed MMM: Bullish (Slope: 0.0007)\n",
      "Processed C: Bullish (Slope: 0.0206)\n",
      "Processed TXN: Bullish (Slope: 0.0496)\n",
      "Processed F: Bearish (Slope: -0.0035)\n",
      "Processed MU: Bullish (Slope: 0.0654)\n",
      "Processed DUK: Bullish (Slope: 0.0169)\n",
      "Processed UL: Bullish (Slope: 0.0221)\n",
      "Processed TM: Bullish (Slope: 0.0589)\n",
      "Processed SHEL: Bullish (Slope: 0.0367)\n",
      "Processed RIO: Bullish (Slope: 0.0132)\n",
      "Processed NVS: Bullish (Slope: 0.0530)\n",
      "Processed HSBC: Bullish (Slope: 0.0282)\n",
      "Processed SONY: Bearish (Slope: -0.0004)\n",
      "Processed GIS: Bearish (Slope: -0.0006)\n",
      "Processed EOG: Bullish (Slope: 0.0412)\n",
      "Processed SLB: Bullish (Slope: 0.0153)\n",
      "Stock graphs saved in 'stock_graphs' directory.\n"
     ]
    }
   ],
   "source": [
    "import yfinance as yf\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.linear_model import LinearRegression\n",
    "import os\n",
    "from datetime import datetime, timedelta\n",
    "\n",
    "# Create a directory to save the images\n",
    "if not os.path.exists('stock_graphs'):\n",
    "    os.makedirs('stock_graphs')\n",
    "\n",
    "# List of stocks (mix of bullish and bearish)\n",
    "stocks = ['AAPL', 'MSFT', 'AMZN', 'GOOGL', 'NVDA', 'JPM', 'V', 'UNH', 'HD', 'DIS',\n",
    "          'PYPL', 'NFLX', 'INTC', 'VZ', 'T', 'BA', 'GE', 'XOM', 'WBA', 'CSCO', 'DIS','META','JNJ','V',\n",
    "          'JPM', 'WMT', 'PG', 'MA', 'BAC', 'CMCSA', 'KO', 'PEP', 'NKE', 'ORCL', 'CVX', 'WFC', 'ABBV', 'ABT',\n",
    "          'MCD', 'HON', 'MDT', 'RTX', 'MMM', 'C', 'TXN', 'F', 'MU', 'DUK', 'UL', 'TM', 'SHEL', 'RIO', 'NVS', 'HSBC', 'SONY', \n",
    "          'GIS', 'EOG', 'SLB']\n",
    "\n",
    "# Function to fetch stock data and determine if it's bullish or bearish based on slope\n",
    "def get_stock_data_and_trend(ticker, start_date, end_date):\n",
    "    stock = yf.Ticker(ticker)\n",
    "    data = stock.history(start=start_date, end=end_date)\n",
    "    \n",
    "    # Prepare data for linear regression\n",
    "    X = np.array(range(len(data))).reshape(-1, 1)\n",
    "    y = data['Close'].values\n",
    "\n",
    "    # Perform linear regression\n",
    "    model = LinearRegression()\n",
    "    model.fit(X, y)\n",
    "    \n",
    "    # Get the slope\n",
    "    slope = model.coef_[0]\n",
    "    \n",
    "    # Determine if it's bullish or bearish based on slope\n",
    "    is_bullish = slope > 0\n",
    "    label = 'Bullish' if is_bullish else 'Bearish'\n",
    "    \n",
    "    return data['Close'], label, slope\n",
    "\n",
    "# Set end date and calculate start date (3 years before)\n",
    "end_date = datetime.now().strftime(\"%Y-%m-%d\")\n",
    "start_date = (datetime.now() - timedelta(days=3*365)).strftime(\"%Y-%m-%d\")\n",
    "\n",
    "# Fetch data and create graphs\n",
    "for stock in stocks:\n",
    "    # Fetch data for a 3-year period and determine trend\n",
    "    data, label, slope = get_stock_data_and_trend(stock, start_date, end_date)\n",
    "    \n",
    "    # Plot and save individual stock graph\n",
    "    plt.figure(figsize=(12, 6))\n",
    "    data.plot()\n",
    "    plt.title(f'{stock} Stock Price (3-Year Period) - {label} (Slope: {slope:.4f})')\n",
    "    plt.xlabel('Date')\n",
    "    plt.ylabel('Close Price')\n",
    "    plt.savefig(f'stock_graphs/{stock}_{label.lower()}_3year.png')\n",
    "    plt.close()\n",
    "\n",
    "    print(f\"Processed {stock}: {label} (Slope: {slope:.4f})\")\n",
    "\n",
    "print(\"Stock graphs saved in 'stock_graphs' directory.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Dataset split complete. Images copied to dataset_even2\n",
      "Train set: Bullish = 436, Bearish = 350\n",
      "Val set: Bullish = 93, Bearish = 75\n",
      "Test set: Bullish = 94, Bearish = 75\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "import random\n",
    "from shutil import copyfile\n",
    "from PIL import Image, ImageEnhance\n",
    "\n",
    "# Directories\n",
    "image_dir = 'stock_graphs'\n",
    "output_dir = 'dataset_even2'\n",
    "\n",
    "# Create directories for the split\n",
    "for split in ['train', 'val', 'test']:\n",
    "    for label in ['bullish', 'bearish']:\n",
    "        os.makedirs(os.path.join(output_dir, split, label), exist_ok=True)\n",
    "\n",
    "# Data augmentation function\n",
    "def augment_image(image_path, output_path):\n",
    "    with Image.open(image_path) as img:\n",
    "        # Random rotation\n",
    "        if random.random() > 0.5:\n",
    "            angle = random.randint(-30, 30)\n",
    "            img = img.rotate(angle)\n",
    "        \n",
    "        # Random brightness adjustment\n",
    "        if random.random() > 0.5:\n",
    "            factor = random.uniform(0.7, 1.3)\n",
    "            enhancer = ImageEnhance.Brightness(img)\n",
    "            img = enhancer.enhance(factor)\n",
    "        \n",
    "        # Random horizontal flip\n",
    "        if random.random() > 0.5:\n",
    "            img = img.transpose(Image.FLIP_LEFT_RIGHT)\n",
    "        \n",
    "        img.save(output_path)\n",
    "\n",
    "# List all image files\n",
    "image_files = [f for f in os.listdir(image_dir) if f.endswith('.png')]\n",
    "\n",
    "# Separate bullish and bearish images\n",
    "bullish_images = [f for f in image_files if 'bullish' in f.lower()]\n",
    "bearish_images = [f for f in image_files if 'bearish' in f.lower()]\n",
    "\n",
    "# Calculate required augmentations\n",
    "target_per_class = 150\n",
    "bullish_augmentations = max(0, target_per_class - len(bullish_images))\n",
    "bearish_augmentations = max(0, target_per_class - len(bearish_images))\n",
    "\n",
    "# Perform augmentations\n",
    "if bullish_images:\n",
    "    for _ in range(bullish_augmentations):\n",
    "        original = random.choice(bullish_images)\n",
    "        new_name = f\"aug_bullish_{random.randint(1000, 9999)}.png\"\n",
    "        augment_image(os.path.join(image_dir, original), os.path.join(image_dir, new_name))\n",
    "        bullish_images.append(new_name)\n",
    "else:\n",
    "    print(\"No bullish images found for augmentation\")\n",
    "\n",
    "    for _ in range(bearish_augmentations):\n",
    "        original = random.choice(bearish_images)\n",
    "        new_name = f\"aug_bearish_{random.randint(1000, 9999)}.png\"\n",
    "        augment_image(os.path.join(image_dir, original), os.path.join(image_dir, new_name))\n",
    "        bearish_images.append(new_name)\n",
    "\n",
    "# Shuffle the images\n",
    "random.shuffle(bullish_images)\n",
    "random.shuffle(bearish_images)\n",
    "\n",
    "# Split ratios\n",
    "train_ratio, val_ratio, test_ratio = 0.7, 0.15, 0.15\n",
    "\n",
    "# Split the dataset\n",
    "def split_data(images, split_ratios):\n",
    "    total = len(images)\n",
    "    train_split = int(total * split_ratios[0])\n",
    "    val_split = int(total * (split_ratios[0] + split_ratios[1]))\n",
    "    return images[:train_split], images[train_split:val_split], images[val_split:]\n",
    "    \n",
    "\n",
    "bullish_train, bullish_val, bullish_test = split_data(bullish_images, (train_ratio, val_ratio, test_ratio))\n",
    "bearish_train, bearish_val, bearish_test = split_data(bearish_images, (train_ratio, val_ratio, test_ratio))\n",
    "\n",
    "# Function to copy files to their respective directories\n",
    "def copy_files(files, split, label):\n",
