{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.12.12","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"kaggle":{"accelerator":"none","dataSources":[{"sourceType":"datasetVersion","sourceId":15752375,"datasetId":10094590,"databundleVersionId":16695744}],"dockerImageVersionId":31329,"isInternetEnabled":true,"language":"python","sourceType":"notebook","isGpuEnabled":false}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"markdown","source":"<div style=\"background-color: #e3f2fd; padding: 15px; border-radius: 10px; border-left: 6px solid #2196f3; color: #0d47a1; font-family: sans-serif;\">\n    <strong style=\"font-size: 18px;\"> Tip:</strong> \n    <span style=\"font-size: 16px;\">Make sure to <b>turn on a GPU</b> \n</div>","metadata":{}},{"cell_type":"markdown","source":"#### areej alqarni ","metadata":{}},{"cell_type":"markdown","source":"# Enhancing Aspect Based Sentiment Analysis of Saudi E-Commerce Reviews Using BERT Models","metadata":{},"attachments":{"fe43dd1d-1374-493b-b4f0-e1baf612aae3.png":{"image/png":"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"},"881e2309-29cc-4947-a1db-b42bc84ee4cd.png":{"image/png":"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"},"69570d86-8b0b-40d4-bed4-90a46fb5a4b9.png":{"image/png":"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"},"f209277a-9a7a-4efc-b615-b1b06f3b1648.png":{"image/png":"iVBORw0KGgoAAAANSUhEUgAACSoAAAB7CAMAAABOgty+AAAAAXNSR0IArs4c6QAAAARnQU1BAACxjwv8YQUAAABjUExURQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAGZodN4AAAAhdFJOUwAwQBAocIBQ5/+HOO/3GI9gv7don0gIxyDfr6d4z9dYl+bFGsMAAAAJcEhZcwAAFxEAABcRAcom8z8AACu6SURBVHhe7Z3rWts8s4YhgNkGSAspUGg5/6Ncl6QZ7WYjOyT56Hqf+xdYsjyanWRZdk5OAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAD/S05Xq9VZfxD8VzhfrU77Y+CofNsIPFtdTJf/U++AdwLwb7Igdr9tBqw5v7oO3PTHD8rZv6CZ/wS3d8H668v++LH5D3vE1yPwMMq7f4i+cX19teqLjkXvnYfpKDgk/6zN/lnBvwd97Dp8PQMehceUDjf98QNy/yNc8e5nfxwcnymZ/6k/flz+0x7x1Qg8kPJOaaJ0fX39oy87Fq13Hqij4ID8szb7ZwX/LiwYWb6aAZdxOklmTYmPK2bkJl1yPUu+nXieNoHp57Yv+WY0Zvu5Or64Y4deTZtfocrVj+m8L9sXh/eI/XAQv/pqBB5GeS/r1GzgtS9MHN4zWu/0O7pbBnz5+bYJPX3dTM992X+LVn+rl758N3ybHZjb6WbzWwbrz+n35mG67Y52HEzwYcr3HNkx0vaiKavR5nvbx1j0biycnT2nUy929oPxyJL5agZcBBm2ZepraRxVzEScr19fXw+cdXcuswoe+qLvhTDbx2Nf5cCMHPp0U4u3OZDJDu4R++EwfvXVCDyM8rjVgDpVOoZntN7pdlSEUmCUAZ8/6tqvF3Lc+s8g9PfnYR/PXV2bHZr0XOf6vT36Nx1dt0d7DiW40HOf8kWFADmyKKuMRGlERVryOa8Zr7Vc9piL7+ZuN+oZjSwVX82Ai3jlntXMkfK4YiZYWGnAPfE7q+Dueyc/xWxX70cV2Xfos7devMPceh/cI/bDYfzqqxF4EOWd566GbN6XHsszWu90O6qEkunWRDvZC/zZdVz499H09+Pri4WuzQ4NXbubMPOcoj3acyjBFT23KV+pICKgho1EsaIil5XqewR5Q3FaLSn7MWTjjywNX82Aiyg9q5gj5XHFTHCGOlReOqsMPWdX2aG4iE9r/vaHK4qcFUfdGeI69Flz0+1UXIbUy6E9wkbKYnMgv/pqBB5Eee/U6I/L1eNbd1t+OM/oab3T7WgtSsaV6bIyJ7PPKfC/Ra+JyPqir7YU12aHhq6901TpUIJTsy11yu/LIuTI/eEIGcmbKsnJUNmIeH29vu9L6/ugXfOSO7K0fDUDLqLqWmGOlMcVM/GQLnl1qLRUL0X+7guPx32SwFvorQSt0FZED4Xr0GLlwKq4CEUvh/YIE0UWmwP51Vcj8CDKo0aV9aTIYTxD0Hqn29FGFsaT6aKvHNn3yPjP0CuC+OpaoWuzQ0N92GmqdCjB6eIdVcrviyLeVImM5E2V5KpSU9xPiO/rm4hd85I7srR8NQMuoupaQSpI4ahiJu5jov3Y99Jmpt5msd63q8/nhUToj1dUgtYcccOS59Bla87HZkO3IWrFZSh6ObRHmCiy2BzIr74agQdRHs2FjOW2A3mGoPVOt6NZohonA9bz3gq98f8AvSKIK33X72xcmx0a6sNOU6VDCZ4121JSfl8SIUfuDxPRSD/7oxWyF03xn66wuYvYNS95I0vHVzPgIqhbT6uaWdn8qGJmXva9W66CVgqII047OmYMw5WMZ6vnfKveu+4B8Rya16B/xwXa7d+wVUetuAxdL4f0CBtdFp1D+dUeInD/youvtmk5NnIgzxAI77Q7ShLNzYAv5QnEx/S8ep4+0z4Qo8P//2n0d/GU1xU++4qLsW12aKgLO02VDiU4XdxO+XRAdeSmrDPSNr/wxo0+8AH5AJ1bos1P3VpqerpOZbvmJRG7NnvIgPNJ1+qdYg5HFfMo0PSaXn846s6fhhnDMNXgYTffrB/vMYDj0Dw1KK7x8rR27tPnMkMvR2OJLIfyq28ZgdRL3REP5RkCxzt7SKS5GZCmgvX3NS/Dsf/6VIn1d8+b0a66ev8SXZ+I2VOlg0AXt1M+/a86cldmGIk76L3nT1Xolbp2y8dtOkjBt2teWhC7R82A6Vq6fn2OKuZRSLe8V9QxuWftWMwYhqlGXqGgB+TWc48D4Dg0R/FXtyv0zNDL0Vgiy6H86ltGIOlFz7aH8gyB4509JNLMDJhHqI/6AdP7+npvVv3X6PWX96vYK3Pfnr5PiW81VZIpXxc60ZfpRlowVbpNDbRTrc90bJWq7JqXFsTuUTNgupau3/ADK5QQXt6nzc3U3DnVYr5ciOLEy+M0vW0+p4t+SdttmrmNq4B/qyz0slpVi5tzGrm9jB/s6g8r0Oj3ye8rybvds9XP6cfn9Fi/CnuaFznvL6a3p+lRe0Z/O02/36bpUs0eopczhmGqkeOG5vPd0x1T+0ZfEo6wZ4/T51vqouPQ/PDbGz2Mi7gWVfViecT59PA2VV843D5Ob5/TX6VT+5NFZ+BX7mWoQLdjmyhOV/3X/+5Xq8oEmsFb5RmVVO4fp+nHWxucEdKLnm3HnmF0VfldKHHA807R0QKJpGdAAS8q/WnD/KwVxdTOQv9cWD0xcufLm83DY10oElA5rjXUIfSXBsx+YVFr7EU+6qx+FESzmdbKdtV7vjhwu1pVHj12ctGniDZV0tqyUpIZ37MGKbq4nfJ1oROiTDXSgqnSC+2/rG96tunZ9I0xVTKDIuDEbkKzvDZV0gyyH9K1VP2ex56H31e55S0GV9WrzkVMtTg08Emr8YG7z0pDg6YjL095V8CGnOgltUf77mc0Ij+A8qrZKUGvOa94F25v6tPy9s7rRBZLMoRHulmG9Wc3Wco/iXV9ff3WP5joe7kVX8AQ6SJAZTlMzuhAtapka9/oS8QT9j5/HujhzHDoCL8vbi5xWRdxLGroxfKIZx7SfiXXOfukK4q3mPcni4XrV85luIZpxypR3P+Jf37Upkzt0e8jaQZvlWdUUrmoXvn/U76/mIQoiN9mGniG3dXnNM0sdwKr/oDnnaKjNXSSlgElp3yNPlvVGNpZ7J8Lq0eG7vyShMsvYvYJiLEaElCdoj+eD1dGVhs7TVapw2Gb+hmeUSs2U1uhqLorQ2PaW3xVvCcJxDXmODmVD6dKWltWSrLie+4gRWV2yqf/VUcWZYqRlk2VqG6d/+kTnafqVMkMioATu6lctbwyVdIMsi+oXU2/SY7Nybb+1GepmMXklcC2+OS894C7YpZB0yFo6lavr99it1vNDBspBqgwsnTek3ZleMy2GCFAqSZVfcpT4Uj7PbqLYuXAU3NDJHqZc3FG/QIAleW44Q//5THb077Rl4AnbPlOa1hkVRya4YcU1q4c8yKORQ29zPCIoKPnaghu+rRPWSxcv3IuE/HsWHWeslM9gtOhuO1TN3iXaPRKCpfdVJFzf7t9XftigOsZXleFt4kDnneKjFpD52gZUMLxqnYgYWnHMrbtnwurB4buTPMT3g0sE9CoIQHVEANDtXtLb4x9tlqYeE5HgnDSZnorPD8v16dq5QAN0lGeeU5OpYOpkt7WjJS00yBFRXbKp/9VRxZl0kjVwTlTJVojr1/oTZPWP2zYOtycoBjEbsCwfK9rwyD7glrV9EtWPqtfda56QGL+ak2di5tPLBA5KAZNn5xs2xJ6aqF6odnImbj5D7SPqCrI9T7LFwOb/f+tFfiDOkmGp07c18rXnpqSoLCy6KT0kvNHQZuMiLjhb+5z2672jb4MhKUbe2JN+UeTLn+wWSv0LuJY1NDLHI94zikgUYfwPmUx8P3KuUzAtWPV+S3Vq4ZwWmKPTekG7xKNXknCS0MVqUfNh7oD4ndNPM9wuyrSZ3/A9c6uoy10ipYBJXxrbGYRWzumsU3/XFh9jjvfc05MplES0KghAVUo+uPZV06CRmPss9WrcuSzb5rNjFb4+C+uxzcy+UxefonV5zk5lQ6mSnpbc1LSLoNUX9SnfEPohCgTRgpwB+dMlbj3RVS6U5qUqZIXFIPYdSzf69owyL6gVjX9Jjk++s/r8pJJG7AZLta+1pA/yzVoWnTaGRjtRkQbEXPlnJJvmGaTX9c3xeK7c+lwkuFHL0OO2/xMuKJMdoWET3OHYSpjP92SAHl1w9W+0RdfWG0ks6TLVX8oK/fORRyLGnqZ4xGvvTKKC+xVFgPfr5zLBHrRA10Upc5TOql+pJNycJhtGAZvlWdUErTrEEQcYnn/VkZMlTzPcLvaz4z6A753dhm1hapqGVDAo24adjVs7ZjGNv1zYfUZ7vwrv7+XTKMkoFFDAioXizjjrELhUG0NJp8Naa23mdkKTRmySdiP83oHLWXGlmY6ORX6UyWjrTkpaZdBiorMlG8InRBlvZEi3MFZUyXSe7k5o/nQi5wquUExiF3H8r2uDYPsC2pU0y8rriNM+WcU5ydSH5syceYnz4NzZaftgVHAjVAb66dp+l2MsTbdIC3kxrjllsvzb86RH2EDZHxQkI4bMuRQ4CcOr2/T9MbL5hwIWi/FcNP/ZmOCyihuzmhl7y4/4/a0b/XFF1aPaX2SUN1CvfXb67yLGNoMFjX0MssjevJEdr+yGLh+5V0m4Nmx6TxP33ICol2f4Qc3LIM3yrMq9dAzko516FKeSDBybuJ4htvV0VTJ987WSzqoqpYBBeww5ovwjnZsY/ewfy6svsid41RJS0CjhgRUnPXHdmQXthtjmfIjJFoQivP9zmZ2K7xWKxIVL0nSXUQQcK6TUzV3qmS1NSsl7TBIUZmZ8g2hE31Zb6QEC6tdnslVeFEwC5Ci9lfORTnc/KAYxK5j+R0z2K6wHPEXrSJvrKjKyuu3dO2IjPummFQX55g/0psW24nU2s0GA1rT+TezNu+r1eVN0I8/VdIaSZa7ipOW8zRePTUvQrTQ0BJXg/n6xevI2mS+l+mObpgrGe4eprJDkhaVt2TaTbzluad7OlpV0Ht5sdnwrV80x2/1BpZqvMV3V9iBqt9y9rRv9cUVlt+3CN2cqnskdap0Xt8nbJoc617Es6iuF9Mj7p4e2ju5j+ktW4f0tGdZdHy/ci8zsGPbeZpe5PxEeyOCfSyDN+dblTpYZ9evF7cnL3/zHV/MueF38ej/X5vNRq4ceZ7hdnUwVRp4Z+slHVRVy4ACXtApq8YtrnYW++fC6nPdOVUyE9CgIQG1QT59llcQxlKJcCCjxoWK1mZOK/xZU36Ol9XC6y0UGUGemU4uZhYJ1mH6z2rLTElafC8YpOhEM+W7jkxllpEIFtb0/3qqxL8BznczNEH6KaZKg6AYxK5n+Z0y2M6QXDX83kqxcvy27jnf69G81iqm+9rz6/VTUTnZhsPNOpea5rudH2kNdfu+Tka2vFBrhO5QKBsn61VTcAENLWmbG938Fl3nxUViS38VGZ6iKfmxKv16JnWEFynZ7MlBrV7OeRGdajRsqgjztG/1xRWWHTy9JP035yN1qlRWCCIf1cZN9yK+RTW9mB5R3ZgG7oII+adaKbr3LIuO71ejy3h2bDvPKmcXp92uod+WwZvzrUodebJAs0OWP38tiv7Xz/Y8w+3qYKo08M7WSzq4boV4c49h6a2nUQPtLPTPhdVnuvP6/eVkNdXPZfsE5DYkoEY2YQS/4R+rKTMXrzEKh/zBaepQ9P3WZl4r5LkUVOWdC/IditRYPNPJZ02VrLZawf34XjJI5YYq6pSvVciOTP9bRiJYWEMrkVKFlns4RNPQFzy9myoNgmIQu57ld8pgO0Ny1XA+ylbm943pX3Igq5gn8+fV3nje8nVH/1rnUtOkjvLq8/lD1KPhhWoj6VEqX5Ca1N4sIFIFWlnnpvPSME2BRLboZcj7JFMoUOzn08juaTZp9XLOMEw1ajaNcI72rb54wtInM/KG9VP2aH2qtO1e69jk8PcuIrTZuYWiF8Mj6IR8A0Ni81NxctE9y6Lj+9XoMp4d286zQPQskDZQxym7ZfDmfKtSB+ms/Bgo79jkZ5D0r5WmbM9wu+pPlUbe2XpJB1Wt0Z262nRhzaUG2lnonwurz3Pnah3CSkBuQwJuuCWr0GuMhSKzkQun/XatzbxW+F2B1ArP//LuJWoozgpmOvmsqZLVlpGS1PheMkhxQxVtyu9Layv0BYnezzt7qJQqHHVpukbP48KUppsq+UExil3P8jtlsJ0huWr6qVKeetI0gP63ivX3aEkB3cSyPzf9z+4kYlP3Qr2RZI+8Up6WPjvfryDzUlv82nPeM0drA+JukmXITsfLifFMaqYsXVNUx0Kzl3OGYarR8lreRuhotG/0xRWW01HelHNKabqPNobDgUhLzIOLDCyq6cX3CDZG/ugKjdNp6XffsqgM/Gp0mQ41iqjztLjfLh1G8xgGb8+3KrV0E7LqKQoLTBXsbGt4Rk/bVX+qNPLOVlEddG6N5dR50bj/mhEx0k5vbN8/F1af587VLN1KQH5DAm64Zp014DbWvRJKyw9pAGls5rZSbwZvdr6kA2SzuH45z8nnTZWstvyUtPsgRQ211Cm/LxtNlYqRGBbWDt4mvkmzaRJI54Z5XjtVGgTFIHZ9y++QwXYnNd/AViU5ipSU5NpAFsV6QiJ3JiNY56amqbRMQxnVC91G8v7L9DyizwoFumPjz0zQuJNP59csfnWJnWT4U0SlW7UoBC1RFuPRzrNYaPZyzjBMNXo+ZWORRvtGX1xheWdeWQOo9sKoPPPjgcQ6ada9yMCiml4GHkEn5D0xdEuSKuxbFpWBX40u06FGUZeU0zhKz/TjAGEYvD3fqtTCbyhXDwlovOY7JGrGyba6Z/S0XfWnSiPvbBXVQefW6PPUaqpk3JOMtCOMTc3p/rmw+ix3rsdhKwH5DQmo4Yq7h6IAvzGaN7ST/aTcxmZ+KxRj6Z/8AIetnxwpvQ83z8nnTZWsttSoNOI7lc4bpNJ5gpLy+5LakfuC1kgMd9AJ3jq+KezSk89kvKjmdqo0CIpB7PqW3yGD7Q41X7Hm/QMiwahOYBV3eEm+P0CuLxtaIgDdM5EFaHJqflWZHoFmp+Xnq5zGyydjNs1vAggZOJvGOwVar69+6zkdiA/nzV7OGYaphsCYUjfaN/riCkv6qLazisFL8MjP5SNXcX3AvYjUZntA0cvAI0iCPBhT4KXg3rcsGiO/Gl2mw4sifiMl5ncSLw2DhsHb861KLd0WogB1iTecUCNetlU9o2fJVGnkna5OU1kNZ0CeGEVibuYjxo3/SDtCDtc/F1af5c7ld+jtBOQ3JEiFNT+qHTR+YyRVWjvkjxo2z82SdH4rNNrG51gp7aeoi3qhSEhz1XlOPm+qZLU1SEnNgSWDVCqU5JTfF1RDuVJWG4nhDnrBW1XhfUPBH0nNMeDaqdIgKAax61t+hwy2O9S4Fvu+lUfFkfOfD9Wbv1aS7w/QGqochZcIsE3/0M0u3b6YGqRoy7NwflKSZag+7lDvPhUysKFjmDa5tiJ+G8vs5ZxhmGrEG7Cz1WW5ULVP1tS+3hdXWLmGJAYvhcf6ly6ibt2LSG22BxS9DDyiH1soztLYsm9ZNIZ+NbhMxLJjV5X6EztHCYiuqxu8O9+o1EJOWy8w0HjN33ujJqwGCOkZCaurwtuaAyPvVHRaSGVKBiwbhCNhPKNmlcqRkXaEHK5/Lqy+0J3tBOQ3JOirNUP0oDF+FBPzWPuhnkZgvxW+SQirCWmhjU6Ob6umP2mpZpaTWz5BrXLMG20NUlJzYMkglQqdlE//qr6ZKxcqIzHcQVsxbXyTBoIHUQjGNtup0iAoBrHrW36HDLY71LSmX9/Ko+JwoEl6ET3JdwdoIVbunVwkAJno1/P5Of+EkvWOb7ZIeRZAMtCbbP1XVfOqp5AhrxuEv6uvyLSEQrOXc4ZhqpEfBNzza6jlUaCtfb0vrrC0ZFoJKwYvlcvqYUu4unsRqc32gKKXgUe4Y8u+ZdEY+tXgMq4du6rUuXhXSpeh6+oG7843KrVQu/W2Fb4s3RrTf8NM1XtGwO6q8LbmwMg7hU5rUpmSAbljRJCE91kZz6JG2hFyuP65sPpCd7YTkN+QgAqfVqvniZYZrtd5ZWTQGBXXS3aU0xqB57USTBiz/pp+pTFsKaNhwE19Air3p0pGW4OU1B5YMEilUifl07/SkYdGYriDXvDWVcj/QjKrV8/bqdIgKAax61t+hwy2O9Supt+BlQfF/JmIFj3JdwfEr/qotQaNnNxXT60T0jcYvi+R5H0B2TMjf2j5UsjQxFP+Nm5HXIw0ezlnGKYaZc/EPQcAPaj1tK/3xRWWCithxeBl8DerNuz4cy8itdkeUPQy8Ah3bNm3LApjvxpcxrVjfy719iHfrefnfqrB+/P1Si10idppWRMkVPufR+sZfleFtzUHRt7ZK6ohlSlxqEyVOLhpLtMz0o6Qw/XPhdUXurOdgPyGBFQam+EPI5Y7tkFjJFbY4kKvQvH35huBB63Q7emGA+6NFhjCDCy1mmWf4+T5o/LdRna2Pv+vtzVISe2BBYMUldspn/4T9gzUZdJIDHfQC96mCknwl1fPk8LaqdIgKAax61t+hwy2O9Sspt+BlQfF6SeVe4wk3x4g5YrbnWUC5NXyjPHeSrUjTFJt/rlg1wy8JjcTMjSrSo3lKqIgZi/nDMNUo9peyr2lYcfTvt4XV1h5ByoGL4tbFiWc7F5EarM9oOhl4BHu2LJvWRTGfjW4jGvH/lwySflh3urJlmJwcb5aqUXJe/xe1uKpUusZfleFtzUHRt4pOlqTypQMyCYmwhCen8nJLbGBkXaEHK5/Lqy+0J3tBOQ3JKDS1Gn+uY1cedBY9alt6gy/HtAIPGiFLLXeUiMX9IjnLlusMsnYybNqusVDelZUfapda2uQkroD8wcpKjdTvu3IfZkwEkOyucHbVKGdRG+8NpSmecOpUh0Ug9j1LS+0qxlkX1Crmn6FHEucYJtvETcP4btXlAeNJK8GhhyFlwhQJt2M8bJrwPi2Ot3vMNuf1X1v8/pvlYHYf8Pf1JHPsictkpRg9nLOMEw1qrih57/J53ztxxqiL66wtDheCSsGLxOWLCQe9yJSm+0BRS8Dj5gztuxNFoWxX/mX8e3Yn8syrThzled+msHF+WqlFsVpeTrYPgbwsm2m9gy/q8LbmgMj75QdrUhlWgZ8L9893mxSbuaFsH6QSYy0I+Rw/XNh9YXurEpbHzcaEqTarD+eTPJ7daPG6AHLUx52OaU1Ao9aobH/mex+ylOwFW9VGkRCD7XXbWSnqKqXFJW2BimpPbBgkKIKVsrvDdHSlvVGYkg2N3ibKjTnWdNH+Elf6gM4KygGsetbXmhXM8i+oEY1/Qo5ljgBa+M3dYqi3OqiFhgysy0RgFth+p+cqslf+Fdo3wu+LdvM9OTH8RRvPaiybjGzl3OGYapRCcf7/2Os+donur64wpL7Wu8p+NBFQk/di0httgcUvQw8wh1b9i2LZIZf+Zfx7SjO5axEz/36XyobO69SqYWuUH87jQajZdu6CaocJPC7KrytOTDyTrWjTCpTM6CEk4r+MGqkHSGH658Lqy90ZzsB+Q0JUu2sP747oMnkqDGS64o7l3/vuRF41Aqt9zykPoVX4VLnJhInf+SR8J28mLk9ShOoTmF9W4OU1B5YMEhRFSvlC0M0dGWdkRiSzQ3etgp5JGmGmlenSlZQDGLXt7zQbqA3yL6gNjX9CjmWOAHNGPLzq0GS1wKjn/P2p/mNJHM83d+EW8Lp3f3WAruIRv9FTf4VmvSEmM4shuaPj8bT6EapbyJh9nLOMEw1qrjhoSa+6uFrP9P0xRWWNVTe65ZvLljQax5x7cC9yMCiml4GHuGOLfuWRTLDr/zL+Ha0zr0jX5D5RXXecr5WqYV0lnelFz/mVujc3tVUas/wu0r7qUucNfl05J1GRxN0rpYBJdSu8fGbkXaEHK5/Lqy+0J3tBOQ3JEi1s/54xYL6MGqM7yZWdF6u2Ag8aoUU8cFbldg/NpSOZZbynLy8S9XcJ3PPxMKPF1VC782BJYMUXcNK+cIQDV1ZZySG48gL3rZKjof6YDtVGgTFIHZ9ywvtJnzj7gq1qelXyLHECUjYHNXzknyTUboFnb7WoJEkgHwhUoNWgf/UC+78Qzn1p7Mi9HZFekLMoubpebMoSv+IZBQxe5mHYfV7Mwl5Lkd37LGv/ULdF1dY/lBYvqD3te77tgmOyzByuxcZWFTTy8Aj3LFl37JIZviVfxnfjuJc3kXed7ugOW85X6vUwtm5FPGva/BmA6ogXC3ieMasruZ8yz9/kNxv5J1WRyN0rpYBJdzb66u2g/fvQbKRdoQcvaG+MlVa6M52AvIbElArWX+8UJJeXBg2RtVvaO6bhWkEHrZC7pAMEJJxejS0Js9QPNlx8pLFmwtyx2RYOVEl9K7E97xBKp1npnxpiJq+rDUSwx6hB2+irdJ8UINvZNqp0iAoBrHrW15ol3CNuyskqKZfIccSJyBZ8zyDnslaSb45wAuLZZX79iPORZYIkP6epSv+2E377gE7orh/rB+qs3Oxpbc8n43uy+OpurPB7GW+2XbmxFSjxA25J2nc135F1RdXWC78RS77ws/YtanSx91UezavhwZbuBcZWFTTy8Aj3LFl37II5viVfxnfjuLcrOpId8uYUJy3Ol+p1JL9IB/h39xiHdO/0tUCjmf4XeWxi2zV/VDUyDvNjgaoqpYBFfjtoevX6lnJdlrHgB9pR8jh+ufC6gvd2U5AfkMCqp31x0NjWggYNsafI0yJMz9/awUetkJmSftUYkwmD6EnQ9oNje3k+ZPQzcIsm7LevcrYUSX03hxIf88apEYpXxqipi9rjcSQbEbwJroq9VurPFK2U6VBUAxi17e80C7jGXdXSBJNv0KOJU5AUwaa8G45KVpJvj3Aj1LfkvrOJrLpEgHS3PTu89Lcj5ihaO2GFvbE2NzlTVlcIunihJydizXImZRmwRSp6+YR9C01ZfUyDxPyAUqGTmVX2F7w2JFe2nC1b/XFFZangOkBySq/raRMlUKeuXpnrW/pdpFSjHsR36KaXgYe4Y4t+5ZFMMOv+la/FkUlewbySyWWwZvzrUodfDvKdn+ke8A83lK5GnSeZ/hd5WH9LpqyfGaAxBh4p1RUBVXVMqBCdRv9QFPg889wvRiKA+0IOXz/XFh9mTs7CchtSECNZP3xN5zpF3ZGjfH9RKIM3a3Ao1ZKHuaNSXlOW/d6ppNX72P8oive5w/88PM3q61BSmoOLBikRilfGqKmL+uMRLAWPWG6KhzD9fOXdqo0CopB7LqW3yWD7QzJ9Vo/IqBvnvtWHhSTrGkqeZZ3r1lJvj2Q89HrdLn6+RC0F0t2EIB53UyP2r1FgIzVP4Nmbwoyf1yvP8lYz2TpaJcqRD8unh/yHJva4pvh9Q3Nmc8uP+/Yt61eFu95ulxNG3V5liq8hXeFprdNeY8irXK42rf64gqbf+R087x6rHSrTJXoRuDXzftqdXmTRUsqcS/iW1TTy8Aj/LFlz7IIZvhV36ruxLod5bm5w5G8mGUZvDnfqtTBOrt+u3w52a7o7q26f6P/1WzreYbf1ay09ds0VXthyf0G3qkoqkBVtQyoUY0N19cfT28sTFTAQDtCDt8/F1Zf5s5OAnIbElDlMkJzr2cEfKDZ2VyWAVqBR63kZaA8dagyc/kp1plOfnLykifj683NxeMNPzavJgRWW4OUpMU34w1So5TvOzKVWUYiWGdq8BJdlfJrImWa202VBkExiF3X8rtksJ3JktUkKX0rD4pZ66/T88+37Hdmku8OdC7EJbsIUHE3qX7Ia3z9M2i2aQi0mKQ+nqaSp82rBDieyhvQ138202aT/h30snuO0o+0kbpCDS1yuNq3+uIKm29DOpSpUp22CvRLX+5FfItqehl4hD+27FmWnjl+1bdqOLFqR3luM5CXxSzL4M35VqWO/IS5pawG0AE123qe4Xc1L9p3kPsNvFNRVKE/J6K4NdMM7YWY+AfaEXL4/rmw+jJ3DhgJyG1IQDXLKMwzsLSMPGyMP0QXTykJuhV42Eql+PSshqMvkCe+M528nxHX5KdlVluDlKTGd4UxSBl+Wq9r9wURcmT6zzISweKowUv0VfgbTZVmuqnSICgGsetafpcMtjNZjhrN5uKAX9x+ZDxjJfnuQLNbLLL4AZyaXO80J6Bpbf9qdf5QVlgolPkx3bMrvh7ItzH0wYmWQS+7AjVr1xUq+JmLq32rL76wQp0pRjTp+pqB/ItD3kUGFlX0MvAIf2zZsyw9c/yqb/VrUdQMEOWjbpbBm/OtSj0v+f664rXcvtERLdBcz/C7KhLqOsnLive9U1FUoT2P0KtGyopXQ/IxXztCDt8/F1Zf5s4BKwF5DQmouHruw0NjmrOMGqufwFVbZzqBR61UUxtymTKGl5sGaTrdye1f1FgYVX03+gPCb61ByvDTWqC+JELXof9MIyVINiN4E32VvChUJl3dVGkQFFIH7cjiWX6nDLYrmhx06YGVB8XNtonr6480G7SSfH8gG4CJHrFIgK1UXdk8VkNOI3edcPJ+UQKHnI4u+adMfQPVJ74vFTfJCUHvZXe7J76mG1DNdlfc3tO+1ZeBsN0t6GNKTtokQd6sXt+VR1TORQYWVfQy8IjB2LJfWXrm+JVo9UtR1OSLkicsgzfnW5UEZTdB5k+VW+mQnm09z3C72sfK3W3nfq53aorKqKGkV03kfVQ1H5RYXO0IOXz/XFh9mTtHrATkNCSg4spd6Fq8kXfUWJWnq224vcCDVsqsj7WRH/RUa76znfzk5EzUDeR7YLutQUpqD8wepAw/rVK+XoGuQ//ZRmqO6cGb6Kvkm7MyEvRTJT8oBrHrWn63DLYjSvJiQQZWHhS3YfhJPmEleXmgNXzaebhEgK3QXKTydYKfCvTPScpHIx7LK43EmpvhS56Ve5i8TTKh/LJVeadO7WV3krr3XzFbu3LraN/qy0DYRqFX/GFcbap0JsaSH/VDY/sivkU1vQw8YjS27FWWjll+JVrtDzh27KvWx+o+mgZvzrcqSYTO3mrHo2N6tnU9w+tq98Tzz2nvfq53aorKKKFkzwsS8md9P7MKPO0IOXz/XFhdu7jtznR4RgIixJvABBVX4xJ/W46NN2isPIGrnr9JgQet5FVHdoky9S63afOdvPfHyEe9TGG1NUhJzYHZg5Tup23KVypkR6b/HCMFOHnowZsQVXh5qmhZTJWk7ZqU4caudrbu1ZZB9sWziPqc+W9TGBWR6YsHdGBQfHJymjVwd0nvSHMsDM89Obnnl6qDvckM7Wl+I5RZk75uJzao3JxIH0FTftCZv/AQhrTb6o2Kq+KhxVbvnHSu+jcU/9bTqOvXh3rLqNbLwDZ/26t6tlHRm21z89zdinja1/sS8YT9mffWhpEheaY+cz9/qu8EfvcvIlgX8S0a6PUy8IjU0fL2BC3m1k6wP1k6ZvpV32p/Gc+O/bn1o/16ocsweKc8vZJGM1X43X4vIanzzjrf8wynq4FHPvMu7PBMrztX7ud4p6aoTB9KEW3uW/NYZ/fXh8b+tnaEHL5/Lqweme3OhJWAzIYEtC5Sv/FOT8PWue9uY/knyZopgiKw2wpPuO7ydIYXbOrp4QInDx/LalzjrXt5w2hrkJJ2G6QUPxUpv68QIUeeY6T8DQ7aVqqTWqq+c0SLPpWW6YFbs5xuB0XAid2IZfmdM9i+OQu/tlL7Y3tgUBxU/x4+QzqlfHi7Wq2y+KKqOHBysr2cfm82P6a/pdPzBaCN82XAkL91yMSzeuNFXurfmzlbTU/xm6qrykOrae3ZYyyWiwgnJ/exJ5vNw/QsJuxKLyPnf6e3zW+1uZnY2tf7QnjC/n0IBT9TQ0E7ZgI9Ob2Yfmw2T9Ol2gPjIp5FiU4vA484DQeqPmq23pssHdq1ErVfiVbFAduOomp1b9Ue1g3ena9XUnl5nILK3qZLUTVIaGgk4XiG3dXIZbjo02O65Lnou+2dmqK+ynN4D2lzMz2KKLG1I+Tw/XNh9cRcd2asBGQ1JAiG6DwmOXjdoNvYfazeGVsT2GtlmxqpjiQpuiYWOHlo4m+64Kca5HpbreCiG9WBBYPUV5llpJOTcERNWYXQUrsJqNd7OtIbyAyKiB27CcPyu2cwwKTVyGryy281NrW+jFhfBeB/Ci3Dd7/0CQD4bhxpkALAIa3MVVMY2vfHv+q5JzBVAt8K/tEA/cEoAODbcKRBCgAbegO1+o0+WttUto58BUyVwLeC96Jqj1oAAN+HYw1SADjQFlB+pSN8tD9ivcexI5gqgW8F7Z7E8zcAvjtHGqQAcOA3J9e/w3fgf/PLM8YvFO8MpkrgO8FvSOP5GwDfnSMNUgA48CciWtbaewxfAVMl8J3g12X7108AAN+NIw1SAHiIj92134zeE5gqgW8Eb+rGZgcAvj/HGaQAcLnov8h1iK9SYaoEvhH8Hb59f6gWAHAAjjJIATDgufp65/Wfi0M8AaafPhUfwQXgfwD97tUfpFsA/gmOMEgBMOb0OWyYm/zPB3+F88dpej9Y6wAs4vbvND3u9we1AQCH5OCDFAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA/Lf5P+4JRd5QoXkFAAAAAElFTkSuQmCC"}}},{"cell_type":"markdown","source":"##  1.Dataset Overview","metadata":{}},{"cell_type":"markdown","source":"##### In this project, a dataset containing customer reviews and opinions written in the Saudi dialect was used. The texts were collected from e-commerce websites,Sentiment classification was performed using the Aspect-Based Sentiment Analysis (ABSA) approach, which analyzes sentiment for each specific aspect within the text rather than evaluating the sentiment of the entire text as a whole,The aspects include, for example, price, quality, size, and others. The dataset consists of 12 different aspects,Sentiment is represented using numerical values to accurately determine the sentiment polarity of each aspect within the text instead of assigning a general sentiment to the entire review,The dataset used in this project was obtained from IEEE DataPort.\n<span style=\"color:green; font-weight:bold;\">+1</span> represents positive sentiment  \n<br>\n<span style=\"color:green; font-weight:bold;\">0</span> represents neutral sentiment  \n<br>\n<span style=\"color:green; font-weight:bold;\">-1</span> represents negative sentiment","metadata":{}},{"cell_type":"code","source":"import pandas as pd\nimport numpy as np\nimport torch\nimport matplotlib.pyplot as plt \nimport seaborn as sns \nimport nltk as nltk \nimport re\nimport torch\nimport torch.nn as nn\nfrom torch.optim import AdamW\nfrom transformers import get_linear_schedule_with_warmup\nfrom tqdm import tqdm\nimport matplotlib.pyplot as plt\nfrom kaggle_secrets import UserSecretsClient\nfrom huggingface_hub import login\nfrom sklearn.metrics import accuracy_score, classification_report\nfrom torch.utils.data import Dataset, DataLoader\nfrom transformers import AutoTokenizer, AutoModelForSequenceClassification\nfrom torch.utils.data import Dataset, DataLoader\nfrom transformers import AutoTokenizer, AutoModel\nfrom sklearn.model_selection import train_test_split\nfrom torch.optim import AdamW\nfrom tqdm import tqdm\nimport torch\nfrom transformers import get_linear_schedule_with_warmup\nfrom torch.utils.data import Dataset, DataLoader\nfrom transformers import AutoTokenizer, AutoModelForSequenceClassification\nimport string\nfrom datasets import Dataset\n\nfrom sklearn.metrics import (\n    accuracy_score,\n    precision_score,\n    recall_score,\n    f1_score,\n    classification_report\n)\n\ndevice = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\nprint(f\"Using device: {device}\")\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-04-26T12:13:12.621526Z","iopub.execute_input":"2026-04-26T12:13:12.621878Z","iopub.status.idle":"2026-04-26T12:13:37.697871Z","shell.execute_reply.started":"2026-04-26T12:13:12.621837Z","shell.execute_reply":"2026-04-26T12:13:37.696797Z"}},"outputs":[{"name":"stdout","text":"Using device: cpu\n","output_type":"stream"}],"execution_count":3},{"cell_type":"code","source":"try:\n    user_secrets = UserSecretsClient()\n    hf_token = user_secrets.get_secret(\"HF_TOKEN\")\n    \n    # تسجيل الدخول\n    login(token=hf_token)\n    \n    print(\"done  Hugging Face\")\n    \nexcept Exception as e:\n    print(f\"❌ حدث خطأ: {e}\")\n    print(\"تأكدي من تفعيل علامة الصح بجانب الـ Secret في القائمة العلوية Add-ons -> Secrets\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-04-16T12:25:33.619032Z","iopub.execute_input":"2026-04-16T12:25:33.620188Z","iopub.status.idle":"2026-04-16T12:25:33.913484Z","shell.execute_reply.started":"2026-04-16T12:25:33.620136Z","shell.execute_reply":"2026-04-16T12:25:33.912571Z"}},"outputs":[{"name":"stdout","text":"✅ تم تسجيل الدخول بنجاح إلى Hugging Face\n","output_type":"stream"}],"execution_count":2},{"cell_type":"markdown","source":"## 2. Reading the Dataset","metadata":{}},{"cell_type":"code","source":"train_df = pd.read_csv(\"/kaggle/input/datasets/areejhaddaj/finalwork/train_finalsmell.csv\", sep=';')\nval_df = pd.read_csv(\"/kaggle/input/datasets/areejhaddaj/finalwork/val_finalsmeel.csv\", sep=';')\ntest_df = pd.read_csv(\"/kaggle/input/datasets/areejhaddaj/finalwork/test_finalsmell.csv\", sep=';')\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-04-26T12:12:58.955283Z","iopub.execute_input":"2026-04-26T12:12:58.955672Z","iopub.status.idle":"2026-04-26T12:12:58.963613Z","shell.execute_reply.started":"2026-04-26T12:12:58.955641Z","shell.execute_reply":"2026-04-26T12:12:58.962215Z"}},"outputs":[{"traceback":["\u001b[0;31m---------------------------------------------------------------------------\u001b[0m","\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)","\u001b[0;32m/tmp/ipykernel_55/387902370.py\u001b[0m in \u001b[0;36m<cell line: 0>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mtrain_df\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mpd\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mread_csv\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"/kaggle/input/datasets/areejhaddaj/finalwork/train_finalsmell.csv\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msep\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m';'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      2\u001b[0m \u001b[0mval_df\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mpd\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mread_csv\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"/kaggle/input/datasets/areejhaddaj/finalwork/val_finalsmeel.csv\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msep\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m';'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      3\u001b[0m \u001b[0mtest_df\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mpd\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mread_csv\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"/kaggle/input/datasets/areejhaddaj/finalwork/test_finalsmell.csv\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msep\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m';'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      4\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n","\u001b[0;31mNameError\u001b[0m: name 'pd' is not defined"],"ename":"NameError","evalue":"name 'pd' is not defined","output_type":"error"}],"execution_count":2},{"cell_type":"code","source":"for df in [train_df, val_df, test_df]:\n    df.replace(r'^\\s*$', np.nan, regex=True, inplace=True)\n    df['Review'] = df['Review'].fillna(\"\").astype(str)\n    df.isnull().sum()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-04-16T12:25:33.973341Z","iopub.execute_input":"2026-04-16T12:25:33.974076Z","iopub.status.idle":"2026-04-16T12:25:33.991073Z","shell.execute_reply.started":"2026-04-16T12:25:33.974023Z","shell.execute_reply":"2026-04-16T12:25:33.990012Z"}},"outputs":[],"execution_count":4},{"cell_type":"code","source":"emoji_pattern = re.compile(\n    \"[\"\n    u\"\\U0001F600-\\U0001F64F\"\n    u\"\\U0001F300-\\U0001F5FF\"\n    u\"\\U0001F680-\\U0001F6FF\"\n    u\"\\U0001F1E0-\\U0001F1FF\"\n    u\"\\U00002700-\\U000027BF\"\n    u\"\\U000024C2-\\U0001F251\"\n    \"]+\", flags=re.UNICODE\n)\n\ndef full_clean(text):\n    text = str(text)\n    text = re.sub(r'http\\S+|www\\S+|https\\S+', '', text)\n    text = re.sub(r'(.)