    "    for file in files:\n",
    "        src = os.path.join(image_dir, file)\n",
    "        dst = os.path.join(output_dir, split, label, file)\n",
    "        copyfile(src, dst)\n",
    "\n",
    "# Copy files to their respective directories\n",
    "copy_files(bullish_train, 'train', 'bullish')\n",
    "copy_files(bearish_train, 'train', 'bearish')\n",
    "copy_files(bullish_val, 'val', 'bullish')\n",
    "copy_files(bearish_val, 'val', 'bearish')\n",
    "copy_files(bullish_test, 'test', 'bullish')\n",
    "copy_files(bearish_test, 'test', 'bearish')\n",
    "\n",
    "print(f\"Dataset split complete. Images copied to {output_dir}\")\n",
    "\n",
    "# Print summary of the split\n",
    "for split in ['train', 'val', 'test']:\n",
    "    bullish_count = len(os.listdir(os.path.join(output_dir, split, 'bullish')))\n",
    "    bearish_count = len(os.listdir(os.path.join(output_dir, split, 'bearish')))\n",
    "    print(f\"{split.capitalize()} set: Bullish = {bullish_count}, Bearish = {bearish_count}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CSV file created: train_dataset.csv\n",
      "Total entries in train set: 1073\n",
      "CSV file created: val_dataset.csv\n",
      "Total entries in val set: 435\n",
      "CSV file created: test_dataset.csv\n",
      "Total entries in test set: 425\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "import pandas as pd\n",
    "\n",
    "# Define the dataset directory\n",
    "dataset_dir = 'dataset_even'\n",
    "\n",
    "# Initialize dictionaries to store file paths and labels for each split\n",
    "data_splits = {'train': [], 'val': [], 'test': []}\n",
    "\n",
    "# Loop through the dataset directory\n",
    "for split in ['train', 'val', 'test']:\n",
    "    for label in ['bullish', 'bearish']:\n",
    "        folder_path = os.path.join(dataset_dir, split, label)\n",
    "        if os.path.exists(folder_path):\n",
    "            for file_name in os.listdir(folder_path):\n",
    "                if file_name.endswith('.png'):\n",
    "                    file_path = os.path.join(folder_path, file_name)\n",
    "                    data_splits[split].append({'file_path': file_path, 'label': label})\n",
    "\n",
    "# Save each split to a separate CSV file\n",
    "for split, data in data_splits.items():\n",
    "    split_df = pd.DataFrame(data)\n",
    "    csv_file_path = f'{split}_dataset.csv'\n",
    "    split_df.to_csv(csv_file_path, index=False)\n",
    "    print(f\"CSV file created: {csv_file_path}\")\n",
    "    print(f\"Total entries in {split} set: {len(split_df)}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test reference dataset created: test_reference_dataset.csv\n",
      "Total test entries: 425\n",
      "                                         file_path  label\n",
      "0   dataset_even/test/bullish/aug_bullish_1289.png      1\n",
      "1   dataset_even/test/bullish/aug_bullish_6900.png      1\n",
      "2   dataset_even/test/bullish/aug_bullish_3059.png      1\n",
      "3  dataset_even/test/bullish/XOM_bullish_3year.png      1\n",
      "4   dataset_even/test/bullish/aug_bullish_8454.png      1\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "import pandas as pd\n",
    "\n",
    "# Define the dataset directory\n",
    "dataset_dir = 'dataset_even'\n",
    "\n",
    "# Initialize a list to store file paths and binary labels for the test set\n",
    "test_data = []\n",
    "\n",
    "# Loop through the test set directory\n",
    "split = 'test'\n",
    "for label in ['bullish', 'bearish']:\n",
    "    folder_path = os.path.join(dataset_dir, split, label)\n",
    "    if os.path.exists(folder_path):\n",
    "        for file_name in os.listdir(folder_path):\n",
    "            if file_name.endswith('.png'):\n",
    "                file_path = os.path.join(folder_path, file_name)\n",
    "                binary_label = 1 if label == 'bullish' else 0\n",
    "                test_data.append({'file_path': file_path, 'label': binary_label})\n",
    "\n",
    "# Convert the data into a DataFrame\n",
    "test_reference_df = pd.DataFrame(test_data)\n",
    "\n",
    "# Save the test reference dataset to a CSV file\n",
    "csv_file_path = 'test_reference_dataset.csv'\n",
    "test_reference_df.to_csv(csv_file_path, index=False)\n",
    "\n",
    "print(f\"Test reference dataset created: {csv_file_path}\")\n",
    "print(f\"Total test entries: {len(test_reference_df)}\")\n",
    "print(test_reference_df.head())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Random Dummy Baseline Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Random Baseline Model Classification Report:\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "     bearish       0.40      0.49      0.44        75\n",
      "     bullish       0.55      0.46      0.50       101\n",
      "\n",
      "    accuracy                           0.47       176\n",
      "   macro avg       0.47      0.47      0.47       176\n",
      "weighted avg       0.49      0.47      0.47       176\n",
      "\n",
      "Accuracy: 0.47\n",
      "Precision: 0.55\n",
      "Recall: 0.46\n",
      "F1-Score: 0.50\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "import random\n",
    "from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score, classification_report\n",
    "\n",
    "# Assuming you have a 'test' directory with subdirectories 'bullish' and 'bearish'\n",
    "test_dir = 'dataset/test'\n",
    "\n",
    "# Get the list of test images and their true labels\n",
    "def load_test_data(test_dir):\n",
    "    test_images = []\n",
    "    true_labels = []\n",
    "    \n",
    "    for label in ['bullish', 'bearish']:\n",
    "        label_dir = os.path.join(test_dir, label)\n",
    "        for img_file in os.listdir(label_dir):\n",
    "            test_images.append(img_file)\n",
    "            true_labels.append(label)\n",
    "    \n",
    "    return test_images, true_labels\n",
    "\n",
    "# Function to generate random predictions (either 'bullish' or 'bearish')\n",
    "def random_predictions(num_samples):\n",
    "    return [random.choice(['bullish', 'bearish']) for _ in range(num_samples)]\n",
    "\n",
    "# Load the test data\n",
    "test_images, true_labels = load_test_data(test_dir)\n",
    "\n",
    "# Generate random predictions\n",
    "predictions = random_predictions(len(test_images))\n",
    "\n",
    "# Evaluate the random baseline model\n",
    "accuracy = accuracy_score(true_labels, predictions)\n",
    "precision = precision_score(true_labels, predictions, pos_label='bullish', average='binary')\n",
    "recall = recall_score(true_labels, predictions, pos_label='bullish', average='binary')\n",
    "f1 = f1_score(true_labels, predictions, pos_label='bullish', average='binary')\n",
    "\n",
    "# Print classification report and metrics\n",
    "print(\"Random Baseline Model Classification Report:\")\n",
    "print(classification_report(true_labels, predictions))\n",
    "print(f\"Accuracy: {accuracy:.2f}\")\n",
    "print(f\"Precision: {precision:.2f}\")\n",
    "print(f\"Recall: {recall:.2f}\")\n",
    "print(f\"F1-Score: {f1:.2f}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Simple CNN Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 786 images belonging to 2 classes.\n",
      "Found 168 images belonging to 2 classes.\n",
      "Found 169 images belonging to 2 classes.\n"
     ]
    }
   ],
   "source": [
    "import tensorflow as tf\n",
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Conv2D, MaxPooling2D, Flatten, Dense\n",
    "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
    "from sklearn.metrics import classification_report\n",
    "\n",
    "# Define directories for training and validation data\n",
    "train_dir = 'dataset_even2/train'\n",
    "val_dir = 'dataset_even2/val'\n",
    "test_dir = 'dataset_even2/test'\n",
    "\n",
    "# Image data generator for loading images and applying augmentations\n",
    "train_datagen = ImageDataGenerator(rescale=1./255)\n",
    "val_datagen = ImageDataGenerator(rescale=1./255)\n",
    "test_datagen = ImageDataGenerator(rescale=1./255)\n",
    "\n",
    "# Load training data\n",
    "train_generator = train_datagen.flow_from_directory(\n",
    "    train_dir,\n",
    "    target_size=(150, 150),\n",
    "    batch_size=32,\n",
    "    class_mode='binary'\n",
    ")\n",
    "\n",
    "# Load validation data\n",
    "val_generator = val_datagen.flow_from_directory(\n",
    "    val_dir,\n",
    "    target_size=(150, 150),\n",
    "    batch_size=32,\n",
    "    class_mode='binary'\n",
    ")\n",
    "\n",
    "# Load test data\n",
    "test_generator = test_datagen.flow_from_directory(\n",
    "    test_dir,\n",
    "    target_size=(150, 150),\n",
    "    batch_size=32,\n",
    "    class_mode='binary',\n",
    "    shuffle=False  # Important for evaluation\n",
    ")\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/10\n",
      "25/25 [==============================] - 20s 758ms/step - loss: 0.7891 - accuracy: 0.5433 - val_loss: 0.8316 - val_accuracy: 0.5536\n",
      "Epoch 2/10\n",
      "25/25 [==============================] - 19s 759ms/step - loss: 0.6734 - accuracy: 0.5700 - val_loss: 0.6706 - val_accuracy: 0.5298\n",
      "Epoch 3/10\n",
      "25/25 [==============================] - 19s 769ms/step - loss: 0.6543 - accuracy: 0.5840 - val_loss: 0.6696 - val_accuracy: 0.5238\n",
      "Epoch 4/10\n",
      "25/25 [==============================] - 19s 743ms/step - loss: 0.6569 - accuracy: 0.5674 - val_loss: 0.6786 - val_accuracy: 0.5179\n",
      "Epoch 5/10\n",
      "25/25 [==============================] - 19s 730ms/step - loss: 0.6539 - accuracy: 0.5420 - val_loss: 0.6663 - val_accuracy: 0.6667\n",
      "Epoch 6/10\n",
      "25/25 [==============================] - 19s 738ms/step - loss: 0.6504 - accuracy: 0.5471 - val_loss: 0.6665 - val_accuracy: 0.6488\n",
      "Epoch 7/10\n",
      "25/25 [==============================] - 19s 767ms/step - loss: 0.6458 - accuracy: 0.5840 - val_loss: 0.6694 - val_accuracy: 0.5298\n",
      "Epoch 8/10\n",
      "25/25 [==============================] - 18s 729ms/step - loss: 0.6462 - accuracy: 0.5789 - val_loss: 0.6769 - val_accuracy: 0.6548\n",
      "Epoch 9/10\n",
      "25/25 [==============================] - 18s 702ms/step - loss: 0.6424 - accuracy: 0.5980 - val_loss: 0.6551 - val_accuracy: 0.5417\n",
      "Epoch 10/10\n",
      "25/25 [==============================] - 18s 701ms/step - loss: 0.6317 - accuracy: 0.6260 - val_loss: 0.6498 - val_accuracy: 0.6012\n",
      "6/6 [==============================] - 3s 413ms/step - loss: 0.6530 - accuracy: 0.5621\n",
      "Test Accuracy: 0.5621301531791687\n",
      "6/6 [==============================] - 3s 379ms/step\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.50      0.84      0.63        75\n",
      "           1       0.73      0.34      0.46        94\n",
      "\n",
      "    accuracy                           0.56       169\n",
      "   macro avg       0.62      0.59      0.55       169\n",
      "weighted avg       0.63      0.56      0.54       169\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Define a simple CNN model\n",
    "model = Sequential([\n",
    "    Conv2D(32, (3, 3), activation='relu', input_shape=(150, 150, 3)),\n",
    "    MaxPooling2D(pool_size=(2, 2)),\n",
    "    \n",
    "    Conv2D(64, (3, 3), activation='relu'),\n",
    "    MaxPooling2D(pool_size=(2, 2)),\n",
    "    \n",
    "    Conv2D(128, (3, 3), activation='relu'),\n",
    "    MaxPooling2D(pool_size=(2, 2)),\n",
    "    \n",
    "    Flatten(),\n",
    "    \n",
    "    Dense(128, activation='relu'),\n",
    "    Dense(1, activation='sigmoid')  # Binary classification: bullish vs bearish\n",
    "])\n",
    "\n",
    "# Compile the model\n",
    "model.compile(optimizer='adam', loss='binary_crossentropy', metrics=['accuracy'])\n",
    "\n",
    "# Train the model\n",
    "model.fit(train_generator, epochs=10, validation_data=val_generator)\n",
    "\n",
    "# Evaluate on test set\n",
    "test_loss, test_acc = model.evaluate(test_generator)\n",
    "print(f\"Test Accuracy: {test_acc}\")\n",
    "\n",
    "# Predict on test set and generate classification report\n",
    "y_pred = model.predict(test_generator)\n",
    "y_pred_labels = (y_pred > 0.5).astype(int)  # Convert probabilities to binary labels\n",
    "\n",
    "# Get true labels from generator\n",
    "y_true = test_generator.classes\n",
    "\n",
    "print(classification_report(y_true, y_pred_labels))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Incorporating Technical Indicators"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "def generate_random_features(num_samples, num_features=5):\n",
    "    return pd.DataFrame(\n",
    "        np.random.rand(num_samples, num_features),\n",
    "        columns=[f'feature_{i}' for i in range(num_features)]\n",
    "    )\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Feature Extraction from Stock Charts \n",
    "a. Image processing to exstract price and volume data from the chart \n",
    "b. Calculuate technical indicators based on this extracted data "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import cv2\n",
    "from tensorflow.keras.utils import Sequence\n",
    "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
    "\n",
    "class CombinedDataGenerator(Sequence):\n",
    "    def __init__(self, directory, batch_size=32, img_size=(150, 150), shuffle=True):\n",
    "        self.image_generator = ImageDataGenerator(rescale=1./255).flow_from_directory(\n",
    "            directory,\n",
    "            target_size=img_size,\n",
    "            batch_size=batch_size,\n",
    "            class_mode='binary',\n",
    "            shuffle=shuffle\n",
    "        )\n",
    "        self.batch_size = batch_size\n",
    "        self.shuffle = shuffle\n",
    "        self.img_size = img_size\n",
    "\n",
    "    def __len__(self):\n",
    "        return len(self.image_generator)\n",
    "\n",
    "    def __getitem__(self, index):\n",
    "        X_image, y = self.image_generator[index]\n",
    "        X_numerical = np.array([self.extract_features(img) for img in X_image])\n",
    "        return [X_image, X_numerical], y\n",
    "\n",
    "    def extract_features(self, img):\n",
    "        # Convert to grayscale\n",
    "        gray = cv2.cvtColor(img, cv2.COLOR_RGB2GRAY)\n",
    "        \n",
    "        # Extract price line\n",
    "        edges = cv2.Canny(gray, 50, 150)\n",
    "        lines = cv2.HoughLinesP(edges, 1, np.pi/180, 30, minLineLength=20, maxLineGap=5)\n",
    "        \n",
    "        # Calculate features\n",
    "        price_data = self.extract_price_data(lines) if lines is not None else np.zeros(50)\n",
    "        \n",
    "        # Calculate technical indicators\n",
    "        slope = self.calculate_slope(price_data)\n",
    "        volatility = self.calculate_volatility(price_data)\n",
    "        trend_strength = self.calculate_trend_strength(price_data)\n",