\\1{2,}', r'\\1\\1', text)\n    text = emoji_pattern.sub(r'', text)\n    text = \"\".join([char for char in text if char not in string.punctuation]) \n    text = re.sub(r'[إأآا]', 'ا', text)\n    text = re.sub(r'ى', 'ي', text)\n    text = re.sub(r'ة', 'ه', text)\n    text = re.sub(r'[ًٌٍَُِّْ]', '', text)  \n    text = re.sub(r'ـ', '', text)            \n    text = re.sub(r'\\s+', ' ', text).strip() \n    text = re.sub(r'(.)\\1{2,}', r'\\1\\1', text)\n    return text\n\n\ntrain_df['Review_after_clean'] = train_df['Review'].apply(full_clean)\nval_df['Review_after_clean'] = val_df['Review'].apply(full_clean)\ntest_df['Review_after_clean'] = test_df['Review'].apply(full_clean)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-04-16T12:25:33.992270Z","iopub.execute_input":"2026-04-16T12:25:33.992651Z","iopub.status.idle":"2026-04-16T12:25:34.133216Z","shell.execute_reply.started":"2026-04-16T12:25:33.992601Z","shell.execute_reply":"2026-04-16T12:25:34.132586Z"}},"outputs":[],"execution_count":5},{"cell_type":"code","source":"df[\"Review_after_clean\"].isnull().sum()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-04-16T12:25:34.134164Z","iopub.execute_input":"2026-04-16T12:25:34.134531Z","iopub.status.idle":"2026-04-16T12:25:34.141076Z","shell.execute_reply.started":"2026-04-16T12:25:34.134508Z","shell.execute_reply":"2026-04-16T12:25:34.140072Z"}},"outputs":[{"execution_count":6,"output_type":"execute_result","data":{"text/plain":"np.int64(0)"},"metadata":{}}],"execution_count":6},{"cell_type":"code","source":"aspect_columns = ['Size', 'Color', 'Price', 'Smell', 'Quality', 'Fabric', 'Style', 'Image']\n\ndef prepare_xy(df):\n    X = df['Review_after_clean']\n    y = df[aspect_columns].fillna(0).astype(int)\n    y = y + 1 # تحويل (-1, 0, 1) إلى (0, 1, 2)\n    return X.reset_index(drop=True), y.reset_index(drop=True)\n\nX_train, y_train = prepare_xy(train_df)\nX_val, y_val = prepare_xy(val_df)\nX_test, y_test = prepare_xy(test_df)\n\nnum_aspects = 8\nnum_classes = 3","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-04-16T12:25:34.142076Z","iopub.execute_input":"2026-04-16T12:25:34.142461Z","iopub.status.idle":"2026-04-16T12:25:34.160911Z","shell.execute_reply.started":"2026-04-16T12:25:34.142435Z","shell.execute_reply":"2026-04-16T12:25:34.160206Z"}},"outputs":[],"execution_count":7},{"cell_type":"code","source":"model_name = \"faisalq/SaudiBERT\"\ntokenizer = AutoTokenizer.from_pretrained(model_name)\n\nmodel = AutoModelForSequenceClassification.from_pretrained(\n    model_name, \n    num_labels=num_aspects * num_classes\n)\nmodel = model.to(device)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-04-16T12:25:34.161960Z","iopub.execute_input":"2026-04-16T12:25:34.162358Z","iopub.status.idle":"2026-04-16T12:25:42.715801Z","shell.execute_reply.started":"2026-04-16T12:25:34.162327Z","shell.execute_reply":"2026-04-16T12:25:42.715062Z"}},"outputs":[{"output_type":"display_data","data":{"text/plain":"config.json:   0%|          | 0.00/630 [00:00<?, ?B/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"7404af77736940549076b0c6ee85d3dd"}},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"tokenizer_config.json: 0.00B [00:00, ?B/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"85cab1a79bdf47d4a2f17a9e6e309d04"}},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"tokenizer.json: 0.00B [00:00, ?B/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"523388d9eb37408eb45b883a04d901e1"}},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"special_tokens_map.json:   0%|          | 0.00/970 [00:00<?, ?B/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"9013ecd9e480490c9a9861817454ae7a"}},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"model.safetensors:   0%|          | 0.00/575M [00:00<?, ?B/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"fd5d3af71b6d4a0cbda7f83a10cf5e29"}},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"Loading weights:   0%|          | 0/197 [00:00<?, ?it/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"7655daea169546d1bf2e2a32aec5e94d"}},"metadata":{}},{"name":"stderr","text":"BertForSequenceClassification LOAD REPORT from: faisalq/SaudiBERT\nKey                                        | Status     | \n-------------------------------------------+------------+-\ncls.predictions.transform.LayerNorm.weight | UNEXPECTED | \ncls.predictions.bias                       | UNEXPECTED | \ncls.predictions.transform.dense.weight     | UNEXPECTED | \ncls.predictions.transform.dense.bias       | UNEXPECTED | \ncls.predictions.transform.LayerNorm.bias   | UNEXPECTED | \nclassifier.bias                            | MISSING    | \nbert.pooler.dense.bias                     | MISSING    | \nbert.pooler.dense.weight                   | MISSING    | \nclassifier.weight                          | MISSING    | \n\nNotes:\n- UNEXPECTED\t:can be ignored when loading from different task/architecture; not ok if you expect identical arch.\n- MISSING\t:those params were newly initialized because missing from the checkpoint. Consider training on your downstream task.\n","output_type":"stream"}],"execution_count":8},{"cell_type":"code","source":"class ABSADataset(Dataset): \n    def __init__(self, texts, labels, tokenizer, max_len):\n        self.texts = texts.values \n        self.labels = labels.values\n        self.tokenizer = tokenizer\n        self.max_len = max_len\n\n    def __len__(self):\n        return len(self.texts) \n\n    def __getitem__(self, idx):\n        text = str(self.texts[idx])\n        label = self.labels[idx]\n\n        encoding = self.tokenizer(\n            text,\n            max_length=self.max_len,\n            padding='max_length',\n            truncation=True,\n            return_tensors='pt'\n        )\n\n        return {\n            'input_ids': encoding['input_ids'].flatten(),\n            'attention_mask': encoding['attention_mask'].flatten(),\n            'labels': torch.tensor(label, dtype=torch.long)\n        }\n\n\ntrain_dataset = ABSADataset(X_train, y_train, tokenizer, max_len=128)\nval_dataset = ABSADataset(X_val, y_val, tokenizer, max_len=128)\ntest_dataset = ABSADataset(X_test, y_test, tokenizer, max_len=128)\n\ntrain_loader = DataLoader(train_dataset, batch_size=16, shuffle=True)\nval_loader = DataLoader(val_dataset, batch_size=16, shuffle=False)\ntest_loader = DataLoader(test_dataset, batch_size=16, shuffle=False)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-04-16T12:25:42.716809Z","iopub.execute_input":"2026-04-16T12:25:42.717652Z","iopub.status.idle":"2026-04-16T12:25:42.725612Z","shell.execute_reply.started":"2026-04-16T12:25:42.717627Z","shell.execute_reply":"2026-04-16T12:25:42.724865Z"}},"outputs":[],"execution_count":9},{"cell_type":"code","source":"\nlearning_rate = 5e-6       \nepochs = 8                \nweight_decay = 0.05        \n\noptimizer = AdamW(model.parameters(), lr=learning_rate, weight_decay=weight_decay)\nloss_fn = nn.CrossEntropyLoss()\n\nnum_training_steps = len(train_loader) * epochs\nnum_warmup_steps = int(0.3 * num_training_steps) \n\nscheduler = get_linear_schedule_with_warmup(\n    optimizer, \n    num_warmup_steps=num_warmup_steps, \n    num_training_steps=num_training_steps\n)\n\ntrain_losses, val_losses = [], []\nbest_val_loss = float('inf') \npatience = 3  \ncounter = 0\n\n\nfor epoch in range(epochs):\n    model.train()\n    total_train_loss = 0\n    \n    progress_bar = tqdm(train_loader, desc=f\"Epoch {epoch+1}/{epochs}\")\n    for batch in progress_bar:\n        optimizer.zero_grad()\n        \n        ids = batch['input_ids'].to(device)\n        mask = batch['attention_mask'].to(device)\n        labels = batch['labels'].to(device)\n        \n        outputs = model(ids, attention_mask=mask)\n        logits = outputs.logits.view(-1, num_aspects, 3) \n        \n        loss = sum([loss_fn(logits[:, i, :], labels[:, i]) for i in range(num_aspects)])\n        \n        loss.backward()\n        \n        torch.nn.utils.clip_grad_norm_(model.parameters(), 1.0)\n        \n        optimizer.step()\n        scheduler.step()\n        \n        total_train_loss += loss.item()\n        progress_bar.set_postfix({'loss': loss.item()})\n\n    avg_train_loss = total_train_loss / len(train_loader)\n    \n    model.eval()\n    total_val_loss = 0\n    with torch.no_grad():\n        for batch in val_loader:\n            ids = batch['input_ids'].to(device)\n            mask = batch['attention_mask'].to(device)\n            labels = batch['labels'].to(device)\n            \n            outputs = model(ids, attention_mask=mask)\n            logits = outputs.logits.view(-1, num_aspects, 3)\n            \n            v_loss = sum([loss_fn(logits[:, i, :], labels[:, i]) for i in range(num_aspects)])\n            total_val_loss += v_loss.item()\n\n    avg_val_loss = total_val_loss / len(val_loader)\n    \n    train_losses.append(avg_train_loss)\n    val_losses.append(avg_val_loss)\n\n    print(f\" Epoch {epoch+1}: Train Loss = {avg_train_loss:.4f} | Val Loss = {avg_val_loss:.4f}\")\n\n    # التوقف التلقائي وحفظ أفضل موديل\n    if avg_val_loss < best_val_loss:\n        best_val_loss = avg_val_loss\n        torch.save(model.state_dict(), 'best_model.pt')\n        counter = 0\n    else:\n        counter += 1\n        if counter >= patience:\n            print(f\" Early Stopping) {epoch+1}\")\n            break\n\nplt.figure(figsize=(10, 5))\nplt.plot(train_losses, label='Training Loss', marker='o')\nplt.plot(val_losses, label='Validation Loss', marker='o')\nplt.title('Stable & Smooth Training Process')\nplt.xlabel('Epochs')\nplt.ylabel('Loss')\nplt.legend()\nplt.grid(True)\nplt.show()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-04-16T12:25:42.727748Z","iopub.execute_input":"2026-04-16T12:25:42.728044Z","iopub.status.idle":"2026-04-16T12:38:34.741121Z","shell.execute_reply.started":"2026-04-16T12:25:42.727976Z","shell.execute_reply":"2026-04-16T12:38:34.740422Z"}},"outputs":[{"name":"stdout","text":"🚀 بدء التدريب الهادئ بـ 8 إيبوك...\n","output_type":"stream"},{"name":"stderr","text":"Epoch 1/8: 100%|██████████| 243/243 [01:25<00:00,  2.85it/s, loss=4.98]\n","output_type":"stream"},{"name":"stdout","text":"\n✅ Epoch 1: Train Loss = 7.0399 | Val Loss = 4.6941\n💎 تم حفظ أفضل نسخة للموديل!\n","output_type":"stream"},{"name":"stderr","text":"Epoch 2/8: 100%|██████████| 243/243 [01:33<00:00,  2.61it/s, loss=5.79]\n","output_type":"stream"},{"name":"stdout","text":"\n✅ Epoch 2: Train Loss = 4.2758 | Val Loss = 3.3219\n💎 تم حفظ أفضل نسخة للموديل!\n","output_type":"stream"},{"name":"stderr","text":"Epoch 3/8: 100%|██████████| 243/243 [01:32<00:00,  2.62it/s, loss=4.53]\n","output_type":"stream"},{"name":"stdout","text":"\n✅ Epoch 3: Train Loss = 2.9690 | Val Loss = 2.4140\n💎 تم حفظ أفضل نسخة للموديل!\n","output_type":"stream"},{"name":"stderr","text":"Epoch 4/8: 100%|██████████| 243/243 [01:32<00:00,  2.62it/s, loss=2.45]\n","output_type":"stream"},{"name":"stdout","text":"\n✅ Epoch 4: Train Loss = 2.2660 | Val Loss = 2.1799\n💎 تم حفظ أفضل نسخة للموديل!\n","output_type":"stream"},{"name":"stderr","text":"Epoch 5/8: 100%|██████████| 243/243 [01:32<00:00,  2.62it/s, loss=1.18]\n","output_type":"stream"},{"name":"stdout","text":"\n✅ Epoch 5: Train Loss = 1.9072 | Val Loss = 2.0278\n💎 تم حفظ أفضل نسخة للموديل!\n","output_type":"stream"},{"name":"stderr","text":"Epoch 6/8: 100%|██████████| 243/243 [01:32<00:00,  2.62it/s, loss=1.27]\n","output_type":"stream"},{"name":"stdout","text":"\n✅ Epoch 6: Train Loss = 1.6922 | Val Loss = 1.9666\n💎 تم حفظ أفضل نسخة للموديل!\n","output_type":"stream"},{"name":"stderr","text":"Epoch 7/8: 100%|██████████| 243/243 [01:32<00:00,  2.62it/s, loss=1.73] \n","output_type":"stream"},{"name":"stdout","text":"\n✅ Epoch 7: Train Loss = 1.5580 | Val Loss = 1.9331\n💎 تم حفظ أفضل نسخة للموديل!