    "        momentum = self.calculate_momentum(price_data)\n",
    "        support_resistance = self.calculate_support_resistance(price_data)\n",
    "        \n",
    "        return np.array([slope, volatility, trend_strength, momentum, support_resistance])\n",
    "\n",
    "    def extract_price_data(self, lines):\n",
    "        y_points = []\n",
    "        for line in lines:\n",
    "            x1, y1, x2, y2 = line[0]\n",
    "            y_points.extend([y1, y2])\n",
    "        return np.array(y_points)\n",
    "\n",
    "    def calculate_slope(self, price_data):\n",
    "        if len(price_data) < 2:\n",
    "            return 0\n",
    "        x = np.arange(len(price_data))\n",
    "        return np.polyfit(x, price_data, 1)[0]\n",
    "\n",
    "    def calculate_volatility(self, price_data):\n",
    "        return np.std(price_data) if len(price_data) > 1 else 0\n",
    "\n",
    "    def calculate_trend_strength(self, price_data):\n",
    "        if len(price_data) < 2:\n",
    "            return 0\n",
    "        diff = np.diff(price_data)\n",
    "        return np.mean(np.abs(diff))\n",
    "\n",
    "    def calculate_momentum(self, price_data):\n",
    "        if len(price_data) < 10:\n",
    "            return 0\n",
    "        return price_data[-1] - price_data[0]\n",
    "\n",
    "    def calculate_support_resistance(self, price_data):\n",
    "        if len(price_data) < 2:\n",
    "            return 0\n",
    "        return np.max(price_data) - np.min(price_data)\n",
    "\n",
    "    def on_epoch_end(self):\n",
    "        if self.shuffle:\n",
    "            self.image_generator.on_epoch_end()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 786 images belonging to 2 classes.\n",
      "Found 168 images belonging to 2 classes.\n",
      "Found 169 images belonging to 2 classes.\n"
     ]
    }
   ],
   "source": [
    "train_gen = CombinedDataGenerator('dataset_even2/train', batch_size=32)\n",
    "val_gen = CombinedDataGenerator('dataset_even2/val', batch_size=32)\n",
    "test_gen = CombinedDataGenerator('dataset_even2/test', batch_size=32, shuffle=False)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"model_7\"\n",
      "__________________________________________________________________________________________________\n",
      " Layer (type)                   Output Shape         Param #     Connected to                     \n",
      "==================================================================================================\n",
      " input_11 (InputLayer)          [(None, 150, 150, 3  0           []                               \n",
      "                                )]                                                                \n",
      "                                                                                                  \n",
      " conv2d_42 (Conv2D)             (None, 148, 148, 32  896         ['input_11[0][0]']               \n",
      "                                )                                                                 \n",
      "                                                                                                  \n",
      " max_pooling2d_42 (MaxPooling2D  (None, 74, 74, 32)  0           ['conv2d_42[0][0]']              \n",
      " )                                                                                                \n",
      "                                                                                                  \n",
      " conv2d_43 (Conv2D)             (None, 72, 72, 64)   18496       ['max_pooling2d_42[0][0]']       \n",
      "                                                                                                  \n",
      " max_pooling2d_43 (MaxPooling2D  (None, 36, 36, 64)  0           ['conv2d_43[0][0]']              \n",
      " )                                                                                                \n",
      "                                                                                                  \n",
      " conv2d_44 (Conv2D)             (None, 34, 34, 128)  73856       ['max_pooling2d_43[0][0]']       \n",
      "                                                                                                  \n",
      " max_pooling2d_44 (MaxPooling2D  (None, 17, 17, 128)  0          ['conv2d_44[0][0]']              \n",
      " )                                                                                                \n",
      "                                                                                                  \n",
      " input_12 (InputLayer)          [(None, 5)]          0           []                               \n",
      "                                                                                                  \n",
      " flatten_14 (Flatten)           (None, 36992)        0           ['max_pooling2d_44[0][0]']       \n",
      "                                                                                                  \n",
      " dense_34 (Dense)               (None, 64)           384         ['input_12[0][0]']               \n",
      "                                                                                                  \n",
      " concatenate_6 (Concatenate)    (None, 37056)        0           ['flatten_14[0][0]',             \n",
      "                                                                  'dense_34[0][0]']               \n",
      "                                                                                                  \n",
      " dense_35 (Dense)               (None, 128)          4743296     ['concatenate_6[0][0]']          \n",
      "                                                                                                  \n",
      " dense_36 (Dense)               (None, 1)            129         ['dense_35[0][0]']               \n",
      "                                                                                                  \n",
      "==================================================================================================\n",
      "Total params: 4,837,057\n",
      "Trainable params: 4,837,057\n",
      "Non-trainable params: 0\n",
      "__________________________________________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "from tensorflow.keras.models import Model\n",
    "from tensorflow.keras.layers import Input, Conv2D, MaxPooling2D, Flatten, Dense, concatenate\n",
    "\n",
    "# Image input branch\n",
    "img_input = Input(shape=(150, 150, 3))\n",
    "x = Conv2D(32, (3, 3), activation='relu')(img_input)\n",
    "x = MaxPooling2D(pool_size=(2, 2))(x)\n",
    "x = Conv2D(64, (3, 3), activation='relu')(x)\n",
    "x = MaxPooling2D(pool_size=(2, 2))(x)\n",
    "x = Conv2D(128, (3, 3), activation='relu')(x)\n",
    "x = MaxPooling2D(pool_size=(2, 2))(x)\n",
    "x = Flatten()(x)\n",
    "\n",
    "# Numerical input branch\n",
    "num_features = 5  # Update this to match the number of features you're extracting\n",
    "num_input = Input(shape=(num_features,))\n",
    "y = Dense(64, activation='relu')(num_input)\n",
    "\n",
    "# Combine branches\n",
    "combined = concatenate([x, y])\n",
    "z = Dense(128, activation='relu')(combined)\n",
    "output = Dense(1, activation='sigmoid')(z)\n",
    "\n",
    "model = Model(inputs=[img_input, num_input], outputs=output)\n",
    "\n",
    "# Compile the model\n",
    "model.compile(optimizer='adam', loss='binary_crossentropy', metrics=['accuracy'])\n",