\n","output_type":"stream"},{"name":"stderr","text":"Epoch 8/8: 100%|██████████| 243/243 [01:32<00:00,  2.61it/s, loss=1.63] \n","output_type":"stream"},{"name":"stdout","text":"\n✅ Epoch 8: Train Loss = 1.4897 | Val Loss = 1.9248\n💎 تم حفظ أفضل نسخة للموديل!\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"<Figure size 1000x500 with 1 Axes>","image/png":"iVBORw0KGgoAAAANSUhEUgAAA0EAAAHWCAYAAACxAYILAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjAsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvlHJYcgAAAAlwSFlzAAAPYQAAD2EBqD+naQAAj/dJREFUeJzs3Xd4FNX+x/H37KZ3EkJCCSHUEECkiXSQDgIqiqIIqCgXQa96/Ylcrwr2XlHsICp2sdIVpEtV6TWE0EtIQnrZ+f2xJBASIGCS2SSf1/Psk+zs7Mx39ySwn5w55ximaZqIiIiIiIhUEjarCxARERERESlLCkEiIiIiIlKpKASJiIiIiEilohAkIiIiIiKVikKQiIiIiIhUKgpBIiIiIiJSqSgEiYiIiIhIpaIQJCIiIiIilYpCkIiIiIiIVCoKQSIip0ybNg3DMFizZs0F9+3atStdu3Yt/aKEPXv2YBgGL730ktWlnNPEiRMxDOOSnpv3c7dnz56SLUpERM5JIUhEXN6GDRu4/vrriYyMxMvLi5o1a9KzZ0/efPPNAvs988wzfP/999YUWQIcDgcvvfQSDRo0wNvbm3r16jFmzBhSUlIu6jhLly6lb9++1KxZEy8vL2rXrs2AAQOYMWNGKVVeMmbNmsXEiRNL9Jh16tTBMIwL3qZNm1ai5y0v8sJb3s3Hx4eYmBj+97//kZycbHV5IiKlxjBN07S6CBGRc1m+fDndunWjdu3ajBgxgvDwcOLj41m5ciW7du1i586d+fv6+flx/fXXX/IH2mnTpnHbbbexevVqWrdufd5983qBFi1adEnnKsqrr77KAw88wDXXXEPfvn2Ji4vj888/57fffqNOnTrFOsbXX3/NjTfeyOWXX85NN91ElSpViI2NZfHixbi7u7Nw4cISq7ekjRs3jrfeeouz/1vas2cPUVFRvPjiizz44IMXdczvv/++QIicNWsWn3/+Oa+++ipVq1bN396+fXvq1q17ybXn5OSQk5ODl5fXRT83NzeX7OxsPD09L7k36VJNnDiRSZMmMWXKFPz8/EhJSWHevHnMnDmTdu3asWzZsjKvSUSkLLhZXYCIyPk8/fTTBAYGsnr1aoKCggo8duTIEWuKKiVffPEFTZo04bvvvsv/4Pnkk0/icDiKfYyJEycSExPDypUr8fDwKPBYRXu/iuOaa64pcP/QoUN8/vnnXHPNNecNlqmpqfj6+hb7PG5ubri5Xdp/qXa7HbvdfknPLSnXX399fij817/+xeDBg/nuu+9YuXIl7dq1K/I5aWlp+Pj4lGWZIiIlRpfDiYhL27VrF02aNCkUgACqVauW/71hGKSmpvLxxx/nX9ozcuRIAOLi4rj77rtp1KgR3t7ehISEcMMNN5xzDEZaWhqjR48mJCSEgIAAhg8fzokTJy5Ya2ZmJo8//jj169fH09OTiIgIHnroITIzM4v1Wm02Gw6Ho8Bf3m0220V9uN61axdt2rQpFICg4Pt15jibt956i7p16+Lj40OvXr2Ij4/HNE2efPJJatWqhbe3N4MGDSIhIaHQMd9++22aNGmCp6cnNWrUYOzYsSQmJhba7+uvv6ZVq1Z4e3tTtWpVhg0bxv79+/MfHzlyJG+99RZAgcuzzvbee+9Rr149PD09adOmDatXry72e3MuI0eOxM/Pj127dtGvXz/8/f255ZZbAFiyZAk33HADtWvXzm/T+++/n/T09ALHKGpMkGEYjBs3ju+//56mTZvi6elJkyZNmDNnToH9ihoTVKdOHa6++mqWLl3KFVdcgZeXF3Xr1mX69OmF6v/777/p0qUL3t7e1KpVi6eeeoqpU6f+o3FGV111FQCxsbGAs+ezadOmrF27ls6dO+Pj48N///tfwBmu77jjDsLCwvDy8qJ58+Z8/PHHhY7pcDh4/fXXadasGV5eXoSGhtKnT59CY/A+/fTT/J+V4OBgbrrpJuLj4wvss2PHDgYPHkx4eDheXl7UqlWLm266iaSkpPx95s+fT8eOHQkKCsLPz49GjRrl1ywiop4gEXFpkZGRrFixgo0bN9K0adNz7vfJJ58watQorrjiCu666y4A6tWrB8Dq1atZvnw5N910E7Vq1WLPnj1MmTKFrl27snnz5kJ/zR43bhxBQUFMnDiRbdu2MWXKFOLi4li0aNE5Lw1yOBwMHDiQpUuXctddd9G4cWM2bNjAq6++yvbt24s1Vum2225j9OjRvPvuu4wePbqY71BBkZGR/Prrr+zbt49atWpdcP/PPvuMrKws7rnnHhISEnjhhRcYMmQIV111FYsWLWL8+PHs3LmTN998kwcffJCPPvoo/7l5l1L16NGDMWPG5L9Xq1evZtmyZbi7uwOnLzNs06YNzz77LIcPH+b1119n2bJlrF+/nqCgIEaPHs2BAweYP38+n3zySZG1zpgxg5MnTzJ69GgMw+CFF17guuuuY/fu3fnnulQ5OTn07t2bjh078tJLL+X/THz99dekpaUxZswYQkJCWLVqFW+++Sb79u3j66+/vuBxly5dynfffcfdd9+Nv78/b7zxBoMHD2bv3r2EhISc97k7d+7k+uuv54477mDEiBF89NFHjBw5klatWtGkSRMA9u/fT7du3TAMgwkTJuDr68sHH3yAp6fnP3o/du3aBVCgxuPHj9O3b19uuukmhg0bRlhYGOnp6XTt2pWdO3cybtw4oqKi+Prrrxk5ciSJiYn8+9//zn/+HXfcwbRp0+jbty+jRo0iJyeHJUuWsHLlyvzLT59++mkeffRRhgwZwqhRozh69ChvvvkmnTt3zv9ZycrKonfv3mRmZnLPPfcQHh7O/v37+fnnn0lMTCQwMJBNmzZx9dVXc9lll/HEE0/g6enJzp07WbZs2T96X0SkAjFFRFzYvHnzTLvdbtrtdrNdu3bmQw89ZM6dO9fMysoqtK+vr685YsSIQtvT0tIKbVuxYoUJmNOnT8/fNnXqVBMwW7VqVeD4L7zwggmYP/zwQ/62Ll26mF26dMm//8knn5g2m81csmRJgfO88847JmAuW7bsgq/14YcfNj08PEy73W5+9913F9y/KB9++KEJmB4eHma3bt3MRx991FyyZImZm5tbYL/Y2FgTMENDQ83ExMT87RMmTDABs3nz5mZ2dnb+9qFDh5oeHh5mRkaGaZqmeeTIEdPDw8Ps1atXgWNPnjzZBMyPPvrINE3TzMrKMqtVq2Y2bdrUTE9Pz9/v559/NgHzsccey982duxYs6j/lvJqDQkJMRMSEvK3//DDDyZg/vTTT8V+f1588UUTMGNjY/O3jRgxwgTMhx9+uND+Rf3sPPvss6ZhGGZcXFz+tscff7xQ7XntsHPnzvxtf/31lwmYb775Zv62vJ+7M2uKjIw0AXPx4sX5244cOWJ6enqa//nPf/K33XPPPaZhGOb69evztx0/ftwMDg4udMyi5NW9bds28+jRo2ZsbKz57rvvmp6enmZYWJiZmppqmqbz5x0w33nnnQLPf+2110zA/PTTT/O3ZWVlme3atTP9/PzM5ORk0zRN87fffjMB89577y1Ug8PhME3TNPfs2WPa7Xbz6aefLvD4hg0bTDc3t/zt69evNwHz66+/PufrevXVV03APHr06Hlfv4hUXrocTkRcWs+ePVmxYgUDBw7kr7/+4oUXXqB3797UrFmTH3/8sVjH8Pb2zv8+Ozub48ePU79+fYKCgli3bl2h/e+6664CPQtjxozBzc2NWbNmnfMcX3/9NY0bNyY6Oppjx47l3/IuK7rQhARvvPEGr7zyCsuWLWPo0KHcdNNNzJs3r8A+np6ePProo+c9zu23386cOXPo2rUrS5cu5cknn6RTp040aNCA5cuXF9r/hhtuIDAwMP9+27ZtARg2bFiBy/Datm1LVlZW/iVsCxYsICsri/vuuw+b7fR/JXfeeScBAQH88ssvAKxZs4YjR45w9913F5g0oH///kRHR+fvVxw33ngjVapUyb/fqVMnAHbv3l3sY5zPmDFjCm0782cnNTWVY8eO0b59e0zTZP369Rc8Zo8ePfJ7JAEuu+wyAgICilVzTExM/msECA0NpVGjRgWeO2fOHNq1a8fll1+evy04ODj/cr7iatSoEaGhoURFRTF69Gjq16/PL7/8UqCX1NPTk9tuu63A82bNmkV4eDhDhw7N3+bu7s69995LSkoKv//+OwDffvsthmHw+OOPFzp3Xu/qd999h8PhYMiQIQV+h8LDw2nQoEH+71Dez+vcuXNJS0sr8vXkXT77ww8/XNSYOhGpPBSCRMTltWnThu+++44TJ06watUqJkyYwMmTJ7n++uvZvHnzBZ+fnp7OY489RkREBJ6enlStWpXQ0FASExMLjCHI06BBgwL3/fz8qF69+nnHV+zYsYNNmzYRGhpa4NawYUPg/JMSpKen8/jjjzNq1Chat27N1KlTueqqq7j22mtZunRp/vGzsrLyQ8r59O7dm7lz55KYmMjixYsZO3YscXFxXH311YXqqF27doH7eR8wIyIiityeNzYqLi4OcH54PpOHhwd169bNf/xc+wFER0fnP14cZ9eaF4iKM17rQtzc3Iq8fHDv3r2MHDmS4OBg/Pz8CA0NpUuXLgBF/uxcqGZw1l2cmovz3Li4OOrXr19ov6K2nc+3337L/PnzWbRoETt37mTjxo20atWqwD41a9YsNNYsLi6OBg0aFAjCAI0bN85/HJyX19WoUYPg4OBz1rBjxw5M06RBgwaFfo+2bNmS/7MbFRXFAw88wAcffEDVqlXp3bs3b731VoH2uPHGG+nQoQOjRo0iLCyMm266ia+++kqBSETyaUyQiJQbHh4etGnThjZt2tCwYUNuu+02vv766yL/unyme+65h6lTp3LffffRrl07AgMDMQyDm266qcQ+FDkcDpo1a8Yrr7xS5ONnh4ozbdmyhcTERK688krA+YH8m2++4aqrrqJ///4sXLiQzz//nGrVqtGzZ89i1+Tj40OnTp3o1KkTVatWZdKkScyePZsRI0bk73OuWcnOtd20cFWF0qzJ09Oz0Af53NxcevbsSUJCAuPHjyc6OhpfX1/279/PyJEji/Wz809qLss26Ny5c4Epw4tyZq9YacibFGT27NlFvnY/P7/8719++WVGjhzJDz/8wLx587j33nt59tlnWblyZf5kHosXL2bhwoX88ssvzJkzhy+//JKrrrqKefPmWT4bn4hYTyFIRMqlvIHUBw8ezN92rkkLvvnmG0aMGMHLL7+cvy0jI6PIWczA+Rfpbt265d9PSUnh4MGD9OvX75z11KtXj7/++ovu3btf9LoqefufOQOWr68vs2bNomPHjvTu3ZuMjAyeeuqpSx7wXtT79U9ERkYCsG3btgLr62RlZREbG0uPHj0K7Zd3aWCebdu25T8O524/q2zYsIHt27fz8ccfM3z48Pzt8+fPt7CqgiIjIwuslZWnqG2ldf6///4bh8NRIERu3bo1/3Fw/n7MnTuXhISEc/YG1atXD9M0iYqKyu9BPZ9mzZrRrFkz/ve//7F8+XI6dOjAO++8w1NPPQU4Z1bs3r073bt355VXXuGZZ57hkUceYeHChfk/nyJSeelyOBFxaQsXLizyL99543POvMzK19e3yGBjt9sLHePNN98kNze3yHO+9957ZGdn59+fMmUKOTk59O3b95x1DhkyhP379/P+++8Xeiw9PZ3U1NRzPrdZs2aEhYUxefLkAperhYSEMHXqVI4dO0Z6ejoDBgw45zHy/Prrr0VuL+r9+id69OiBh4cHb7zxRoH39sMPPyQpKYn+/fsDzvBVrVo13nnnnQJThc+ePZstW7bk7wfkr8tzrnBa1vJ6C858faZp8vrrr1tVUiG9e/dmxYoV/Pnnn/nbEhIS+Oyzz8rk/P369ePQoUN8+eWX+dtycnJ488038fPzy790cPDgwZimyaRJkwodI+/9ve6667Db7UyaNKnQ76tpmhw/fhyA5ORkcnJyCjzerFkzbDZb/s9YUdO5542bKu6U9SJSsaknSERc2j333ENaWhrXXnst0dHRZGVlsXz5cr788kvq1KlTYKB2q1atWLBgAa+88go1atQgKiqKtm3bcvXVV/PJJ58QGBhITEwMK1asYMGCBeecojgrK4vu3bszZMgQtm3bxttvv03Hjh0ZOHDgOeu89dZb+eqrr/jXv/7FwoUL6dChA7m5uWzdupWvvvqKuXPn5vfGnM3NzY3Jkydz44030qxZM0aPHk1kZCRbtmzho48+olmzZuzbt49BgwaxbNkyAgICzlnHoEGDiIqKYsCAAdSrV4/U1FQWLFjATz/9RJs2bYoVpIojNDSUCRMmMGnSJPr06cPAgQPz36s2bdowbNgwwDlI/vnnn+e2226jS5cuDB06NH+K7Dp16nD//ffnHzNvDMq9995L7969sdvt3HTTTSVS76WIjo6mXr16PPjgg+zfv5+AgAC+/fbbEhmDVFIeeughPv30U3r27Mk999yTP0V27dq1SUhIKPXetbvuuot3332XkSNHsnbtWurUqcM333zDsmXLeO211/D39wegW7du3Hrrrbzxxhvs2LGDPn364HA4WLJkCd26dWPcuHHUq1ePp556igkTJrBnzx6uueYa/P39iY2NZebMmdx11108+OCD/Pbbb4wbN44bbriBhg0bkpOTwyeffILdbmfw4MEAPPHEEyxevJj+/fsTGRnJkSNHePvtt6lVqxYdO3Ys1fdERMqJsp+QTkSk+GbPnm3efvvtZnR0tOnn52d6eHiY9evXN++55x7z8OHDBfbdunWr2blzZ9Pb29sE8qfLPnHihHnbbbeZVatWNf38/MzevXubW7duNSMjIwtMqZ03VfHvv/9u3nXXXWaVKlVMPz8/85ZbbjGPHz9e4FxnT5Ftms6pgZ9//nmzSZMmpqenp1mlShWzVatW5qRJk8ykpKQLvtbFixebvXv3NgMCAkxPT0+zadOm5rPPPmumpaWZs2fPNm02m9mrV68CU1ef7fPPPzdvuukms169eqa3t7fp5eVlxsTEmI888kj+dMWmeXra6RdffLHA8xcuXFjk9MN5783q1asLbJ88ebIZHR1turu7m2FhYeaYMWPMEydOFKrryy+/NFu0aGF6enqawcHB5i233GLu27evwD45OTnmPffcY4aGhpqGYeRPOX2uWk3TOQ31448/fs7342znmiLb19e3yP03b95s9ujRw/Tz8zOrVq1q3nnnnfnTXE+dOjV/v3NNkT127NhCxzzXz93ZU2T379+/0HOL+rlbv3692alTJ9PT09OsVauW+eyzz5pvvPGGCZiHDh0695txRt0Xmkq6S5cuZpMmTYp87PDhw/m/Xx4eHmazZs0KvDd5cnJyzBdffNGMjo42PTw8zNDQULNv377m2rVrC+z37bffmh07djR9fX1NX19fMzo62hw7dqy5bds20zRNc/fu3ebtt99u1qtXz/Ty8jKDg4PNbt26mQsWLMg/xq+//moOGjTIrFGjhunh4WHWqFHDHDp0qLl9+/bzvk4RqTwM07RwlKuIiIiUuPvuu493332XlJQUTQIgIlIEjQkSEREpx9LT0wvcP378OJ988gkdO3ZUABIROQeNCRIRESnH2rVrR9euXWncuDGHDx/mww8/JDk5+YIL64qIVGYKQSIiIuVYv379+Oabb3jvvfcwDIOWLVvy4Ycf0rlzZ6tLExFxWRoTJCIiIiIilYrGBImIiIiISKWiECQiIiIiIpVKuR4T5HA4OHDgAP7+/qW+IJyIiIiIiLgu0zQ5efIkNWrUwGY7f19PuQ5BBw4cICIiwuoyRERERETERcTHx1OrVq3z7lOuQ5C/vz/gfKEBAQGW1pKdnc28efPo1asX7u7ultYiTmoT16M2cS1qD9ejNnE9ahPXovZwPa7UJsnJyURERORnhPMp1yEo7xK4gIAAlwhBPj4+BAQEWP4DIE5qE9ejNnEtag/XozZxPWoT16L2cD2u2CbFGSajiRFERERERKRSUQgSEREREZFKRSFIREREREQqlXI9JkhEREREXE9ubi7Z2dklftzs7Gzc3NzIyMggNze3xI8vF68s28Rut+Pm5lYiS+MoBImIiIhIiUlJSWHfvn2YplnixzZNk/DwcOLj47VGpIso6zbx8fGhevXqeHh4/KPjKASJiIiISInIzc1l3759+Pj4EBoaWuIfih0OBykpKfj5+V1wMUwpG2XVJqZpkpWVxdGjR4mNjaVBgwb/6HwKQSIiIiJSIrKzszFNk9DQULy9vUv8+A6Hg6ysLLy8vBSCXERZtom3tzfu7u7ExcXln/NS6adHREREREqULlWT0lJSQUshSEREREREKhWFoBKQ6zD5IzaBtccM/ohNINdR8gMBRURERESkZCgE/UNzNh6k4/O/MeyjNUzfYWfYR2vo+PxvzNl40OrSRERERMqlXIfJil3H+eHP/azYdbxc/oG5Tp06vPbaa8Xef9GiRRiGQWJiYqnVJKdpYoR/YM7Gg4z5dB1n/1oeSspgzKfrmDKsJX2aVrekNhEREZHyaM7Gg0z6aTMHkzLyt1UP9OLxATH0igkr8fNdaPzS448/zsSJEy/6uKtXr8bX17fY+7dv356DBw8SGBh40ee6GIsWLaJbt26cOHGCoKCgUj2XK1MIukS5DpNJP20uFIAATMAAJv20mZ4x4dhtGhwoIiIiciEX+gPzWze3oH1tnxI958GDp6/e+fLLL3nsscfYtm1b/jY/P7/8703TJDc3Fze3C3+EDg0Nvag6PDw8CA8Pv6jnyKWz9HK4OnXqYBhGodvYsWOtLKtYVsUmFPgLxdlM4GBSBqtiE8quKBEREREXYpomaVk5xbqdzMjm8R83nfMPzABP/LyZlIziHa+4i7WGh4fn3wIDAzEMI//+1q1b8ff3Z/bs2bRq1QpPT0+WLl3Krl27GDRoEGFhYfj5+dGmTRsWLFhQ4LhnXw5nGAYffPAB1157LT4+PjRo0IAff/wx//GzL4ebNm0aQUFBzJ07l8aNG+Pn50efPn0KhLacnBzuvfdegoKCCAkJYfz48YwYMYJrrrmmWK+9KCdOnGD48OFUqVIFHx8f+vbty44dO/Ifj4uLY8CAAVSpUgVfX1+aNWvGvHnz8p97yy235E+R3qBBA6ZOnXrJtZQmS3uCVq9eTW5ubv79jRs30rNnT2644QYLqyqeIyfPHYAuZT8RERGRiiY9O5eYx+aWyLFM4FByJh1f+6NY+29+ojc+HiXzUffhhx/mpZdeom7dulSpUoX4+Hj69evH008/jaenJ9OnT2fAgAFs27aN2rVrn/M4kyZN4oUXXuDFF1/kzTff5JZbbiEuLo7g4OAi909LS+Oll17ik08+wWazMWzYMB588EE+++wzAJ5//nk+++wzpk6dSuPGjXn99df5/vvv6dat2yW/1pEjR7Jjxw5+/PFHAgICGD9+PP369WPz5s24u7szduxYsrKyWLx4Mb6+vmzcuBG73Q7Ao48+yubNm5k9ezZVq1Zl586dpKenX3ItpcnSEHR2N+Fzzz1HvXr16NKli0UVFV81/+ItzlTc/URERETENT3xxBP07Nkz/35wcDDNmzfPv//kk08yc+ZMfvzxR8aNG3fO44wcOZKhQ4cC8Mwzz/DGG2+watUq+vTpU+T+2dnZvPPOO9SrVw+AcePG8cQTT+Q//uabbzJhwgSuvfZaACZPnsysWbMu+XXmhZ9ly5bRvn17AD777DMiIiL4/vvvueGGG9i7dy+DBw+mWbNmgLPHKzk5GYC9e/fSokULWrdunf+Yq3KZMUFZWVl8+umnPPDAA+ccoJaZmUlmZmb+/bw3PDs7m+zs7DKpM0+LWv6EB3hyODmzyG5bAwgP9KRFLf8yr02c8t53vf+uQ23iWtQerkdt4nrUJhcnOzsb0zRxOBw4HA487QYbJ/a88BNxDjW4/eO1F9zvrRsa0zmm5gUnNPC0GzgcjmKdO0/e/md/bdmyZYFjpaSkMGnSJGbNmsXBgwfJyckhPT2duLi4AvvlvRd5mjZtmn/f29ubgIAADh06lP9+5Z0z7+bj40NUVFT+Y2FhYRw5cgSHw0FSUhKHDx+mdevW+Y8bhpFf67le+9nnOdOmTZtwc3OjTZs2+Y9VqVKFRo0asXnzZhwOB+PGjWPs2LHMmzeP7t27c+211xIVFYVpmowePZobbriBdevW0bNnTwYNGpQfpkqKw+HANE2ys7Pze6DyXMzvqcuEoO+//57ExERGjhx5zn2effZZJk2aVGj7vHnz8PEp2UFyxdEv3OCj5LxhVQV/EU1M+oalMXfO7DKvSwqaP3++1SXIWdQmrkXt4XrUJq5HbVI8bm5uhIeHk5KSQlZW1kU9t3mYJ2H+Hhw5mXXOPzBX8/fgyqgq5GZe+BKrSxmRkJGRgWma+X9oT0tLA5wfvPO2Adx///0sWrSIJ598kqioKLy9vRkxYgQpKSn5+zkcDjIyMgo8Lycnp8D9vHMkJyfnn+vkyZPYbDYyMjJwc3MrsP+Z9eVtT01NLXSOs+s9+3xnnqeox5KTkwsEjNzcXDIzM0lOTmbIkCG0b9+eefPmsXDhQp577jmeeuop7rrrLjp06MDff//N/PnzWbhwIT179mTUqFE8+eSTF3zviysrK4v09HQWL15MTk5OkfUXh8uEoA8//JC+fftSo0aNc+4zYcIEHnjggfz7ycnJRERE0KtXLwICAsqizAL6AS03HeapWVs5lJxZ4DEPNzu3D+pCWIAuh7NKdnY28+fPp2fPnri7u1tdjqA2cTVqD9ejNnE9apOLk5GRQXx8PH5+fnh5XfxnoMcHNGHsjPUYUCAIGfmPx2C3Gfj7+1+wJ+hSeHl5YRhG/ufKvD+y+/v7F/isuWbNGm677TZuvvlmwNkzFB8fj4eHR/5+NpsNLy+vAs/L6/3Jf12Gkb/P2ec6u5a85wMEBAQQEBBAWFgYW7ZsoW/fvoAzrGzYsIHmzZuf87PxuV4TQKtWrcjJyWHLli35PTjHjx9n586dXH755fn7x8TEEBMTw3333ceECRP4+OOP+c9//pNf7+jRoxk9ejTvvvsu48eP5/XXXy9+I1xARkYG3t7edO7cudDP2LmCX1FcIgTFxcWxYMECvvvuu/Pu5+npiaenZ6Ht7u7ulv3DdPXlteh7WU1W7DzCvCV/0LPjFbyyYCfr45N45dddvDLkckvqktOs/PmQoqlNXIvaw/WoTVyP2qR4cnNzMQwDm81WqJehOPpdVoMpNqPQOkHhZ6wTlJycnH+OkpZ3zKK+nnm+Bg0aMHPmTAYOHIhhGDz66KM4HI5CdZ19v6j3JW/b2ec6u4ai6rrnnnt47rnnaNCgAdHR0bz55pucOHHivO9/3vZNmzbh7+9foNbmzZszaNCg/ADj7+/Pww8/TM2aNbn22mux2Wzcd9999O3bl4YNG3LixAl+//13GjVqhGEYTJw4kVatWtGkSRMyMzOZNWsWjRs3LtG2stlsGIZR5O/kxfyOukQImjp1KtWqVaN///5Wl3JJ7DaDtlHBHN9icmXdECYO9GLQW8v4bt1+br0ykha1q1hdooiIiEi50KdpdXrGhLMqNoEjJzOo5u/FFVHB2G0XP8antLzyyivcfvvttG/fnqpVqzJ+/PiL6oUoKePHj+fQoUMMHz4cu93OXXfdRe/evQuNlSlK586dC9y32+3k5OQwdepU/v3vf3P11VeTlZVF586dmTVrVn7AyM3NZezYsezbt4+AgAB69+6dP1zFw8ODCRMmsGfPHry9venUqRNffPFFyb/wEmCYxZ1EvZQ4HA6ioqIYOnQozz333EU9Nzk5mcDAQJKSkiy5HO5M2dnZzJo1i379+uHu7s6DX//FN2v3cXlEEN+NaY9NC6aWubPbRKynNnEtag/XozZxPWqTi5ORkUFsbCxRUVGXdDncheSNdQkICCiVnqDyzuFw0LhxY4YMGVKi43AudM6ybJPz/YxdTDaw/KdnwYIF7N27l9tvv93qUkrUQ70b4eth58/4RH74a7/V5YiIiIhIBRMXF8f777/P9u3b2bBhA2PGjCE2NjZ/rJKcm+UhqFevXpimScOGDa0upURVC/Bi7FX1AXhu9lZSM3Mu8AwRERERkeKz2WxMmzaNNm3a0KFDBzZs2MCCBQto3Lix1aW5PJcYE1RR3d4hii9WxbM3IY0pi3bxYO9GVpckIiIiIhVEREQEy5Yts7qMcsnynqCKzMvdziP9nUn8vSW7iU8o/tzlIiIiIiJSOhSCSlmvmDA61A8hK8fBM7O2WF2OiIiIiEilpxBUygzD4LGrm2AzYPbGQ6zYddzqkkREREREKjWFoDLQKNyfYVdGAjDpp03kOiydlVxEREREpFJTCCoj9/doSKC3O1sPneSL1XutLkdEREREpNJSCCojVXw9uL9HAwBemruNpLRsiysSEREREamcFILK0C1XRtKgmh8n0rJ5/dcdVpcjIiIi4pocuRC7BDZ84/zqyLW6ogvq2rUr9913X/79OnXq8Nprr533OYZh8P333//jc5fUcSoThaAy5G638diAGACmr9jDziMpFlckIiIi4mI2/wivNYWPr4Zv73B+fa2pc3spGDBgAH369CnysSVLlmAYBn///fdFH3f16tXcdddd/7S8AiZOnMjll19eaPvBgwfp27dviZ7rbNOmTSMoKKhUz1GWFILKWKcGofRoHEaOw+SpXzZbXY6IiIiI69j8I3w1HJIPFNyefNC5fctPJX7KO+64g/nz57Nv375Cj02dOpXWrVtz2WWXXfRxQ0ND8fHxKYkSLyg8PBxPT88yOVdFoRBkgUf6N8bdbrBo21EWbj1idTkiIiIipcM0ISu1eLeMZJj9EFDULLrObcachyHzZPGOZxZvNt6rr76a0NBQpk2bVmB7SkoKX3/9NXfccQfHjx9n6NCh1KxZEx8fH5o1a8bnn39+3uOefTncjh076Ny5M15eXsTExDB//vxCzxk/fjwNGzbEx8eHunXr8uijj5Kd7RxHPm3aNCZNmsRff/2FYRgYhpFf89mXw23YsIGrrroKb29vQkJCuOuuu0hJOX0F0siRI7nmmmt46aWXqF69OiEhIYwdOzb/XJdi7969DBo0CD8/PwICAhgyZAiHDx/Of/yvv/6iW7du+Pv7ExAQQKtWrVizZg0AcXFxDBgwgCpVquDr60uTJk2YNWvWJddSHG6lenQpUlRVX27vEMW7i3fz5M+b6VC/Kh5uyqMiIiJSwWSnwTM1SuhgJsbJAwRNaVq83f97ADx8L7ibm5sbw4cPZ9q0aTzyyCMYhgHA119/TW5uLkOHDiUlJYVWrVoxfvx4AgIC+OWXX7j11lupV68eV1xxxQXP4XA4uO666wgLC+OPP/4gKSmpwPihPP7+/kybNo0aNWqwYcMG7rzzTvz9/XnooYe48cYb2bhxI3PmzGHBggUABAYGFjpGamoqvXv3pl27dqxevZojR44watQoxo0bVyDoLVy4kOrVq7Nw4UJ27tzJjTfeyOWXX86dd955wddT1Ou79tpr8fPz4/fffycnJ4exY8dy4403smjRIgBuueUWWrRowZQpU7Db7fz555+4u7sDMHbsWLKysli8eDG+vr5s3rwZPz+/i67jYigEWWTcVfX5dt0+dh9LZfqKPYzqVNfqkkREREQqpdtvv50XX3yR33//na5duwLOS+EGDx5MYGAggYGBPPjgg/n733PPPcydO5evvvqqWCFowYIFbN26lblz51KjhjMUPvPMM4XG8fzvf//L/75OnTo8+OCDfPHFFzz00EN4e3vj5+eHm5sb4eHh5zzXjBkzyMjIYPr06fj6OkPg5MmTGTBgAM8//zxhYWEAVKlShcmTJ2O324mOjqZ///78+uuvlxSCfv/9dzZs2EBsbCwREREATJ8+nSZNmrB69WratGnD3r17+b//+z+io6MBaNCgQf7z9+7dy+DBg2nWrBkAdeuW/udihSCL+Hu581DvaB769m9eX7CDa1rUpKqfruUUERGRCsTdx9kjUxxxy+Gz6y+4W8o1H+MT3R2b7QJX0bgXfzxOdHQ07du356OPPqJr167s3LmTJUuW8MQTTwCQm5vLM888w1dffcX+/fvJysoiMzOz2GN+tmzZQkRERH4AAmjXrl2h/b788kveeOMNdu3aRUpKCjk5OQQEBBT7deSdq3nz5vkBCKBDhw44HA62bduWH4KaNGmC3W7P36d69eps2LDhos6VZ/v27UREROQHIICYmBiCgoLYsmULbdq04YEHHmDUqFF88skn9OjRgxtuuIF69eoBcO+99zJmzBjmzZtHjx49GDx48CWNw7oYugbLQte3qkWzmoGczMzh5XnbrS5HREREpGQZhvOStOLc6l0FATUA41wHwwyoSU7tTsU7nnGu4xTtjjvu4Ntvv+XkyZNMnTqVevXq0aVLFwBefPFFXn/9dcaPH8/ChQv5888/6d27N1lZWf/s/TnDihUruOWWW+jXrx8///wz69ev55FHHinRc5wp71K0PIZh4HA4SuVc4JzZbtOmTfTv35/ffvuNmJgYZs6cCcCoUaPYvXs3t956Kxs2bKB169a8+eabpVYLKARZymYzePzUlNlfrN7LpgNJFlckIiIiYhGbHfo8f+rO2QHGed/s/axzv1IwZMgQbDYbM2bMYPr06dx+++3544OWLVvGoEGDGDZsGM2bN6du3bps3178P2A3btyY+Ph4Dh48mL9t5cqVBfZZvnw5kZGRPPLII7Ru3ZoGDRoQFxdXYB8PDw9yc8+/ZlLjxo3566+/SE1Nzd+2bNkybDYbjRo1KnbNF6Nhw4bEx8cTHx+fv23z5s0kJiYSExNTYL/777+fefPmcd111zF16tT8xyIiIvjXv/7Fd999x3/+8x/ef//9Uqk1j0KQxVrXCWZA8xqYJkz6aTNmMWcyEREREalwYgbCkOkQUL3g9oAazu2NB5Taqf38/LjxxhuZMGECBw8eZOTIkfmPNWjQgPnz57N8+XK2bNnC6NGjC8x8diE9evSgYcOGjBgxgr/++oslS5bwyCOPFNinQYMG7N27ly+++IJdu3bxxhtv5PeU5KlTpw6xsbH8+eefHDt2jMzMzELnuuWWW/Dy8mLEiBFs3LiRhQsXcs8993DrrbfmXwp3qXJzc/nzzz8L3LZs2ULXrl1p1qwZt9xyC+vWrWPVqlUMHz6cLl260Lp1a9LT0xk3bhyLFi0iLi6OZcuWsXr1aho3bgzAfffdx9y5c4mNjWXdunUsXLgw/7HSohDkAh7uG42Xu41VsQnM2nDI6nJERERErBMzEO7bCCN+hsEfOr/et8G5vZTdcccdnDhxgt69excYv/O///2Pli1b0rt3b7p27Up4eDjXXHNNsY9rs9mYOXMm6enpXHHFFYwaNYqnn366wD4DBw7k/vvvZ9y4cVx++eUsX76cRx99tMA+gwcPpk+fPnTr1o3Q0NAip+n28fFh7ty5JCQk0KZNG66//nq6d+/O5MmTL+7NKEJKSgotWrQocBs0aBCGYTBz5kyqVKlC586d6dGjB3Xr1uXLL78EwG63c/z4cYYPH07Dhg0ZMmQIffv2ZdKkSYAzXI0dO5bGjRvTp08fGjZsyNtvv/2P6z0fwyzHXQ/JyckEBgaSlJR00YPGSlp2djazZs2iX79+ha6xLI7XFmzntQU7qBnkza//6YKXe+l09VYm/7RNpOSpTVyL2sP1qE1cj9rk4mRkZBAbG0tUVBReXl4lfnyHw0FycjIBAQEXnhhBykRZt8n5fsYuJhvop8dFjO5cjxqBXuxPTOe9xbutLkdEREREpMJSCHIR3h52JvRzXvs4ZdEuDialW1yRiIiIiEjFpBDkQq6+rDpt6lQhPTuX52dvtbocEREREZEKSSHIhRiGwWNXN8Ew4Ps/D7A2LsHqkkREREREKhyFIBfTrFYgQ1o5V9ud9NNmHI5yO2+FiIiIVFLleN4tcXEl9bOlEOSCHuzdCD9PN/7el8S36/ZZXY6IiIhIsdjtztlts7KyLK5EKqq0tDSAfzxbo1tJFCMlK9Tfk3u71+eZWVt5Ye42+jarjp+nmkpERERcm5ubGz4+Phw9ehR3d/cSnzLZ4XCQlZVFRkaGpsh2EWXVJqZpkpaWxpEjRwgKCsoP3JdKn6xd1Mj2UXy+Kp7YY6m8tXAn4/tEW12SiIiIyHkZhkH16tWJjY0lLi6uxI9vmibp6el4e3tjGEaJH18uXlm3SVBQEOHh4f/4OApBLsrDzcYj/RozavoaPlwSy01tIogM8bW6LBEREZHz8vDwoEGDBqVySVx2djaLFy+mc+fOWrzWRZRlm7i7u//jHqA8CkEurHvjanRqUJUlO47x9C9beG94a6tLEhEREbkgm82Gl5dXiR/XbreTk5ODl5eXQpCLKK9toospXZhzyuwY7DaDeZsPs3THMatLEhEREREp9xSCXFyDMH9uvTISgCd+3kROrsPiikREREREyjeFoHLg/h4NqeLjzvbDKXy+aq/V5YiIiIiIlGsKQeVAoI87D/RqBMDL87eTmKa590VERERELpVCUDkxtE0EjcL8SUzL5rUFO6wuR0RERESk3FIIKifc7DYeGxADwCcr49h++KTFFYmIiIiIlE8KQeVIh/pV6d0kjFyHyZM/b8Y0TatLEhEREREpdxSCyplH+sXgYbexZMcxFmw5YnU5IiIiIiLljkJQOVM7xIdRnaIAePqXzWTm5FpckYiIiIhI+aIQVA7d3a0+of6e7DmexrRle6wuR0RERESkXFEIKof8PN0Y3ycagDd/28mRkxkWVyQiIiIiUn4oBJVT17WoSfNagaRk5vDS3G1WlyMiIiIiUm4oBJVTNpvBYwOaAPD12n1s2JdkcUUiIiIiIuWDQlA51iqyCte2qIlpwqSfNmnKbBERERGRYlAIKufG94nG293OmrgT/PT3QavLERERERFxeQpB5Vx4oBd3d60HwLOztpCepSmzRURERETORyGoArizc11qBnlzMCmDd37fZXU5IiIiIiIuTSGoAvByt/NI/8YAvPP7LvYnpltckYiIiIiI61IIqiD6Ng2nbVQwmTkOnp21xepyRERERERclkJQBWEYBo8NiMFmwM9/H2RVbILVJYmIiIiIuCSFoAqkSY1AbmxTG3BOmZ3r0JTZIiIiIiJnUwiqYB7s1RB/Lzc2HUjmm7XxVpcjIiIiIuJyFIIqmBA/T/7dvQEAL87dRnJGtsUViYiIiIi4FstD0P79+xk2bBghISF4e3vTrFkz1qxZY3VZ5drwdnWoG+rLsZQsJv+20+pyRERERERciqUh6MSJE3To0AF3d3dmz57N5s2befnll6lSpYqVZZV7Hm42Hr06BoCpy2KJPZZqcUUiIiIiIq7DzcqTP//880RERDB16tT8bVFRURZWVHF0a1SNbo1CWbjtKE//spkPRrSxuiQREREREZdgaQj68ccf6d27NzfccAO///47NWvW5O677+bOO+8scv/MzEwyMzPz7ycnJwOQnZ1Ndra1Y1/yzm91HWd6uHdDluw4xoItR/ht80E6NahqdUllyhXbpLJTm7gWtYfrUZu4HrWJa1F7uB5XapOLqcEwTdOyeZS9vLwAeOCBB7jhhhtYvXo1//73v3nnnXcYMWJEof0nTpzIpEmTCm2fMWMGPj4+pV5veTRzj41FB22EeZuMvywXu+WjwERERERESl5aWho333wzSUlJBAQEnHdfS0OQh4cHrVu3Zvny5fnb7r33XlavXs2KFSsK7V9UT1BERATHjh274AstbdnZ2cyfP5+ePXvi7u5uaS1nSk7PpsdrSzmRls3/+jViRLtIq0sqM67aJpWZ2sS1qD1cj9rE9ahNXIvaw/W4UpskJydTtWrVYoUgSy+Hq169OjExMQW2NW7cmG+//bbI/T09PfH09Cy03d3d3fI3PY8r1QIQ4u7O//WO5r8zN/DGb7u4rlVtgn09rC6rTLlam4jaxNWoPVyP2sT1qE1ci9rD9bhCm1zM+S29OKpDhw5s27atwLbt27cTGVl5eivKwo1tImhcPYDkjBxenb/d6nJERERERCxlaQi6//77WblyJc888ww7d+5kxowZvPfee4wdO9bKsiocu83g8QHOHrfP/ohj66FkiysSEREREbGOpSGoTZs2zJw5k88//5ymTZvy5JNP8tprr3HLLbdYWVaFdGXdEPo1C8dhwqQfN2PhUDAREREREUtZOiYI4Oqrr+bqq6+2uoxKYULfxizYcoQVu48zd9Nh+jQNt7okEREREZEypwmTK5GIYB9Gd64LwNOzNpORnWtxRSIiIiIiZU8hqJIZ07Ue4QFexCek8+HSWKvLEREREREpcwpBlYyPhxsP940G4K2FOzmcnGFxRSIiIiIiZUshqBIadHkNWtQOIi0rlxfmbLvwE0REREREKhCFoErIMAweH9AEgG/X7ePP+ERrCxIRERERKUMKQZXU5RFBDG5ZC4CJP27C4dCU2SIiIiJSOSgEVWLj+zTC18POn/GJ/PDXfqvLEREREREpEwpBlVi1AC/GXlUfgOdmbyU1M8fiikRERERESp9CUCV3e4coagf7cDg5k3d+32V1OSIiIiIipU4hqJLzcrfz336NAXh38W7iE9IsrkhEREREpHQpBAm9m4TRvl4IWTkOnp29xepyRERERERKlUKQYBgGjw2IwWbArA2HWLHruNUliYiIiIiUGoUgASA6PIBb2kYC8MTPm8nVlNkiIiIiUkEpBEm+B3o2JNDbnS0Hk/lydbzV5YiIiIiIlAqFIMlXxdeD+3o0AOCledtISs+2uCIRERERkZKnECQFDLsykvrV/EhIzeKNX3dYXY6IiIiISIlTCJIC3O02Hr06BoCPl+9h55EUiysSERERESlZCkFSSJeGofRoXI0ch8lTv2y2uhwRERERkRKlECRFeqR/DO52g0XbjrJw6xGryxERERERKTEKQVKkqKq+3N4hCoAnf9lMVo7D4opEREREREqGQpCc07ir6lPVz4PdR1OZvmKP1eWIiIiIiJQIhSA5J38vd/6vdyMAXv91B8dSMi2uSERERETkn1MIkvO6vlUETWsGcDIjh5fnbbe6HBERERGRf0whSM7LbjN4fEATAL5YvZdNB5IsrkhERERE5J9RCJILalMnmAHNa2Ca8MRPmzFN0+qSREREREQumUKQFMvDfaPxcrfxR2wCszcesrocEREREZFLphAkxVIzyJvRnesB8PQvW8jIzrW4IhERERGRS6MQJMX2ry71qB7oxf7EdN5fvNvqckRERERELolCkBSbt4edCf0aA/D2ol0cSsqwuCIRERERkYunECQXZcBl1WkdWYX07Fyen7PV6nJERERERC6aQpBcFMNwTpltGDBz/X7Wxp2wuiQRERERkYuiECQXrVmtQG5oVQuAJ37ahMOhKbNFREREpPxQCJJL8mDvRvh5uvHXviS+W7/f6nJERERERIpNIUguSTV/L+65qj4Az8/ZSkpmjsUViYiIiIgUj0KQXLKRHepQJ8SHoyczeWvhTqvLEREREREpFoUguWSebnb+1z8GgA+XxLL3eJrFFYmIiIiIXJhCkPwj3RtXo1ODqmTlOnh61maryxERERERuSCFIPlHDMPgsatjsNsM5m46zLKdx6wuSURERETkvBSC5B9rEObPrVdGAvDET5vJyXVYXJGIiIiIyLkpBEmJuK9HA4J83Nl2+CSfr9prdTkiIiIiIuekECQlIsjHg//0bAjAy/O3k5iWZXFFIiIiIiJFUwiSEjP0ito0CvMnMS2b1xbssLocEREREZEiKQRJiXGz23hsgHPK7E9WxrHj8EmLKxIRERERKUwhSEpUh/pV6RUTRq7D5ImfN2OaptUliYiIiIgUoBAkJe6R/o3xsNtYsuMYv245YnU5IiIiIiIFKARJiYsM8eWOTlEAPPXLZjJzci2uSERERETkNIUgKRVju9Un1N+TPcfT+Hj5HqvLERERERHJpxAkpcLP043xfaIBeOPXnRw9mWlxRSIiIiIiTgpBUmqua1GT5rUCScnM4aW526wuR0REREQEUAiSUmSzGTw2oAkAX62NZ8O+JIsrEhERERFRCJJS1iqyCtdcXgPThEk/bdKU2SIiIiJiOYUgKXXj+0bj7W5nTdwJfvr7oNXliIiIiEglZ2kImjhxIoZhFLhFR0dbWZKUguqB3tzdtR4Az83aQnqWpswWEREREetY3hPUpEkTDh48mH9bunSp1SVJKbizc11qBnlzICmDdxfvsrocEREREanELA9Bbm5uhIeH59+qVq1qdUlSCrzc7fy3X2MA3vl9F/sT0y2uSEREREQqKzerC9ixYwc1atTAy8uLdu3a8eyzz1K7du0i983MzCQz8/R6M8nJyQBkZ2eTnZ1dJvWeS975ra7DlfWMDqFNnSqs3nOCZ37ZzGtDLivV86lNXI/axLWoPVyP2sT1qE1ci9rD9bhSm1xMDYZp4XRds2fPJiUlhUaNGnHw4EEmTZrE/v372bhxI/7+/oX2nzhxIpMmTSq0fcaMGfj4+JRFyfIP7UuFl/62Y2Jwb5Mc6gVYXZGIiIiIVARpaWncfPPNJCUlERBw/g+ZloagsyUmJhIZGckrr7zCHXfcUejxonqCIiIiOHbs2AVfaGnLzs5m/vz59OzZE3d3d0trcXX/+2EzX67ZR5Ma/nw3+kpsNqNUzqM2cT1qE9ei9nA9ahPXozZxLWoP1+NKbZKcnEzVqlWLFYIsvxzuTEFBQTRs2JCdO3cW+binpyeenp6Ftru7u1v+pudxpVpc1UN9opm18RCbDpzkh78PM6RNRKmeT23ietQmrkXt4XrUJq5HbeJa1B6uxxXa5GLOb/nECGdKSUlh165dVK9e3epSpBSF+Hny7+4NAHhh7lZOZlh/DamIiIiIVB6WhqAHH3yQ33//nT179rB8+XKuvfZa7HY7Q4cOtbIsKQPD29WhblVfjqVkMfm3onv+RERERERKg6UhaN++fQwdOpRGjRoxZMgQQkJCWLlyJaGhoVaWJWXAw83Go1fHAPDRslhij6VaXJGIiIiIVBaWjgn64osvrDy9WKxbdDW6Ngpl0bajPP3LZj4Y0cbqkkRERESkEnCpMUFS+fyvfwxuNoMFW46wePtRq8sRERERkUpAIUgsVb+aHyPa1wHgyZ83k53rsLYgEREREanwFILEcvd2b0Cwrwc7jqTw2co4q8sRERERkQpOIUgsF+jtzn96NQTglfnbSUjNsrgiEREREanIFILEJdzUpjbR4f4kZ+Tw6vztVpcjIiIiIhWYQpC4BLvN4PEBTQD47I84th5KtrgiEREREamoFILEZbSrF0K/ZuE4THjip82Ypml1SSIiIiJSASkElQRHLkbcUmomrMCIWwqOXKsrKrcm9G2Mh5uN5buOM2/zYavLEREREZEKSCHon9r8I7zWFLdPr6F13BTcPr0GXmvq3C4XLSLYh7s61QXg6V+2kJGtQCkiIiIiJUsh6J/Y/CN8NRySDxTcnnzQuV1B6JKM6VqPsABP9iak8dGyWKvLEREREZEKRiHoUjlyYc54oKhxK6e2zXlYl8ZdAl9PNx7uGw3A5N92ciQ5w+KKRERERKQiUQi6VHHLC/cAFWBC8n7nfnLRBjWvSYvaQaRl5fLC3G1WlyMiIiIiFYhC0KVKKeag/eLuJwXYzpgy+5u1+/grPtHagkRERESkwlAIulR+YSW7nxRyeUQQ17WsCcDEnzZpymwRERERKREKQZcqsj0E1ACMc+/j5gXVLy+riiqk8X2i8fGws35vIj/8eb7LD0VEREREikch6FLZ7NDn+VN3zhGEcjJg+kA4qUviLlVYgBdju9UH4NnZW0jNzLG4IhEREREp7xSC/omYgTBkOgRUL7g9oCZ0+x94B8OBdfBBDziyxZoaK4A7OkYREezN4eRM3vl9l9XliIiIiEg5pxD0T8UMhPs2kjPse9ZEjiFn2Pdw3wbo8n8wagEE14OkvfBhb9i10OpqyyUvdzuP9IsB4L3Fu4lPSLO4IhEREREpzxSCSoLNjhnZkf3B7TAjOzovlQMIqecMQrXbQWYSfHY9rJtuba3lVO8mYbSvF0JmjoPnZm+1uhwRERERKccUgkqbTzAM/wGaDQFHDvx4DyyYCA6H1ZWVK4Zh8NiAGGwG/LLhICt3H7e6JBEREREppxSCyoKbJ1z3HnR52Hl/6avwzW2QnW5tXeVMdHgAN7etDcCknzaT69CU2SIiIiJy8RSCyophQLcJcM07YHOHzd/DxwMh9ZjVlZUrD/RsRICXG1sOJvPl6niryxERERGRckghqKxdPhRunQlegbBvFXzQHY5ut7qqciPY14P7ezYE4OV520hKz7a4IhEREREpbxSCrBDVCe5YAFXqwIk98GEPiF1idVXlxrArI6lfzY/jqVm8+esOq8sRERERkXJGIcgqoQ1h1K9Q6wrISIJProU/Z1hdVbngbrfx6NXOKbOnLd/DrqMpFlckIiIiIuWJQpCVfKvCiJ+gyXXgyIbvx8BvT4OpAf8X0qVhKN2jq5HjMHnq581WlyMiIiIi5YhCkNXcvWDwh9DpP877i1+A7+6EnExr6yoHHunfGHe7wcJtR1m49YjV5YiIiIhIOaEQ5ApsNuj+GAycDDY32PA1TB8EqVoL53zqhvpxW4coAJ78ZTPZuVp7SUREREQu7JJCUHx8PPv27cu/v2rVKu677z7ee++9EiusUmp5Kwz7FjwDYe8K54QJx3dZXZVLG3dVfar6ebD7aCrTV8RZXY6IiIiIlAOXFIJuvvlmFi5cCMChQ4fo2bMnq1at4pFHHuGJJ54o0QIrnbpd4Y55EFQbEnY7p9COW251VS4rwMud/+vdCIDXFmzneIouIxQRERGR87ukELRx40auuOIKAL766iuaNm3K8uXL+eyzz5g2bVpJ1lc5VYt2zhxXsxWkn3BeGvf3V