    "\n",
    "# Print model summary to verify the architecture\n",
    "model.summary()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [],
   "source": [
    "class CombinedDataGenerator(tf.keras.utils.Sequence):\n",
    "    def __init__(self, directory, batch_size=32, img_size=(150, 150), shuffle=True):\n",
    "        self.image_generator = ImageDataGenerator(rescale=1./255).flow_from_directory(\n",
    "            directory,\n",
    "            target_size=img_size,\n",
    "            batch_size=batch_size,\n",
    "            class_mode='binary',\n",
    "            shuffle=shuffle\n",
    "        )\n",
    "        self.batch_size = batch_size\n",
    "        self.shuffle = shuffle\n",
    "        self.img_size = img_size\n",
    "\n",
    "    def __len__(self):\n",
    "        return len(self.image_generator)\n",
    "\n",
    "    def __getitem__(self, index):\n",
    "        X_image, y = self.image_generator[index]\n",
    "        X_numerical = np.array([self.extract_features(img) for img in X_image])\n",
    "        return [X_image, X_numerical], y\n",
    "\n",
    "    def extract_features(self, img):\n",
    "        # Implement feature extraction here\n",
    "        # For now, let's return dummy values\n",
    "        return np.random.rand(5)  # 5 dummy features\n",
    "\n",
    "    def on_epoch_end(self):\n",
    "        if self.shuffle:\n",
    "            self.image_generator.on_epoch_end()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 786 images belonging to 2 classes.\n",
      "Found 168 images belonging to 2 classes.\n",
      "Epoch 1/10\n",
      "25/25 [==============================] - 21s 785ms/step - loss: 0.8519 - accuracy: 0.5433 - val_loss: 0.6898 - val_accuracy: 0.5536\n",
      "Epoch 2/10\n",
      "25/25 [==============================] - 21s 838ms/step - loss: 0.6640 - accuracy: 0.5636 - val_loss: 0.6701 - val_accuracy: 0.5060\n",
      "Epoch 3/10\n",
      "25/25 [==============================] - 19s 765ms/step - loss: 0.6524 - accuracy: 0.5763 - val_loss: 0.6682 - val_accuracy: 0.5298\n",
      "Epoch 4/10\n",
      "25/25 [==============================] - 19s 742ms/step - loss: 0.6480 - accuracy: 0.5840 - val_loss: 0.6711 - val_accuracy: 0.5476\n",
      "Epoch 5/10\n",
      "25/25 [==============================] - 18s 719ms/step - loss: 0.6481 - accuracy: 0.5878 - val_loss: 0.6717 - val_accuracy: 0.5476\n",
      "Epoch 6/10\n",
      "25/25 [==============================] - 17s 688ms/step - loss: 0.6323 - accuracy: 0.6170 - val_loss: 0.6468 - val_accuracy: 0.5774\n",
      "Epoch 7/10\n",
      "25/25 [==============================] - 18s 706ms/step - loss: 0.5824 - accuracy: 0.6858 - val_loss: 0.6553 - val_accuracy: 0.5893\n",
      "Epoch 8/10\n",
      "25/25 [==============================] - 18s 701ms/step - loss: 0.5017 - accuracy: 0.7494 - val_loss: 0.5683 - val_accuracy: 0.7202\n",
      "Epoch 9/10\n",
      "25/25 [==============================] - 18s 707ms/step - loss: 0.4077 - accuracy: 0.8219 - val_loss: 0.4562 - val_accuracy: 0.7917\n",
      "Epoch 10/10\n",
      "25/25 [==============================] - 18s 718ms/step - loss: 0.3248 - accuracy: 0.8511 - val_loss: 0.4681 - val_accuracy: 0.7798\n",
      "Found 169 images belonging to 2 classes.\n",
      "6/6 [==============================] - 2s 393ms/step - loss: 0.4877 - accuracy: 0.7929\n",
      "Test Accuracy: 0.7928994297981262\n"
     ]
    }
   ],
   "source": [
    "# Assuming you've already created your generators\n",
    "train_gen = CombinedDataGenerator('dataset_even2/train', batch_size=32)\n",
    "val_gen = CombinedDataGenerator('dataset_even2/val', batch_size=32)\n",
    "\n",
    "# Train the model\n",
    "history = model.fit(\n",
    "    train_gen,\n",
    "    validation_data=val_gen,\n",
    "    epochs=10,\n",
    "    steps_per_epoch=len(train_gen),\n",
    "    validation_steps=len(val_gen)\n",
    ")\n",
    "\n",
    "# Evaluate on test set\n",
    "test_gen = CombinedDataGenerator('dataset_even2/test', batch_size=32, shuffle=False)\n",
    "test_loss, test_acc = model.evaluate(test_gen, steps=len(test_gen))\n",
    "print(f\"Test Accuracy: {test_acc}\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 169 images belonging to 2 classes.\n",
      "6/6 [==============================] - 3s 405ms/step\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.44      1.00      0.61        75\n",
      "           1       0.00      0.00      0.00        94\n",
      "\n",
      "    accuracy                           0.44       169\n",
      "   macro avg       0.22      0.50      0.31       169\n",
      "weighted avg       0.20      0.44      0.27       169\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/kaylahoffman/miniconda3/lib/python3.9/site-packages/sklearn/metrics/_classification.py:1509: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, f\"{metric.capitalize()} is\", len(result))\n",
      "/Users/kaylahoffman/miniconda3/lib/python3.9/site-packages/sklearn/metrics/_classification.py:1509: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, f\"{metric.capitalize()} is\", len(result))\n",
      "/Users/kaylahoffman/miniconda3/lib/python3.9/site-packages/sklearn/metrics/_classification.py:1509: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, f\"{metric.capitalize()} is\", len(result))\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import classification_report\n",
    "import numpy as np\n",
    "\n",
    "# Assuming you have your test generator ready\n",
    "test_gen = CombinedDataGenerator('dataset_even2/test', batch_size=32, shuffle=False)\n",
    "\n",
    "# Make predictions\n",
    "y_pred = model.predict(test_gen)\n",
    "\n",
    "# Convert predictions to class labels\n",
    "y_pred_classes = np.argmax(y_pred, axis=1)\n",
    "\n",
    "# Get true labels\n",
    "y_true = test_gen.image_generator.classes\n",
    "\n",
    "# Generate and print the classification report\n",
    "print(classification_report(y_true, y_pred_classes))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Save the complete model (architecture + weights)\n",
    "model.save('stock_prediction_model.h5')\n",
    "\n",
    "# Optionally, save just the weights if needed\n",
    "model.save_weights('stock_prediction_weights.h5')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## DQN Agent"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "from collections import deque\n",
    "import random\n",
    "\n",
    "class DQNAgent:\n",
    "    def __init__(self, state_size, action_size):\n",
    "        self.state_size = state_size\n",
    "        self.action_size = action_size\n",
    "        self.memory = deque(maxlen=2000)\n",
    "        self.gamma = 0.95    # discount rate\n",
    "        self.epsilon = 1.0   # exploration rate\n",
    "        self.epsilon_min = 0.01\n",
    "        self.epsilon_decay = 0.995\n",
    "        self.model = self.build_model()\n",
    "\n",
    "    def build_model(self):\n",
    "        model = Sequential([\n",
    "            Conv2D(32, (3, 3), activation='relu', input_shape=self.state_size),\n",
    "            MaxPooling2D(pool_size=(2, 2)),\n",
    "            Conv2D(64, (3, 3), activation='relu'),\n",
    "            MaxPooling2D(pool_size=(2, 2)),\n",
    "            Conv2D(128, (3, 3), activation='relu'),\n",
    "            MaxPooling2D(pool_size=(2, 2)),\n",
    "            Flatten(),\n",
    "            Dense(128, activation='relu'),\n",
    "            Dense(self.action_size, activation='linear')\n",
    "        ])\n",
    "        model.compile(optimizer='adam', loss='mse', metrics=['accuracy'])\n",
    "        return model\n",
    "\n",
    "    def remember(self, state, action, reward, next_state, done):\n",
    "        self.memory.append((state, action, reward, next_state, done))\n",
    "\n",
    "    def act(self, state):\n",
    "        if np.random.rand() <= self.epsilon:\n",
    "            return random.randrange(self.action_size)\n",
    "        act_values = self.model.predict(state)\n",
    "        return np.argmax(act_values[0])\n",
    "\n",
    "    def replay(self, batch_size):\n",
    "        minibatch = random.sample(self.memory, batch_size)\n",
    "        for state, action, reward, next_state, done in minibatch:\n",
    "            target = reward\n",
    "            if not done:\n",
    "                target = reward + self.gamma * np.amax(self.model.predict(next_state)[0])\n",
    "            target_f = self.model.predict(state)\n",
    "            target_f[0][action] = target\n",
    "            self.model.fit(state, target_f, epochs=1, verbose=0)\n",
    "        if self.epsilon > self.epsilon_min:\n",
    "            self.epsilon *= self.epsilon_decay"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "\n",
    "class StockEnvironment:\n",
    "    def __init__(self, data_generator):\n",
    "        self.data_generator = data_generator\n",
    "        self.reset()\n",
    "\n",
    "    def reset(self):\n",
    "        self.current_step = 0\n",
    "        self.data, self.labels = next(self.data_generator)\n",
    "        return self.data[self.current_step]\n",
    "\n",
    "    def step(self, action):\n",
    "        if self.current_step >= len(self.data) - 1:\n",
    "            return None, 0, True, {}\n",
    "        \n",
    "        true_label = self.labels[self.current_step]\n",
    "        \n",
    "        # Assuming 0 is bearish and 1 is bullish\n",
    "        if action == true_label:\n",
    "            reward = 1  # Correct prediction\n",
    "        else:\n",
    "            reward = -1  # Incorrect prediction\n",
    "        \n",
    "        self.current_step += 1\n",
    "        done = self.current_step == len(self.data) - 1\n",
    "        \n",
    "        return self.data[self.current_step], reward, done, {}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "episode: 0/50, score: -5\n",
      "episode: 1/50, score: 7\n",
      "1/1 [==============================] - 0s 67ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 17ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 17ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 17ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 17ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 16ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 36ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 16ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 16ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 17ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 17ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "episode: 2/50, score: -11\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 17ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 17ms/step\n",
      "1/1 [==============================] - 0s 17ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 17ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 17ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 17ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 17ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 17ms/step\n",
      "episode: 3/50, score: -1\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 17ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 17ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 110ms/step\n",
      "1/1 [==============================] - 0s 33ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 30ms/step\n",
      "1/1 [==============================] - 0s 31ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 32ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 17ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "episode: 4/50, score: -1\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 16ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 17ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 17ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 17ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 17ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 17ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "episode: 5/50, score: -11\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "episode: 6/50, score: 5\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 17ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 17ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 30ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "episode: 7/50, score: 5\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "episode: 8/50, score: 7\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 31ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "episode: 9/50, score: -7\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 51ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "episode: 10/50, score: -7\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "episode: 11/50, score: 3\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 35ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "episode: 12/50, score: 1\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 30ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 33ms/step\n",
      "1/1 [==============================] - 0s 41ms/step\n",
      "1/1 [==============================] - 0s 35ms/step\n",
      "1/1 [==============================] - 0s 48ms/step\n",
      "1/1 [==============================] - 0s 34ms/step\n",
      "1/1 [==============================] - 0s 34ms/step\n",
      "1/1 [==============================] - 0s 36ms/step\n",
      "1/1 [==============================] - 0s 34ms/step\n",
      "1/1 [==============================] - 0s 110ms/step\n",
      "1/1 [==============================] - 0s 39ms/step\n",
      "1/1 [==============================] - 0s 36ms/step\n",
      "1/1 [==============================] - 0s 30ms/step\n",
      "1/1 [==============================] - 0s 31ms/step\n",
      "1/1 [==============================] - 0s 32ms/step\n",
      "1/1 [==============================] - 0s 35ms/step\n",
      "1/1 [==============================] - 0s 37ms/step\n",
      "1/1 [==============================] - 0s 32ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 36ms/step\n",
      "1/1 [==============================] - 0s 31ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 35ms/step\n",
      "1/1 [==============================] - 0s 31ms/step\n",
      "1/1 [==============================] - 0s 31ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 35ms/step\n",
      "1/1 [==============================] - 0s 35ms/step\n",
      "1/1 [==============================] - 0s 33ms/step\n",
      "1/1 [==============================] - 0s 32ms/step\n",
      "1/1 [==============================] - 0s 32ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 34ms/step\n",
      "1/1 [==============================] - 0s 32ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "episode: 13/50, score: -3\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 30ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 31ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 72ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 30ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 30ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 55ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "episode: 14/50, score: 3\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 35ms/step\n",
      "1/1 [==============================] - 0s 37ms/step\n",
      "1/1 [==============================] - 0s 37ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 32ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 45ms/step\n",
      "1/1 [==============================] - 0s 30ms/step\n",
      "1/1 [==============================] - 0s 42ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 43ms/step\n",
      "1/1 [==============================] - 0s 30ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 32ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "episode: 15/50, score: -1\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 31ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 30ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 55ms/step\n",
      "1/1 [==============================] - 0s 50ms/step\n",
      "1/1 [==============================] - 0s 31ms/step\n",
      "1/1 [==============================] - 0s 35ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 38ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 30ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 30ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 30ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "episode: 16/50, score: 3\n",
      "1/1 [==============================] - 0s 30ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 30ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 43ms/step\n",
      "1/1 [==============================] - 0s 31ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 17ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "episode: 17/50, score: 3\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 30ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 30ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 30ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 30ms/step\n",
      "1/1 [==============================] - 0s 30ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 33ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 30ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 30ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "episode: 18/50, score: -7\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 30ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 31ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 32ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 35ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 40ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "episode: 19/50, score: 1\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 38ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 31ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 30ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 38ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 54ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "episode: 20/50, score: -7\n",
      "1/1 [==============================] - 0s 36ms/step\n",
      "1/1 [==============================] - 0s 30ms/step\n",
      "1/1 [==============================] - 0s 101ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 50ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 44ms/step\n",
      "1/1 [==============================] - 0s 30ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 32ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 17ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "episode: 21/50, score: -3\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 69ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 30ms/step\n",
      "episode: 22/50, score: -5\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 45ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 51ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "episode: 23/50, score: -1\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 17ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 30ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 17ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 17ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 41ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 17ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "episode: 24/50, score: -1\n",
      "1/1 [==============================] - 0s 33ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 17ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 30ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 41ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 56ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 44ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 17ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "episode: 25/50, score: -5\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 49ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 36ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 51ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "episode: 26/50, score: 7\n",
      "1/1 [==============================] - 0s 42ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 30ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 54ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "episode: 27/50, score: 7\n",
      "1/1 [==============================] - 0s 34ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 31ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 17ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "episode: 28/50, score: 5\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 32ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 17ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 44ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "episode: 29/50, score: -1\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 41ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 65ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 17ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 72ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "episode: 30/50, score: -5\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "episode: 31/50, score: 3\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 179ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 38ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 47ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 87ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 32ms/step\n",
      "1/1 [==============================] - 0s 37ms/step\n",
      "1/1 [==============================] - 0s 31ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 40ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "episode: 32/50, score: 11\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 33ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 35ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 36ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 33ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "episode: 33/50, score: 1\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 49ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 33ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 42ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "episode: 34/50, score: -7\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 44ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 46ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 52ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "episode: 35/50, score: -7\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 43ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 37ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "episode: 36/50, score: 11\n",
      "1/1 [==============================] - 0s 54ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 53ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 53ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 35ms/step\n",
      "1/1 [==============================] - 0s 139ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 105ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 30ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 31ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 36ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 47ms/step\n",
      "1/1 [==============================] - 0s 17ms/step\n",
      "episode: 37/50, score: -9\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 41ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 31ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 46ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 32ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "episode: 38/50, score: -7\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 42ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 52ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 43ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 17ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 41ms/step\n",
      "episode: 39/50, score: -3\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 45ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 45ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 37ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "episode: 40/50, score: -3\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 44ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 45ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 17ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "episode: 41/50, score: -3\n",
      "1/1 [==============================] - 0s 32ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 30ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 49ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 52ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 53ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 37ms/step\n",
      "1/1 [==============================] - 0s 52ms/step\n",
      "episode: 42/50, score: -1\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 34ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 35ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 47ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "episode: 43/50, score: -7\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 43ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 48ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 30ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 48ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 51ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "episode: 44/50, score: 1\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 47ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 48ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 40ms/step\n",
      "1/1 [==============================] - 0s 37ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 53ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "episode: 45/50, score: 1\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 48ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 43ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "episode: 46/50, score: -5\n",
      "1/1 [==============================] - 0s 38ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 52ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 58ms/step\n",
      "episode: 47/50, score: 3\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 53ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 30ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 44ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 55ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "episode: 48/50, score: 7\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 30ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 47ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 30ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 39ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 40ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 47ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 37ms/step\n",
      "episode: 49/50, score: -1\n",
      "1/1 [==============================] - 0s 31ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 43ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 53ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 39ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 38ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 41ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 28ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 30ms/step\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n"
     ]
    }
   ],
   "source": [
    "# Initialize DQN agent\n",
    "state_size = (150, 150, 3)\n",
    "action_size = 2  # binary classification: bullish or bearish\n",
    "agent = DQNAgent(state_size, action_size)\n",
    "\n",
    "# Create environment\n",
    "env = StockEnvironment(train_generator)\n",
    "\n",
    "# Training loop\n",
    "episodes = 50\n",
    "batch_size = 32\n",
    "\n",
    "for e in range(episodes):\n",
    "    state = env.reset()\n",
    "    state = np.expand_dims(state, axis=0)\n",
    "    total_reward = 0\n",
    "    \n",
    "    for time in range(500):  # Limit the number of steps per episode\n",
    "        action = agent.act(state)\n",
    "        next_state, reward, done, _ = env.step(action)\n",
    "        \n",
    "        if next_state is not None:\n",
    "            next_state = np.expand_dims(next_state, axis=0)\n",
    "            agent.remember(state, action, reward, next_state, done)\n",
    "            state = next_state\n",
    "        \n",
    "        total_reward += reward\n",
    "        \n",
    "        if done:\n",
    "            print(f\"episode: {e}/{episodes}, score: {total_reward}\")\n",
    "            break\n",
    "    \n",
    "    if len(agent.memory) > batch_size:\n",
    "        agent.replay(batch_size)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "Test Accuracy: 0.5\n"
     ]
    }
   ],
   "source": [
    "def evaluate_model(agent, test_generator):\n",
    "    correct_predictions = 0\n",
    "    total_predictions = 0\n",
    "    \n",
    "    for i in range(len(test_generator)):\n",
    "        state = test_generator.next()[0][0]\n",
    "        state = np.expand_dims(state, axis=0)\n",
    "        action = agent.act(state)\n",
    "        true_label = test_generator.classes[i]\n",
    "        \n",
    "        if action == true_label:\n",
    "            correct_predictions += 1\n",
    "        total_predictions += 1\n",
    "    \n",
    "    accuracy = correct_predictions / total_predictions\n",
    "    print(f\"Test Accuracy: {accuracy}\")\n",
    "\n",
    "# Evaluate the model\n",
    "evaluate_model(agent, test_generator)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "tf",
   "language": "python",
   "name": "tf"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.18"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}