1ZX5bKubxVBkxoBnMzI4eX5WnNJRERERM7vkkJQdnY2np6eACxYsICBAwcCEB0dzcGDB0uuusrMrxqM+BkaD4DcLOdkCb+/oJnjimC3GTw+oAkAn6/ay6YDSRZXJCIiIiKu7JJCUJMmTXjnnXdYsmQJ8+fPp0+fPgAcOHCAkJCQEi2wUvPwgRumQ/t7nfcXPu2cRjsny9q6XNAVUcFcfVl1TBOe+GkzpsKiiIiIiJzDJYWg559/nnfffZeuXbsydOhQmjdvDsCPP/6Yf5mclBCbDXo9CVe/CoYd/vocPr0O0hKsrszlTOjXGE83G3/EJjBn4yGryxERERERF+V2KU/q2rUrx44dIzk5mSpVquRvv+uuu/Dx8Smx4uQMrW93Tpbw1UjYswQ+7AW3fAXBda2uzGXUDPLmX13q8fqvO3jql8142mNYe8wgJDaBdvWrYbcZVpcoIiIiIi7gknqC0tPTyczMzA9AcXFxvPbaa2zbto1q1aqVaIFyhvo94I65EFALju+AD3rA3j+srsql/KtLPYJ83NmfmMHt09cxfYedYR+toePzvzFno8ariYiIiMglhqBBgwYxffp0ABITE2nbti0vv/wy11xzDVOmTCnRAuUsYU3gzl+h+uWQdhw+HgAbv7W6Kpfx+/YjJKZlF9p+KCmDMZ+uUxASERERkUsLQevWraNTp04AfPPNN4SFhREXF8f06dN54403SrRAKYJ/ONw2Cxr1h9xM+OZ2WPJypZ85LtdhMumnzUU+lvfOTPppM7mOyv0+iYiIiFR2lxSC0tLS8Pf3B2DevHlcd9112Gw2rrzySuLi4kq0QDkHD1+48RO4cqzz/q9PwI/3QG7hXpDKYlVsAgeTMs75uAkcTMpgVawmlRARERGpzC4pBNWvX5/vv/+e+Ph45s6dS69evQA4cuQIAQEBJVqgnIfNDn2egX4vgWGD9Z/Ap4MhPdHqyixx5OS5A9Cl7CciIiIiFdMlhaDHHnuMBx98kDp16nDFFVfQrl07wNkr1KJFixItUIrhijth6Bfg7guxv8NHveFE5euRq+bvVaL7iYiIiEjFdEkh6Prrr2fv3r2sWbOGuXPn5m/v3r07r776aokVJxehYW+4fQ74V4ejW+GD7rBvrdVVlakrooKpHujFhSbCPpKsniARERGRyuySQhBAeHg4LVq04MCBA+zbtw+AK664gujo6BIrTi5S9ctg1K8Q3gxSj8K0frD5B6urKjN2m8HjA2IACgWhM+//+8s/eXHuVhyaIEFERESkUrqkEORwOHjiiScIDAwkMjKSyMhIgoKCePLJJ3E4HCVdo1yMwJpw22xo0AtyMuCrEbDsjUozc1yfptWZMqwl4YEFL3kLD/Ti7ZtbMrqLc3HZtxbu4q5P1nAyo/JOJCEiIiJSWbldypMeeeQRPvzwQ5577jk6dOgAwNKlS5k4cSIZGRk8/fTTJVqkXCRPf7jpc5jzMKx+H+Y/Cgm7nRMo2C+pycuVPk2r0zMmnBU7jzBvyR/06tSWdvWrYbcZ9LusOtHh/oz/dgMLthzhureX88GI1kSG+FpdtoiIiIiUkUvqCfr444/54IMPGDNmDJdddhmXXXYZd999N++//z7Tpk0r4RLlktjdoN+L0Oc5wIC1U2HGEMhItrqyMmG3GbSNCqZVVZO2UcHYbacviLu2RS2+Ht2OsABPdhxJYeDkZSzbeczCakVERESkLF1SCEpISChy7E90dDQJCVqDxWUYBlw5Bm76DNx9YNev8FEfSIy3ujLLNY8I4sdxHWkeEURSejbDP1rF1GWxmJXkskERERGRyuySQlDz5s2ZPHlyoe2TJ0/msssu+8dFSQmL7g+3zQK/MDiyyTlz3IH1VldlubAAL76860qua1mTXIfJpJ828/C3G8jMybW6NBEREREpRZc0QOSFF16gf//+LFiwIH+NoBUrVhAfH8+sWbNKtEApITVaOGeOm3GjMwhN7QeDP4ToflZXZikvdzsv39CcmOoBPDNrC1+uiWfX0RSmDGtFqL+n1eWJiIiISCm4pJ6gLl26sH37dq699loSExNJTEzkuuuuY9OmTXzyySclXaOUlKAI51pC9a6C7DT44mZYOaXSzBx3LoZhMKpTXT4a2QZ/LzfWxJ1g0OSlbNyfZHVpIiIiIlIKLnmdoBo1avD000/z7bff8u233/LUU09x4sQJPvzww5KsT0qaVwDc/BW0ug0wnTPIzX4IcnOsrsxyXRtV4/uxHagb6suBpAyuf2c5P/99wOqyRERERKSEXXIIknLM7g5Xvwo9nwQMWPWes1coM8XqyixXL9SPmXd3oEvDUDKyHYybsZ6X5m7TwqoiIiIiFYjLhKDnnnsOwzC47777rC6lcjAM6HAvDPkY3Lxgx1yY2geS1fMR6O3ORyPbcFdn58KqkxfuZPSna0nJVG+ZiIiISEXgEiFo9erVvPvuu5pZzgoxg2DkL+AbCoc2wPvd4eDfVldlObvN4L/9GvPKkOZ4uNmYv/kw1729jLjjqVaXJiIiIiL/0EXNDnfddded9/HExMSLLiAlJYVbbrmF999/n6eeeuqiny8loFbrUzPHDYGjW51rCd0wFRr2troyy13XshZ1Q/24a/oath9OYdBby3j75pa0r1/V6tJERERE5BJdVAgKDAy84OPDhw+/qALGjh1L//796dGjxwVDUGZmJpmZmfn3k5OTAcjOziY7O/uizlvS8s5vdR2XzK8GDP8F+7e3YduzGPPzm3D0ehZH6zusruySlVSbNAn35bt/teXuGX/y9/5kbv1oFY/0bcSwthEYhlESpVYa5f73pIJRe7getYnrUZu4FrWH63GlNrmYGgzTtG5+5C+++IKnn36a1atX4+XlRdeuXbn88st57bXXitx/4sSJTJo0qdD2GTNm4OPjU8rVVg6GmUPzvdOITFgMwK7Q3mysORQMl7hy0lJZufDlbhtrjjnfi3bVHFwf5cBNb42IiIiI5dLS0rj55ptJSkoiICDgvPtaFoLi4+Np3bo18+fPzx8LdKEQVFRPUEREBMeOHbvgCy1t2dnZzJ8/n549e+Lu7m5pLf+YaWJb/jr2Rc6eOUfDvuQOegc8fK2t6yKVRpuYpsmHy+J4Yd52TBNaRwYx+abmhPhpYdXiqFC/JxWA2sP1qE1cj9rEtag9XI8rtUlycjJVq1YtVgi6qMvhStLatWs5cuQILVu2zN+Wm5vL4sWLmTx5MpmZmdjt9gLP8fT0xNOz8IdNd3d3y9/0PK5Uyz/S9f+gal2YOQbb9tnYPh0EN38J/uFWV3bRSrpNxnRrQHT1QO79fD1r4hIZ/O4q3hveiiY1zn+5qJxWYX5PKgi1h+tRm7getYlrUXu4Hldok4s5v2UX8nTv3p0NGzbw559/5t9at27NLbfcwp9//lkoAIkFmg6GET+BTwgc/NM5c9zhTVZX5RK6RVdj5tgO1K3qy/7EdAZPWc4vfx+0uiwRERERKQbLQpC/vz9NmzYtcPP19SUkJISmTZtaVZacrXZbGLUAQhpA8j74sDfsXGB1VS6hfjU/Zo7tQOdTC6uOnbGOV+ZpYVURERERV6ch3XJhwXXhjnkQ2RGyTsJnQ2DNVKurcgmB3u5MHdmGOztFAfDGbzv5lxZWFREREXFpLhWCFi1adM5JEcRiPsFw60xoPhTMXPj5Ppj3KDgcVldmObvN4JH+Mbx8Q3M87DbmbT7M4LeXs/d4mtWliYiIiEgRXCoEiYtz84BrpkC3R5z3l78BX4+ALH3YBxjcqhZfjL6SUH9Pth0+ycC3lrJ81zGryxIRERGRsygEycUxDOjyEFz3Ptg9YMuP8PHVkHLE6spcQsvaVfhpXEcuqxVIYlo2t364ik9W7MHC5bhERERE5CwKQXJpLhsCw38A7yqwfy180B2ObLW6KpcQHujFV6Pbcc3lNch1mDz6wyb+O3MjWTm6dFBERETEFSgEyaWLbA93LHBOnJC4Fz7sBbsXWV2VS/Byt/PqjZfzcN9oDAM+X7WXYR/8wbGUzAs/WURERERKlUKQ/DNV6zuDUO12kJkEnw6GdZ9YXZVLMAyDf3Wpx0cj2uDv6caqPQkMmryMTQeSrC5NREREpFJTCJJ/zjfEeWlcsxvAkQM/joMFkzRz3CnOhVXbUyfEh/2J6Vw/ZQWzNmhhVRERERGrKARJyXDzdE6W0GW88/7SV+DbOyA7w9q6XET9av78MLYjnRpUJT07l7s/W8cr87drYVURERERCygESckxDOj2X+c02jZ32PQdTB8IqZomGiDQx7mw6qiOpxZW/XUHYz5bS6oWVhUREREpUwpBUvIuvxlu/Q68AiH+D+fMccd2WF2VS3Cz2/jf1TG8eP1leNhtzN10mMFTlhOfoLWWRERERMqKQpCUjqjOzgkTgiLhxB74oAfELrG6KpdxQ+sIPr/LubDq1kMnGTh5KSt2Hbe6LBEREZFKQSFISk9oQ7jzN6h1BWQkwifXwp+fW12Vy2gVWYUfx3WgWc1ATqRlc+uHf/DJyjiryxIRERGp8BSCpHT5VoURP0LMNeDIhu//BQufAVMTAgBUD/Tm63+1Y2DzGuQ4TB79fiOPzNyghVVFRERESpFCkJQ+d2+4fip0fMB5//fn4bu7IEcLh4JzYdXXb7qc8X2cC6t+9sdehn34B8e1sKqIiIhIqVAIkrJhs0GPx2Hgm2Bzgw1fwfRrIC3B6spcgmEYjOlajw+Gt8bP041VsQkMnLyMzQeSrS5NREREpMJRCJKy1XI43PINeAbA3uXOCROO77K6KpfRvXEYM+8+vbDq4CnLma2FVUVERERKlEKQlL163eCOeRBYGxJ2OYNQ3Aqrq3IZDcL8+X5sBzrWdy6sOuazdbyqhVVFRERESoxCkFijWmMYtQBqtIT0BOeiqhu+sboqlxHk48G029pwewfnwqqv/7qDuz9bp4VVRUREREqAQpBYxz8MRv4C0VdDbhZ8ewf8/qJmjjvFzW7jsQExvHBqYdU5mw5pYVURERGREqAQJNby8IEhn0D7e5z3Fz4F398NOVnW1uVChrSO4PO72lLVz7mw6qC3lrFytxZWFREREblUCkFiPZsNej0F/V8Bww5/zYBPr4P0E1ZX5jJaRQbnL6yakJrFsA/+4FMtrCoiIiJySRSCxHW0uQNu/go8/GHPEviwFyTEWl2Vy6gR5M1Xo9sx4NTCqv/7fiP/+34D2blaWFVERETkYigEiWtp0ANunwMBNeHYdufMcfGrrK7KZXh72Hnjpsv5v96NMAz4dOVebv3wDxJSdfmgiIiISHEpBInrCW8Ko36F6s0h7RhMuxo2fmd1VS7DMAzGdqvP+7c6F1ZduTuBgZOXsuWgFlYVERERKQ6FIHFNAdXhttnQqB/kZsI3t8GSVzRz3Bl6xDgXVo0M8WHfCefCqnM2amFVERERkQtRCBLX5eELN34Kbcc47/86CX66F3Kzra3LhTQI8+eHsR3oUD+EtKxc/vXpOl5fsEMLq4qIiIich0KQuDabHfo+B31fAMMG66bDZ9dDRpLVlbmMIB8PPr7tCka2rwPAqwu2M3bGOtKytLCqiIiISFEUgqR8aDsabvoc3H1h9yLnzHEnNEV0Hje7jYkDm/D84Ga42w1mbzzE4Ckr2HdCC6uKiIiInE0hSMqPRn3g9tngXx2ObnXOHLd/rdVVuZQb29Tm8zuvpKqfB1sOJjNw8jJWxSZYXZaIiIiIS1EIkvKlenPnzHFhzSD1CEztD1t+sroql9K6TjA/jOtIkxoBJKRmcfP7K5nxx16ryxIRERFxGQpBUv4E1nT2CDXoBTnp8OWtsPxNzRx3hppB3nzzr/b0v6w6OQ6T/87cwKPfb9TCqiIiIiIoBEl55envHCPUZhRgwrz/wS8PQK4mA8jj7WFn8tAW/F/vRgB8sjJOC6uKiIiIoBAk5ZndDfq9BL2fBQxY8xF8fiNkJIMjFyNuKTUTVmDELQVHrtXVWiJ/YdXhrfH1sLNydwKD3lrK1kNaWFVEREQqL4UgKd8MA9rdDTd9Bu4+sHMBvN0eXmmM26fX0DpuCm6fXgOvNYXNP1pdrWV6xoQxc2wHagf7EJ+QznVvL2fupkNWlyUiIiJiCYUgqRii+8PIX8ArEJLjIeVwwceTD8JXwyt1EGp4amHV9vWcC6uO/mQtb/66A1NjqURERKSSUQiSiqN6c3DzOseDpz7oz3m40l4aB1DF14OPbz+9sOrL87czbsZ6LawqIiIilYpCkFQcccsL9wAVYELyfud+lZj7qYVVn7vOubDqLxsOcr0WVhUREZFKRCFIKo7zBqBL2K+Cu+mK2sy480pCfD3YfDCZQVpYVURERCoJhSCpOPzCirff/jWaSvuUNnWC+fGejsRUD+B4aha3fLCSz1dpYVURERGp2BSCpOKIbA8BNQDj/PutnALvdISdv5ZJWa6uZpA334xpR/9m1cnONZnw3QYe/0ELq4qIiEjFpRAkFYfNDn2eP3Xn7CBkOG8tR4B3MBzdAp9eB58NgaPby7hQ1+Pj4cbkm1vwYK+GAHy8Io7hH67ihBZWFRERkQpIIUgqlpiBMGQ6BFQvuD2ghnP7wDfg3nVw5ViwucGOuTClHcweD2mVezyMYRiMu6oB797aCl8POyt2H2fgW0vZduik1aWJiIiIlCiFIKl4YgbCfRvJGfY9ayLHkDPse7hvg3M7gHcV6PMM3P0HNOoHjhz44x14owX88S7kZltavtV6Nwnnu7s7EBHsfWph1WXM08KqIiIiUoEoBEnFZLNjRnZkf3A7zMiOzkvlzla1Pgz9HIb/ANWaQEYizH4IprSH7fOgEi8i2ijcnx/HdqRd3RBSs3K565O1TP5NC6uKiIhIxaAQJFK3K/xrCVz9GvhUhWPbYcYN8OlgOLLF6uosU8XXg+l3XMHwdpEAvDRvO+M+X096VuVdbFZEREQqBoUgEXD2FLW+zTleqMO/we4Bu36FKR3gl/9A6nGrK7SEu93GE4Oa8sy1zXCzGfzy90Guf2c5+xPTrS5NRERE5JIpBImcySsQej4BY/+AxgPAzIXVHzjHCy2fDDmVc7a0m9s6F1YN9vVg04FkBk1eyuo9lXsiCRERESm/FIJEihJcF278FEb+AuGXQWYSzHsE3m4LW3+plOOFrogK5sdxHWhcPYBjKVnc/P5KvtDCqiIiIlIOKQSJnE+djnDXIhg4GfzCIGE3fHEzTB8IhzZaXV2Zq1XFh2/HtKNfs3Cyc00e/m4DE3/cRI4WVhUREZFyRCFI5EJsdmh5K9yzFjr9B+yeELsY3u0EP94LKUesrrBM+Xi48dbNLXmgp3Nh1WnL9zBiqhZWFRERkfJDIUikuDz9oftjMG41NLkOTAes+xjeaAlLX4OcTKsrLDOGYXBv9wa8M6wVPh52lu08zqC3lrH9sBZWFREREdenECRysapEwg1T4fa5UKMlZJ2EBY/D5Daw+YdKNV6oT9Nwvru7PbWqeLM3IY1r31rGgs2HrS5LRERE5LwUgkQuVe0rYdSvcO274F8dEuPgq+EwrT8c+NPq6spMdHgAP47ryJV1g0nNyuXOT9bw1sKdWlhVREREXJalIWjKlClcdtllBAQEEBAQQLt27Zg9e7aVJYlcHJsNmt/kHC/U5WFw84a4ZfBeV/h+LJw8ZHWFZSLY14NP7mjLrVdGYprw4txt3KOFVUVERMRFWRqCatWqxXPPPcfatWtZs2YNV111FYMGDWLTpk1WliVy8Tx8odsEuGcNNBsCmPDnp87xQotfhOyKv7iou93Gk9c05elrm+JmM/j574Pc8O5yDpxaWDXXYfJHbAJrjxn8EZtArkM9RSIiImINNytPPmDAgAL3n376aaZMmcLKlStp0qSJRVWJ/AOBtWDw+9B2NMx5GPatht+egrUfQ4+J0HQwGIbVVZaqW9pGUj/UjzGfrWPj/mQGTl7GbR0i+XTlXg4mZQB2pu9YQ/VALx4fEEOfptWtLllEREQqGUtD0Jlyc3P5+uuvSU1NpV27dkXuk5mZSWbm6Rm4kpOTAcjOziY7O7tM6jyXvPNbXYecZmmbhDWH4bMwNs/E/tskjKR4+PYOHH+8i6PHU5g1W5Z9TWWoZUQA345uy5jP1rP1cAovzt1eaJ9DSRmM+XQdb97UnN5NwiyoUvTvlutRm7getYlrUXu4Hldqk4upwTAtHr28YcMG2rVrR0ZGBn5+fsyYMYN+/foVue/EiROZNGlSoe0zZszAx8entEsVuSQ2Rxb1j8ymweGfcXM4Q3x8lfZsrjGEDI9gi6srXek58OgaO9nmuXq/TII84PGWudgqdgeZiIiIlLK0tDRuvvlmkpKSCAgIOO++loegrKws9u7dS1JSEt988w0ffPABv//+OzExMYX2LaonKCIigmPHjl3whZa27Oxs5s+fT8+ePXF3d7e0FnFyuTY5eRD7omew/f05AKabN45243BcOc45pqgC+iM2gWEfrbngfp/e3pq2URU7ELoil/sdEbWJC1KbuBa1h+txpTZJTk6matWqxQpBll8O5+HhQf369QFo1aoVq1ev5vXXX+fdd98ttK+npyeenp6Ftru7u1v+pudxpVrEyWXaJLg2XPcOXDka5vwXY+9y7EtexP7nZ9DjceeECraKNWv98bScYu/nEm1USbnM74jkU5u4HrWJa1F7uB5XaJOLOb/LfeJyOBwFentEKpwaLeC2WXDDxxBUG04egJmj4YPusPcPq6srUdX8vYq131er4/krPrF0ixERERE5xdIQNGHCBBYvXsyePXvYsGEDEyZMYNGiRdxyyy1WliVS+gwDmlwDY1c7Z43z8IcD6+CjXvD1bZC41+oKS8QVUcFUD/TiQsN9lu06zqC3lnH9lOXM3nBQ02eLiIhIqbI0BB05coThw4fTqFEjunfvzurVq5k7dy49e/a0siyRsuPuBR3vh3vXQcsRgAGbvoM3W8OvT0DmSasr/EfsNoPHBzjH950dhIxTtwl9o7m2RU3cbAZr4k4w5rN1dHlxIR8s2U1yhvUzzYiIiEjFY+mYoA8//NDK04u4Dr9qMPANaDMK5v4X9iyBJS/D+k+h+2PQ/OZyO16oT9PqTBnWkkk/bT61TpBT+FnrBD3cN5rpK/bw2R972Xcinad+2cJrC3ZwQ+ta3NY+itohmgFSRERESoblEyOIyBmqXwYjfoJts2DuI3AiFn4YC3+8C32ehTodra7wkvRpWp2eMeGs2HmEeUv+oFentrSrXw37GfNihwV48X+9oxnXrQEz1+/no2Wx7DySwtRle5i2fA89G4dxR8corogKxqjgC86KiIhI6Sqff1oWqcgMA6L7w9g/oNdT4BkIh/6Gaf3hy1shIdbqCi+J3WbQNiqYVlVN2kYFFwhAZ/L2sHNz29rMu68z025rQ+eGoZgmzNt8mBvfW8mAyUuZuX4fWTmOMn4FIiIiUlEoBIm4KjdPaH+Pc7xQ6zvAsMGWH+GtK2D+Y5CRbHWFpcpmM+jaqBrTb7+C+fd3ZugVEXi62di4P5n7v/yLjs//xuTfdpCQmmV1qSIiIlLOKASJuDrfqnD1K/CvZVDvKsjNgmWvwxstYM1UcORaXWGpaxDmz7PXXcaKCd15sFdDqvl7cuRkJi/N2067Z39lwncb2HG4fE8iISIiImVHIUikvAiLgWHfwc1fQ0gDSDsGP98H73aG3Yusrq5MBPt6MO6qBiwdfxWvDGlOkxoBZOY4+HzVXnq+upjhH63i9+1HMU1NsS0iIiLnphAkUp4YBjTsBXevgL4vgFcQHN4I0wfB50Ph+C6rKywTHm42rmtZi5/v6ciXd11Jr5gwDAMWbz/KiI9W0evVxXy+ai8Z2RW/l0xEREQunkKQSHlkd4e2o+He9dD2X2Bzc84o91ZbmPNfSD9hdYVlwjAM2tYN4b3hrVn0YFdu61AHXw87O46kMOG7DbR79ldemruNI8kZFz6YiIiIVBoKQSLlmU8w9H0exqyABr3BkQ0r34I3WsKq9yE3x+oKy0xkiC+PD2jCiv9253/9G1MzyJsTadlMXriTDs//xv1f/snG/UlWlykiIiIuQCFIpCIIbQi3fOUcMxTaGNITYNaD8E4H2LnA6urKVICXO6M61eX3/+vK27e0pHVkFbJzTWau38/Vby5lyLsrmLvpELkOjRsSERGprLRYqkhFUr87RHWBddPgt6fh6Fb4dDDU7wm9n4bQRlZXWGbc7Db6NatOv2bV+TM+kY+WxjJrw0FWxSawKjaB2sE+jGxfhyFtIvDz1D+FIiIilYl6gkQqGrsbtBnlHC/UbhzY3GHnfHi7Hcx6CNISrK6wzF0eEcQbQ1uwZHw3xnStR6C3O3sT0nji5820e+ZXnvp5M/EJaVaXKSIiImVEIUikovIOcvb+jP0Doq8GMxdWvetcX2jlFMjNtrrCMlc90JvxfaJZMeEqnrymKXWr+nIyM4cPlsbS5cWF3P3ZWtbsSdAU2yIiIhWcQpBIRRdSD276DIb/CGFNISMR5jzs7BnaNgcq4Qd+Hw83br0ykgUPdGHqyDZ0rF8VhwmzNhzi+ndWcM1by/jhz/1k5zqsLlVERERKgUKQSGVRtwuMXgwD3gDfUDi+Az6/ET65Fg5vtro6S9hsBt2iq/HpqLbMua8TQ1rXwsPNxl/7kvj3F3/S6fmFvL1oJ4lpWVaXKiIiIiVIIUikMrHZodUIuGcddLgP7B6we6FzFrmf74fUY1ZXaJno8ABeuL45yx++ivt7NKSqnyeHkjN4Yc422j37G//7fgO7jqZYXaaIiIiUAIUgkcrIKwB6ToKxqyBmEJgOWPORc7zQsjcgJ9PqCi1T1c+Tf/dowLKHu/Hi9ZcRHe5PenYun67cS/eXf+e2qatYuuOYxg2JiIiUYwpBIpVZcBQMmQ4jZ0H15pCZDPMfhbfawpafKuV4oTyebnZuaB3B7H93YsadbenRuBqGAQu3HWXYh3/Q9/UlfLU6nozsXKtLFRERkYukECQiUKcD3LkIBr0NfuFwIha+HAYfD4CDf1tdnaUMw6B9vap8MKINv/2nK8PbReLtbmfroZM89O3fdHjuN16Zv52jJytv75mIiEh5oxAkIk42G7S4Be5ZC53/D9y8YM8SeLcz/DAOTh62ukLLRVX15YlBTVk5oTsT+kZTI9CL46lZvPHrDjo89xv/+eovNh9ItrpMERERuQCFIBEpyNMPrvofjFsDTa8HTFj/CbzZEpa8DNkZVldouUAfd0Z3qcfih7ox+eYWtKgdRFaug2/X7aPfG0sY+t5KFmw+jMNReS8nFBERcWUKQSJStKAIuP5DuH0e1GwFWSnw6xPwVhvYNLNSjxfK42a3cfVlNZh5dwe+u7s9/S+rjt1msGL3cUZNX8NVLy/i4+V7SM3MsbpUEREROYNCkIicX+22cMcCuO598K8BiXvh65EwtS/sX2d1dS6jZe0qvHVzSxY/1I3Rnevi7+XGnuNpPP7jJto9+yvPztrC/sR0q8sUERERFIJEpDhsNrhsCNyzBrpOAHcf2LsC3u8GM/8FyQesrtBl1AzyZkK/xqyc0J1JA5tQJ8SH5Iwc3l28m84vLGTsjHWs23vC6jJFREQqNYUgESk+D1/o+rBzvNBlNzm3/fU5vNkKfn8BstKsrc+F+Hq6MaJ9HX77T1c+GN6adnVDyHWY/PL3Qa57eznXvr2Mn/8+QE6uw+pSRUREKh2FIBG5eIE14bp3YdRvENEWstNg4dMwuTX8/bXGC53BZjPoERPG53ddyS/3duT6VrXwsNtYvzeRcTPW0/mFhbz7+y6S0rOtLlVERKTSUAgSkUtXqxXcPheu/wgCa0PyfvhuFHzYE+JXF9zXkYsRt5SaCSsw4paCo/ItMtqkRiAv3dCcpQ93497uDQjx9eBAUgbPzt5Ku2d/5fEfNrLnWKrVZYqIiFR4blYXICLlnGFA08HQqB+seAuWvAL7VsOHPaDZDdBjonMChTnjcUs+QGuAuCkQUAP6PA8xAy1+AWWvmr8XD/RsyN1d6/HDn/v5aOketh0+yccr4pi+Mo7u0dW4vWMU7eqGYBiG1eWKiIhUOApBIlIy3L2h84PQYhj89iSs/ww2fA2bfgBHVuH9kw/CV8NhyPRKGYQAvNzt3NimNkNaR7Bs53E+XLqbhduOsmDLERZsOULj6gHc3qEOAy+vgaeb3epyRUREKgxdDiciJcs/HAa9BXctgtrtiw5AAJwaNzTn4Up5adyZDMOgY4OqTL3tCn79TxeGXVkbL3cbWw4m83/f/E2H5xby+oIdHEvJtLpUERGRCkEhSERKR43LoduEC+xkOscRrXgLTsRV+jAEUC/Uj6euacbKCd15qE8jwgO8OJaSyasLttP+ud8Y/83fbDt00uoyRUREyjVdDicipSflSPH2m/+o82b3hOC6EFLPeQuuByH1nd/7hTnHH1USQT4e3N21Pnd2qsusDQf5aGksf+1L4ss18Xy5Jp6O9atyR8coujQMxWarPO+LiIhISVAIEpHS4xdWvP0CIiD1MORmwtEtztvZPPxOBaT6p4NRSH3nNp/gkq3bhbjbbQy6vCYDm9dgbdwJPloWy5yNh1i68xhLdx6jbqgvt3WIYnDLmvh46J90ERGR4tD/mCJSeiLbO2eBSz5I/higAgzn4/f95bybFA/Hd8Lx3c6vCbucXxP3QlYKHPrbeTubd/AZoahewZ4kT7/SfIVlxjAMWtcJpnWdYOIT0vh4+R6+XB3P7qOpPPr9Rl6au42b29ZmRLs6hAd6WV2uiIiIS1MIEpHSY7M7p8H+ajhgUDAInbqEq89zzv0AqtRx3uqfdZycTOeYoTOD0fFdztvJA5CeAPsSnFNzn82/esFglBeUgqPAzbOkX3GZiAj24X9Xx3Bfz4Z8vSaeqcv2sDchjSmLdvH+4t30a1adOzpG0TwiyOpSRUREXJJCkIiUrpiBzmmw54yH5AOntwfUcAag4kyP7eYJoQ2dt7NlpULC7oLBKC8opR2Hkwedt7ilBZ9n2CCw1unL6/LHH9V1Lvxqd/1/Hv083bitQxTD29VhwZbDfLg0llWxCfz41wF+/OsArSOrcEfHKHo1Ccd+1rihXIfJH7EJrD1mEBKbQLv61QrtIyIiUlG5/v/yIlL+xQyE6P7k7F7Mn0vmcnmn3rjV7Xy6B+if8PCF8GbO29nSTzgvrcvvPTojKGWddF5ml7gXdv1W8Hk2d2ePVP7YozMmafCvDjbXmljTbjPo3SSc3k3C2bg/iY+WxvLT3wdYE3eCNXEnqFXFm5Ht6zCkTQQBXu7M2XiQST9t5mBSBmBn+o41VA/04vEBMfRpWt3qlyMiIlLqFIJEpGzY7JiRHdm/KZnmkR1LJgBdiHcVqNXKeTuTaULq0TNCUd5ldqduuZlwfIfzdjZ3n9Mz2J05e11IffAJsXwGu6Y1A3nlxssZ3zeaT1bE8dkfcew7kc5Tv2zhtQU7aBNVhYVbjxZ63qGkDMZ8uo4pw1oqCImISIWnECQilY9hgF815y2yfcHHHA7n2kVnB6PjOyExDrLT4PBG5+1snoGnA9GZs9eF1AOvwLJ5baeEBXjxYO9GjLuqPjPX7+ejpbHsOJJSZAAC52gtA5j002Z6xhS+fE5ERKQiUQgSETmTzQZBEc5bvW4FH8vNdl4+V1QPUtI+yEyCA+uct7P5hhaevS4vJLl7l9rL8XK3M/SK2tzUJoJ3F+/mudlbz7mvCRxMymBVbALt6oWUWk0iIiJWUwgSESkuu/vpAHO27HRIiD1r/NGpCRtSjzgvv0s9CntXFH5uQK3Cs9eF1Icqkc5zlgDDMKhezKmzX5q3lUGX16Rl7SpEh/vjZnetMVAiIiL/lEKQiEhJcPeGsBjn7WwZyQUvrTszKGUkQfI+5y3294LPM+zOIFRg7NGp7wNqXfQEDdX8T4cgGw6usG2lGokcIYhVjmgcOI+3Ni6RtXGJAHi722keEUiryCq0iqxCi4gqVPH1uKjzioiIuBqFIBGR0uYVADVaOG9nMk1ISyh6/aOEXc7xRwm7nbed8ws+1+55erxRgUka6jvHOhUxQcMVUcFUD/Si+cnFPOY+nRpGQv5jB8xgnsgezirvjtx6ZSTr4xNZv/cEJzNyWLk7gZW7T+9bN9SXlrWdoahl7So0qOaHTWOIRESkHFEIEhGximGAb4jzVrttwcdM07m+UYHxR6cur0uIdc5gd3SL83Y2D78iZ6+zB9fl7Zb7aL78tUJPCSeBt91f468r6tKiZ08AHA6TnUdTWBt3gnVxJ1i79wS7j6bm375Zuw8Afy83WtSuQsvaQbSKrMLlEUH4e5XMZXwiIiKlQSFIRMQVGYZzQdmAGhDVueBjuTmQFH/GJXZnBKWkeMhKgYN/OW9naYGBaThngjuTzXBOjNBi0/PQcxjY7NhsBg3D/GkY5s/QK2oDcCI1i/XxJ04Fo0T+jE/kZEYOi7cfZfH2o/mlNwrzp0V+b1EQUVV9MSyePlxERCSPQpCISHljd4PgKOetfo+Cj+Vkwok9Rax/tNPZs4RZKADlMcA5PfgrMc7FYv2qgX/4qenEw8EvjCp+1biqZjhXNagPdjdych1sPXSSdXtP9xbFJ6Sz9dBJth46yeer9gIQ7OtBi4ggWp4aW3RZrUB8PPRfkIiIWEP/A4mIVCRunhDayHk727pP4cexFz5GyiHn7bwM8A3FzS+Mpn7VaOofzvCq1aBOOEluVdh60of1JzxYetDOqoPZJKRm8evWI/y69QgAdptBTPUAWtZ2BqOWtatQq4q3eotERKRMKASJiFQWVSKLt1+f5509QCmHnbeTh09/n3LYOdW36Tg19fcROFzw6YFA21O3fwGmty9ZXlU5YavCgdxAdqX5EJvpz9FDQcQdDGTNyiocMQOx+4XSok5VWtauQsvIKjStGYCnm71k3wMREREUgkREKo/I9s4xRsnOy+IKOzUO6Yo7wXae8OHIhdRjp0LRkVM9R2eGpbxtRyArBSM7Fc/sVMKJIxxoCVDEvAm52QYJOwI4uj2II2YQPxtB4BeGX0hNqtWIJDIyiuCwCPALA0+/knhHRESkklIIEhGpLGx2Zy/PV8NxjgA6Mwidugytz3PnD0B5x/EPc94uJDOlcDA6eepryuH8bWbqUew4CCWJUCOJGOKcz087dYsH/jh92Gy7Nw7fMNwDw7H5h50as5Q3hins9Dgm36oXfj0iIlLpKASJiFQmMQNhyHSYMx6SD5zeHlDDGYBiBpbs+Tz9nLeQeufdzSjQu3QY8+QhEo/s4/jheNJPHMBIOYxf1nFCjUR8jUzcc9MheY/zdt4D28A39FQoOjss5W0Lc43eJUcuRtxSaiaswIgLgLqdFeBEREqJQpCISGUTMxCi+5OzezF/LpnL5Z1642b1B+6zepcMoMqpW56UzBz+jE/k7937id2zmyMH4vDJOk6okUQ14wShJFHNSKSmezJhtiT8chKxmY7TY5nYcP4aPPwKByP/sILhyS+sdHqXNv8Ic8bjlnyA1gBxU04F0+dLPpiKiIhCkIhIpWSzY0Z2ZP+mZJpHdiwXPQ5+nm50qF+VDvWrAs1xOEx2H0vJX7No5t4T7DySAtnO/e3kEsxJIj1P0jY0m8uCMmnom0oNt5N4ph85fUneycOQnepcXykhxbko7fnk9y4VFZbCTl+O5x8OHr4XfmGbfzx1ieJZ47SSDzq3D5muICQiUsIsDUHPPvss3333HVu3bsXb25v27dvz/PPP06hREVO7ioiInMFmM6hfzZ/61fy5sY1zMdfEtCzWxyeyLu4E6/ae4M+9HqzJDGLNPmCf83mGAQ2q+dEqsgotLnOuW1Q3wMTIH6dUxIx4edvyZsbL7126AA+/wsHojHWX8AmBWf9H0RNVmIABcx6G6P7lIqiKiJQXloag33//nbFjx9KmTRtycnL473//S69evdi8eTO+vsX465mIiMgZgnw86NaoGt0aVQMg12Gy7dBJ1u49wfpTi7nGHU9j++EUth9O4fNV8aee5+6cmrt2KC0jG9K8fhC+nkX8F5mbA2lnzIx38tBZYemMbdlpZ/Qu7brEV2Q6F7D99g6o2hDcvcHN2/nV3efU1zNvPqe/unk5v1d4unQapyVSYVkagubMmVPg/rRp06hWrRpr166lc+fOFlUlIiIVhd1mEFMjgJgaAdx6pXOdpGMpmaw7FYjWxyXy175EEtOy+W3rEX47YzHX6HB/Wp1ayLVV5KnFXO1uzt4c//ALnzzz5BmX3B0qOENeXs9SYhxkJl/4WJtm/oM3wfOMgOR1VnjyOStYFRWuzrhfKICd8VhFCwcapyVSobnUmKCkpCQAgoODi3w8MzOTzMzM/PvJyc7/OLKzs8nOzi79As8j7/xW1yGnqU1cj9rEtVTW9gj0tNGtYQjdGoYAkJXjYOuhk6yLT+TPvUmsi0/kYFIGmw4ks+lAMtNXOKfrrurnQYuIIFrWDqJFRCBNawTg6X6eD/42Lwio7bydgxG3FLdPr7lgzbmNrwGfEIzsdMhJg+x05y0nw7kt+9S2nHTIzsDIST/jyZnOW0ZiMd6dS2faPQoFKvPscOXmhZkfqPIClNcZ+50ZuLwwiwppttL/6GJs/Rn7t7cBZt7k8c7XeGqcVu7gqZjRV5d6HXIWRy65sc6eudxdvhBVPsYzVnSu9H/JxdRgmKZZ1IXIZc7hcDBw4EASExNZunRpkftMnDiRSZMmFdo+Y8YMfHx8SrtEERGpJBIzITbFYM9Jg9iTBvtSIdc0CuxjN0xq+UKUv0kdf5MoP5Mgz4s8kemg16YH8MpOwCjqYSDdPZj5TV5xTshwEce1m9nYHVnYHZnYHdmnvmadsS0Lu5lVeNuZN7OIbWfs52ZmXeQL/ucchp1cmye5hgc5Ng8cNg9ybR7k2DzJPfW949Rjzvuntxe4b+R9715wH8Odq7b+F6/sEyXbJvKPVE9cTbN9n+GdnZC/Ld09mA21buFgUBsLK6vkTAchKdvwyk4kwz2I436NLP29SEtL4+abbyYpKYmAgIDz7usyIWjMmDHMnj2bpUuXUqtWrSL3KaonKCIigmPHjl3whZa27Oxs5s+fT8+ePXF3L2IpdClzahPXozZxLWqP4svMzmXjgWTWxSeyfm8S6+MTOZZSOADUCPSiRUQQLWoH0iIiiMbV/XG3n/8DweleBzDOmCAhrw/CpXsdTBNyMs7ohUo73RNVoIeqqB6r9FM9W+mne7ay0894bsHHjCInj7COo2oj8AkGmzvYPZw9VHZ3583m/Gqe+pq/zebm3Nd+6qvNHbOIbQX2K3Csgtso6vhGUdGtfDtnz1x5+B2pwIytP2Of91+Mk6fXnDP9a5Db6xnL2iM5OZmqVasWKwS5xOVw48aN4+eff2bx4sXnDEAAnp6eeHoW/jObu7u7y/wH7kq1iJPaxPWoTVyL2uPC3N3dubK+F1fWd064YJom+06kszbuhHOK7r0n2HIwmQNJGRxIOsQvGw8B4OVu47KaQbSMrELL2s6vVf3O+n+s2bWsP5BMjRWTCON4/ubDBHOw3eO0aHZtmb3OS+LhAZTyHyJNE3IyT4eovECVk3HGtrQCYYoiQlfB/ZyBrdC2YoYt27FtpfuaL9XZwajA92eGKI9T291Of1/kc873/LOOddHPP+v8RV3a5siF+f+lqHYxTs2g6Db/EWgyUJfGlaXNP8KpYHom4+RB3L69zbKp/S/m/zJLQ5Bpmtxzzz3MnDmTRYsWERUVZWU5IiIixWIYBhHBPkQE+3BNi5oApGbm8Ne+vOm5E1m39wSJadms2pPAqj2nL+GJDPGhVe0qp4JRFfYcS2XswqoYvM4Vtq1UI5EjBLHaEY1joY0pNQ/Sp2l1q16qazCMU5M6eJXueUwTdi2ET4sRPLs9AlUbOGcMzM0CRzbknro5sp3bcnMKfp+/37meU8T3uVngyDnr+yznfTO3cF2OU8+3fnjGxTNshYOT6XDOyHhOp2ZQfKcTeAedvhTLME59f+qrYZzne4qxz9nfG6W8v825avRF7V8Gr8WwgcMBP99HeZ/a39IQNHbsWGbMmMEPP/yAv78/hw45/3IWGBiIt7e3laWJiIhcFF9PN9rXq0r7elUB5x/6dh9LZW3cCdbvdfYYbT+cQtzxNOKOp/Hd+v2A83OOCZjYWOmIKXBMA5j002Z6xoRjt1W8y5xcjmFA3S7OWeCSD1L0hzzD+Xin/1j/Ac/hOB2YihucLiVsnev5RT7nAs8/MySezXScnsjjYh3Z9M/fTykhp4Jp3HKI6mR1MedkaQiaMmUKAF27di2wferUqYwcObLsCxIRESkhhmFQL9SPeqF+DGkdAUBSejZ/xifmB6PVexLIyHac8xgmcDApg6d+2UyXhqHUDvahVhUfPNw0IL/U2OzOabC/Gs7piJrnVBDt85z1AQjAZgObJ7hd7IwcLsA0nZe6FdmDdkZwil8Nv9x/4eN1GQ+h0YDpPLZpnvrecRHfc5H7X8q5irP/qfuF9j97+/m+p/T2TzsOSfEXbpPiLChtIcsvhxMREaksAr3d6dIwlC4NQwGYuX4/93/55wWfN3XZHqYu2wOAzYDqgd7UDvZx3kKcXyNPfQ3y8SjFV1BJxAx0jmmYMx6STw/6dq4T9JzWCSoJhnFqMgg359Tn51ItBpa8eOGeuS7jXSOYVgaxS+DjYkx84BdW+rX8Ay4xMYKIiEhlFB5QvDEurSKrkJqZQ9zxNNKzc9mfmM7+xHRW7D5eaN8ALzdqh/gQGexLxBnhqHawD9UDvXC7wGx1ckrMQIjuT87uxfy5ZC6Xd+qNW93O+qBd1spTz1xlEdm+eJeMRrYv68ouikKQiIiIRa6ICqZ6oBeHkjLO9VGC8EAvvhrdDrvNwDRNjqVksTchlb0Jaew9nk5cQirxCc5xRkdOZpKckcPG/cls3J9c6HhuNoNaVbwLhaPawb7UDvHBz1MfCwqw2TEjO7J/UzLNI7Uwp2XUM+daKkgw1b92IiIiFrHbDB4fEMOYT9ed66MEjw+IyZ8UwTAMQv09CfX3pFVkcKHjpWflEn8ijb3H04hLSDsVjpyBKf5EOlk5DvYcT2PP8TSW7ChcT4ivRxEByYfIEF+q+Xti0+QMYhX1zLmWChBMFYJEREQs1KdpdaYMa8mknzZzMCkjf3t4oBePD4i5qOmxvT3sNAzzp2GYf6HHHA6TwycziDuedqoXyfk1LywlpGZx/NTtz/jEQs/3dLMREXx2OHJ+jQj2wctdH0allKlnzrWU82CqECQiImKxPk2r0zMmnBU7jzBvyR/06tSWdvWrlei02DabQfVAb6oHenNl3ZBCj5/MyC4yHMUdT2N/YjqZOQ52Hklh55GUIo8fFuBZeBzSqa8hvh4YhnqRRCqcchxMFYJERERcgN1m0DYqmONbTNpGBZf5ukD+Xu40qRFIkxqBhR7LyXVwIDHDGZIS0gqMQ9p7PI2TmTkcTs7kcHJmgYVh8/h62PN7kU4HJF9qB/tQM8hbU36LSJlTCBIREZHzcrPbnL06IT6FHjNNk8S07LN6j1Lze5UOJmeQmpXL1kMn2XroZKHn5035fXbvUWSwMyQF+riXxUsUkUpGIUhEREQumWEYVPH1oIqvB80jggo9npmTy74T6QUvtTvuDEt7EwpO+b18V+EpvwO93c8KR6fDUvVA71LtMct1mPwRm8DaYwYhsQklfomiiFhHIUhERERKjaebnXqhftQL9Sv0mGmaHE3JLDIcxSWkcfRkJknp2WzYn8SG/UmFnu9uN6hVxTkxQ144OnNMku8/mPJ7zsaDZ0xWYWf6jjVUv4TJKkTENSkEiYiIiCUMw6CavxfV/L1oXafwlN9pWTnEJ6SfCkinxiGdCkn7EtLJynUQeyyV2GOpRR6/qp9HgYCUNw4pMsSHUL9zT/k9Z+NBxny6rtDaTYeSMhjz6TqmDGupICRSzikEiYiIiEvy8XCjUbg/jcILT/md6zA5nJyR33sUl5DK3oR09p4aj3QiLZtjKVkcS8li/d7EQs/3dLOdnu475PSkDTWDvJn446YiF681ca7fNOmnzfSMCdelcSLlmEKQiIiIlDt2m0GNIG9qBHnTrl7hKb+TM7LzL7M781K7uIRUDiRmkJnjYMeRFHacY8rvczGBg0kZzNl4iJ4xYZrZTqScUggSERGRCifAy52mNQNpWrPwlN/ZuQ4OJKYXDEenAtOuoylk5jguePyxM9YB4O/lRlU/T4J9PQjx9SDEz4MQ31P3/TxOP+bnQbCPB252hSYRV6AQJCIiIpWKu91GZIgvkSG+dGpQ8LEVu44x9P0/LngMmwEOE05m5HAyI+ec45LOFuTjTrCvB1XPCErO8HTmfU9C/Dyo4uOhS+5ESolCkIiIiMgpV0SFUD3Qi0NJGUWOCzKA8EAvFv9fN1KzcjiWkkVCahbHUzI5nprF8ZQsElIzOZaaRUJKFsdTMzmeksWJtCwcJiSmZZOYls3uoxcOTYYBVXw8iuxlqurnQfCpsJQXooK83c852YOIFKQQJCIiInKK3Wbw+IAYxny6DgMKBKG8ePH4gBjc3WwEuXkQ5ONRrOPmOkwS05yBKT84nQpIx1MzC25PySQxPRvThIRU57adxTiHzeBUYCq6l+ns4BTgpdAklZdCkIiIiMgZ+jStzpRhLc9YJ8gp/B+sE2S3GYT4eRLi50mDsAvvn5Pr4ERadn4ocvYsneptOrUt4VTP0/HULJLSs3GY5M+IVxxuNudCt8XpZQr29SDAyw3DsC40afFaKUkKQSIiIiJn6dO0Oj1jwlmx8wjzlvxBr05ty/RDt5vdRqi/J6H+nkDhKcLPlp3r4MQFepnODE4nM3PIcZgcPZnJ0ZOZxarJ3W4Uu5cpxM8TXw97iYUmLV4rJU0hSERERKQIdptB26hgjm8xaRsV7NK9Du52G9UCvKgW4FWs/TNzcgv0JCXkh6bTYenMy/NSs3LJzjU5lJzBoeSMC58A8HCzUdXXg+C8yR7yZsk7KyzlbffxKPpjqRavldKgECQiIiJSyXi62ake6E31QO9i7Z+RnZsfkM6cAKKo4HQ8NZOMbAdZOQ4OJGVwIKl4ocnL3ZY/M15eQKri484Xq+O1eK2UOIUgERERETkvL3c7NYO8qRlUvNCUlpVToJepwCx6ecEpNZOElCyOpWaRleMgI9vB/sR09iemF7uuvMVrWzwxj0Afd3zc3fDysOPjbsfHw463h/Orj8f/t3f/MVXVjx/HXwcuPy6IJPJDyN9fzdQCf0CG6Gf5I5WcRTMtR4Vacxqa5txSV0Ll1O/6RS29pVNsXzNKN8ivExlaUbmcv4ZhmlmZWqhomQIqvy7fPxDyiil+Jc/B83xsd7ucc7j3BW+ZvDjv9zmOuuc+9dscV+z3ltPH8ffzS8dQrK6tJa/TogQBAACgWQX4OhQQ4lCHkIDrHltbW6vyyprGlxkvq9TOX//UlwdPXfc1zl2s1rmL1c0R3YOvw6uuGF1Wnv6pOPn7XLb98pLlc5XP8/WWr7eXqReauFktfZ0WJQgAAACmMQxDrfwcauXnUKe2gR77vv35jyaVoP8ee6+6hQfpQmWNzldW60JVjc5X1j0uVFZf9rxG56vqttUfc6HyimOralR7af5dZXXdtL6/VNXsX7e3l3FZufq7ONWVq6tsu0qhaihmV5z5cvo030UpruZ2WKdFCQIAAIAl3dclpEk3r32sf4dmm4ZVW1urimr3pWJU7VmSqq4oVJeXrKqahhLmsf+y7Req6i4wIdVNJSutqFZpRfOfwZLUuBj5OpplmqCfw1uv/O/+Fr9OixIEAAAAS2rqzWub85dtwzDk71M3vS0ksGk3w70RVTXuy0pSdUNRuupZq8oana+qvqxwXe1z/t5WUe1ueJ8LVXXHqLzZv4Rrql+ntePwn4r/r7a39s1vACUIAAAAlvVv3LzWTD7eXgp2einY6dPsr13jrtXFy8tRVfUVU/6qPc5QXXnW6vKzXdeaJtgUJaVNuyqgWShBAAAAsDSzb17bUnh7GQr0cyjQr/l/xa+fJvjVj6c05X92X/f48KCm3bPKLF5mBwAAAACup/7mtf1DrX/z2ttR/TTBYT0jFBnsr3/67huSIoP9dV+XkFsZ74ZRggAAAAA0Sf06LUmNitC/tU7r30AJAgAAANBk9eu02gV7TnlrF+zfIi6PLbEmCAAAAMANaunrtChBAAAAAG5Y/TqtPw60vHVaTIcDAAAAYCuUIAAAAAC2QgkCAAAAYCuUIAAAAAC2QgkCAAAAYCuUIAAAAAC2QgkCAAAAYCuUIAAAAAC2QgkCAAAAYCuUIAAAAAC24jA7wM2ora2VJJ07d87kJFJVVZXOnz+vc+fOycfHx+w4EGNiRYyJtTAe1sOYWA9jYi2Mh/VYaUzqO0F9R7iWFl2CSktLJUkdOnQwOQkAAAAAKygtLVVwcPA1jzFqm1KVLMrtdqu4uFhBQUEyDMPULOfOnVOHDh107NgxtW7d2tQsqMOYWA9jYi2Mh/UwJtbDmFgL42E9VhqT2tpalZaWKioqSl5e117106LPBHl5eal9+/Zmx/DQunVr0/8BwBNjYj2MibUwHtbDmFgPY2ItjIf1WGVMrncGqB4XRgAAAABgK5QgAAAAALZCCWomfn5+SktLk5+fn9lRcAljYj2MibUwHtbDmFgPY2ItjIf1tNQxadEXRgAAAACAG8WZIAAAAAC2QgkCAAAAYCuUIAAAAAC2QgkCAAAAYCuUoGaydOlSde7cWf7+/howYIB27NhhdiTb+uqrrzRmzBhFRUXJMAzl5OSYHcnWFi9erLi4OAUFBSk8PFxJSUk6ePCg2bFszeVyKTo6uuHGdvHx8crNzTU7Fi5ZsmSJDMPQrFmzzI5iW+np6TIMw+Nx9913mx3L9n7//Xc9+eSTatu2rZxOp+69917t2rXL7Fi21blz50Y/J4ZhKDU11exoTUIJagaffPKJZs+erbS0NO3Zs0cxMTEaOXKkSkpKzI5mS+Xl5YqJidHSpUvNjgJJBQUFSk1N1fbt25Wfn6+qqiqNGDFC5eXlZkezrfbt22vJkiXavXu3du3apaFDh+qRRx7R999/b3Y029u5c6c++OADRUdHmx3F9nr37q3jx483PL755huzI9namTNnlJCQIB8fH+Xm5mr//v1688031aZNG7Oj2dbOnTs9fkby8/MlSePGjTM5WdNwiexmMGDAAMXFxem9996TJLndbnXo0EEzZszQ3LlzTU5nb4ZhKDs7W0lJSWZHwSWnTp1SeHi4CgoK9J///MfsOLgkJCREr7/+up555hmzo9hWWVmZ+vXrp2XLlmnhwoXq06ePMjIyzI5lS+np6crJyVFhYaHZUXDJ3LlztW3bNn399ddmR8E/mDVrljZu3KhDhw7JMAyz41wXZ4JuUmVlpXbv3q3hw4c3bPPy8tLw4cP17bffmpgMsKazZ89KqvulG+arqalRVlaWysvLFR8fb3YcW0tNTdXo0aM9/j+BeQ4dOqSoqCh17dpVycnJOnr0qNmRbG3Dhg2KjY3VuHHjFB4err59+2rFihVmx8IllZWVWrNmjSZPntwiCpBECbppp0+fVk1NjSIiIjy2R0RE6MSJEyalAqzJ7XZr1qxZSkhI0D333GN2HFsrKipSq1at5Ofnp6lTpyo7O1u9evUyO5ZtZWVlac+ePVq8eLHZUaC6GR6rV6/W5s2b5XK5dPjwYQ0ePFilpaVmR7OtX375RS6XS927d1deXp6mTZum559/Xh9++KHZ0SApJydHf/31lyZOnGh2lCZzmB0AgH2kpqZq3759zK23gB49eqiwsFBnz57V+vXrlZKSooKCAoqQCY4dO6aZM2cqPz9f/v7+ZseBpMTExIbn0dHRGjBggDp16qRPP/2UKaMmcbvdio2N1aJFiyRJffv21b59+/T+++8rJSXF5HRYuXKlEhMTFRUVZXaUJuNM0E0KDQ2Vt7e3Tp486bH95MmTateunUmpAOuZPn26Nm7cqC+++ELt27c3O47t+fr6qlu3burfv78WL16smJgYvfPOO2bHsqXdu3erpKRE/fr1k8PhkMPhUEFBgd599105HA7V1NSYHdH27rjjDt1111366aefzI5iW5GRkY3+SNOzZ0+mKVrAkSNHtGXLFj377LNmR7khlKCb5Ovrq/79+2vr1q0N29xut7Zu3cr8ekBSbW2tpk+fruzsbH3++efq0qWL2ZFwFW63WxUVFWbHsKVhw4apqKhIhYWFDY/Y2FglJyersLBQ3t7eZke0vbKyMv3888+KjIw0O4ptJSQkNLq9wo8//qhOnTqZlAj1MjMzFR4ertGjR5sd5YYwHa4ZzJ49WykpKYqNjdV9992njIwMlZeXa9KkSWZHs6WysjKPv9YdPnxYhYWFCgkJUceOHU1MZk+pqalau3atPvvsMwUFBTWslQsODpbT6TQ5nT3NmzdPiYmJ6tixo0pLS7V27Vp9+eWXysvLMzuaLQUFBTVaIxcYGKi2bduyds4kc+bM0ZgxY9SpUycVFxcrLS1N3t7emjBhgtnRbOuFF17QwIEDtWjRIo0fP147duzQ8uXLtXz5crOj2Zrb7VZmZqZSUlLkcLSsWtGy0lrU448/rlOnTmnBggU6ceKE+vTpo82bNze6WAJujV27dmnIkCENH8+ePVuSlJKSotWrV5uUyr5cLpck6YEHHvDYnpmZ2aIWUN5OSkpK9PTTT+v48eMKDg5WdHS08vLy9OCDD5odDbCE3377TRMmTNAff/yhsLAwDRo0SNu3b1dYWJjZ0WwrLi5O2dnZmjdvnl599VV16dJFGRkZSk5ONjuarW3ZskVHjx7V5MmTzY5yw7hPEAAAAABbYU0QAAAAAFuhBAEAAACwFUoQAAAAAFuhBAEAAACwFUoQAAAAAFuhBAEAAACwFUoQAAAAAFuhBAEAAACwFUoQAMA2DMNQTk6O2TEAACajBAEAbomJEyfKMIxGj1GjRpkdDQBgMw6zAwAA7GPUqFHKzMz02Obn52dSGgCAXXEmCABwy/j5+aldu3YejzZt2kiqm6rmcrmUmJgop9Oprl27av369R6fX1RUpKFDh8rpdKpt27aaMmWKysrKPI5ZtWqVevfuLT8/P0VGRmr69Oke+0+fPq1HH31UAQEB6t69uzZs2NCw78yZM0pOTlZYWJicTqe6d+/eqLQBAFo+ShAAwDJefvlljR07Vnv37lVycrKeeOIJHThwQJJUXl6ukSNHqk2bNtq5c6fWrVunLVu2eJQcl8ul1NRUTZkyRUVFRdqwYYO6devm8R6vvPKKxo8fr++++04PPfSQkpOT9eeffza8//79+5Wbm6sDBw7I5XIpNDT01n0DAAC3hFFbW1trdggAwO1v4sSJWrNmjfz9/T22z58/X/Pnz5dhGJo6dapcLlfDvvvvv1/9+vXTsmXLtGLFCr344os6duyYAgMDJUmbNm3SmDFjVFxcrIiICN15552aNGmSFi5ceNUMhmHopZde0muvvSaprli1atVKubm5GjVqlB5++GGFhoZq1apV/9J3AQBgBawJAgDcMkOGDPEoOZIUEhLS8Dw+Pt5jX3x8vAoLCyVJBw4cUExMTEMBkqSEhAS53W4dPHhQhmGouLhYw4YNu2aG6OjohueBgYFq3bq1SkpKJEnTpk3T2LFjtWfPHo0YMUJJSUkaOHDg/+trBQBYFyUIAHDLBAYGNpqe1lycTmeTjvPx8fH42DAMud1uSVJiYqKOHDmiTZs2KT8/X8OGDVNqaqreeOONZs8LADAPa4IAAJaxffv2Rh/37NlTktSzZ0/t3btX5eXlDfu3bdsmLy8v9ejRQ0FBQercubO2bt16UxnCwsKUkpKiNWvWKCMjQ8uXL7+p1wMAWA9nggAAt0xFRYVOnDjhsc3hcDRcfGDdunWKjY3VoEGD9NFHH2nHjh1auXKlJCk5OVlpaWlKSUlRenq6Tp06pRkzZuipp55SRESEJCk9PV1Tp05VeHi4EhMTVVpaqm3btmnGjBlNyrdgwQL1799fvXv3VkVFhTZu3NhQwgAAtw9KEADgltm8ebMiIyM9tvXo0UM//PCDpLort2VlZem5555TZGSkPv74Y/Xq1UuSFBAQoLy8PM2cOVNxcXEKCAjQ2LFj9dZbbzW8VkpKii5evKi3335bc+bMUWhoqB577LEm5/P19dW8efP066+/yul0avDgwcrKymqGrxwAYCVcHQ4AYAmGYSg7O1tJSUlmRwEA3OZYEwQAAADAVihBAAAAAGyFNUEAAEtgdjYA4FbhTBAAAAAAW6EEAQAAALAVShAAAAAAW6EEAQAAALAVShAAAAAAW6EEAQAAALAVShAAAAAAW6EEAQAAALCV/wNHGbbTFGKJvgAAAABJRU5ErkJggg==\n"},"metadata":{}}],"execution_count":10},{"cell_type":"code","source":"from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score\n\ndef final_report(model, loader, aspect_names):\n    model.eval()\n    all_preds, all_labels = [], []\n    \n    with torch.no_grad():\n        for batch in tqdm(loader, desc=\"Calculating Metrics\"):\n            ids = batch['input_ids'].to(device)\n            mask = batch['attention_mask'].to(device)\n            labels = batch['labels'].to(device)\n            \n            outputs = model(ids, attention_mask=mask)\n            logits = outputs.logits.view(-1, len(aspect_names), 3)\n            \n            preds = torch.argmax(logits, dim=2)\n            all_preds.append(preds.cpu().numpy())\n            all_labels.append(labels.cpu().numpy())\n    \n    all_preds = np.vstack(all_preds)\n    all_labels = np.vstack(all_labels)\n    \n    res = []\n    for i, col in enumerate(aspect_names):\n        acc = accuracy_score(all_labels[:, i], all_preds[:, i])\n        prec = precision_score(all_labels[:, i], all_preds[:, i], average='macro', zero_division=0)\n        rec = recall_score(all_labels[:, i], all_preds[:, i], average='macro', zero_division=0)\n        f1 = f1_score(all_labels[:, i], all_preds[:, i], average='macro', zero_division=0)\n        \n        res.append({\n            'Aspect': col,\n            'Accuracy': acc,\n            'Precision': prec,\n            'Recall': rec,\n            'F1-Score': f1\n        })\n    \n    report_df = pd.DataFrame(res)\n    \n    avg_row = {\n        'Aspect': '--- Total Average ---',\n        'Accuracy': report_df['Accuracy'].mean(),\n        'Precision': report_df['Precision'].mean(),\n        'Recall': report_df['Recall'].mean(),\n        'F1-Score': report_df['F1-Score'].mean()\n    }\n    \n    report_df = pd.concat([report_df, pd.DataFrame([avg_row])], ignore_index=True)\n    \n    return report_df\n\nfinal_results = final_report(model, test_loader, aspect_columns)\ndisplay(final_results)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-04-16T12:38:34.742457Z","iopub.execute_input":"2026-04-16T12:38:34.742813Z","iopub.status.idle":"2026-04-16T12:38:38.118902Z","shell.execute_reply.started":"2026-04-16T12:38:34.742779Z","shell.execute_reply":"2026-04-16T12:38:38.117829Z"}},"outputs":[{"name":"stderr","text":"Calculating Metrics: 100%|██████████| 26/26 [00:03<00:00,  7.90it/s]\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"                  Aspect  Accuracy  Precision    Recall  F1-Score\n0                   Size  0.903846   0.805497  0.762894  0.776541\n1                  Color  0.947115   0.906481  0.802314  0.845312\n2                  Price  0.987981   0.973641  0.945006  0.958900\n3                  Smell  0.992788   0.978885  0.971126  0.974693\n4                Quality  0.896635   0.706694  0.435723  0.479099\n5                 Fabric  0.882212   0.835148  0.846762  0.840293\n6                  Style  0.790865   0.677439  0.517921  0.555524\n7                  Image  0.954327   0.883657  0.826787  0.852639\n8  --- Total Average ---  0.919471   0.845930  0.763566  0.785375","text/html":"<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>Aspect</th>\n      <th>Accuracy</th>\n      <th>Precision</th>\n      <th>Recall</th>\n      <th>F1-Score</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>Size</td>\n      <td>0.903846</td>\n      <td>0.805497</td>\n      <td>0.762894</td>\n      <td>0.776541</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>Color</td>\n      <td>0.947115</td>\n      <td>0.906481</td>\n      <td>0.802314</td>\n      <td>0.845312</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>Price</td>\n      <td>0.987981</td>\n      <td>0.973641</td>\n      <td>0.945006</td>\n      <td>0.958900</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>Smell</td>\n      <td>0.992788</td>\n      <td>0.978885</td>\n      <td>0.971126</td>\n      <td>0.974693</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>Quality</td>\n      <td>0.896635</td>\n      <td>0.706694</td>\n      <td>0.435723</td>\n      <td>0.479099</td>\n    </tr>\n    <tr>\n      <th>5</th>\n      <td>Fabric</td>\n      <td>0.882212</td>\n      <td>0.835148</td>\n      <td>0.846762</td>\n      <td>0.840293</td>\n    </tr>\n    <tr>\n      <th>6</th>\n      <td>Style</td>\n      <td>0.790865</td>\n      <td>0.677439</td>\n      <td>0.517921</td>\n      <td>0.555524</td>\n    </tr>\n    <tr>\n      <th>7</th>\n      <td>Image</td>\n      <td>0.954327</td>\n      <td>0.883657</td>\n      <td>0.826787</td>\n      <td>0.852639</td>\n    </tr>\n    <tr>\n      <th>8</th>\n      <td>--- Total Average ---</td>\n      <td>0.919471</td>\n      <td>0.845930</td>\n      <td>0.763566</td>\n      <td>0.785375</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}}],"execution_count":11},{"cell_type":"code","source":"import pandas as pd\n\ndef predict_review_to_table(text):\n    model.eval()\n    clean_text = full_clean(text)\n    inputs = tokenizer(clean_text, return_tensors=\"pt\", padding=\"max_length\", truncation=True, max_length=128).to(device)\n    \n    with torch.no_grad():\n        outputs = model(**inputs)\n        logits = outputs.logits.view(len(aspect_columns), 3)\n        preds = torch.argmax(logits, dim=1).cpu().numpy()\n    \n    sentiment_map = {0: \"Negative\", 1: \"Neutral\", 2: \"Positive\"}\n    \n    results = []\n    for i, col in enumerate(aspect_columns):\n        results.append({\n            \"Aspect\": col,\n            \"Sentiment\": sentiment_map[preds[i]]\n        })\n    \n    df_result = pd.DataFrame(results)\n    \n    print(f\"{text}\")\n    return df_result\n\nprediction_table = predict_review_to_table(\"حلو لونه و القماش كمان بس لو كان اوسع شوي \")\ndisplay(prediction_table)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-04-16T12:38:41.946082Z","iopub.execute_input":"2026-04-16T12:38:41.946746Z","iopub.status.idle":"2026-04-16T12:38:41.999555Z","shell.execute_reply.started":"2026-04-16T12:38:41.946719Z","shell.execute_reply":"2026-04-16T12:38:41.998840Z"}},"outputs":[{"name":"stdout","text":"حلو لونه و القماش كمان بس لو كان اوسع شوي \n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"    Aspect Sentiment\n0     Size  Negative\n1    Color  Positive\n2    Price   Neutral\n3    Smell   Neutral\n4  Quality   Neutral\n5   Fabric  Positive\n6    Style   Neutral\n7    Image   Neutral","text/html":"<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>Aspect</th>\n      <th>Sentiment</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>Size</td>\n      <td>Negative</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>Color</td>\n      <td>Positive</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>Price</td>\n      <td>Neutral</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>Smell</td>\n      <td>Neutral</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>Quality</td>\n      <td>Neutral</td>\n    </tr>\n    <tr>\n      <th>5</th>\n      <td>Fabric</td>\n      <td>Positive</td>\n    </tr>\n    <tr>\n      <th>6</th>\n      <td>Style</td>\n      <td>Neutral</td>\n    </tr>\n    <tr>\n      <th>7</th>\n      <td>Image</td>\n      <td>Neutral</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}}],"execution_count":14},{"cell_type":"code","source":"import torch\nfrom tqdm import tqdm\n\nmodel.eval()\n\nall_labels = []\nall_preds = []\n\nwith torch.no_grad():\n    for batch in tqdm(val_loader, desc=\"Calculating Predictions\"):\n        ids = batch['input_ids'].to(device)\n        mask = batch['attention_mask'].to(device)\n        labels = batch['labels'].to(device)\n        \n        outputs = model(ids, attention_mask=mask)\n        \n        logits = outputs.logits.view(-1, 8, 3)\n        preds = torch.argmax(logits, dim=2)\n        \n        all_labels.extend(labels.cpu().numpy())\n        all_preds.extend(preds.cpu().numpy())\n\nall_labels = np.array(all_labels)\nall_preds = np.array(all_preds)\n\nprint(\"done\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-04-16T12:38:42.000714Z","iopub.execute_input":"2026-04-16T12:38:42.001134Z","iopub.status.idle":"2026-04-16T12:38:45.315476Z","shell.execute_reply.started":"2026-04-16T12:38:42.001100Z","shell.execute_reply":"2026-04-16T12:38:45.314602Z"}},"outputs":[{"name":"stderr","text":"Calculating Predictions: 100%|██████████| 26/26 [00:03<00:00,  7.87it/s]","output_type":"stream"},{"name":"stdout","text":"✅ تم استخراج النتائج بنجاح!\n","output_type":"stream"},{"name":"stderr","text":"\n","output_type":"stream"}],"execution_count":15},{"cell_type":"code","source":"import seaborn as sns\nimport matplotlib.pyplot as plt\nfrom sklearn.metrics import confusion_matrix\n\ndef plot_global_confusion_matrix(labels, preds):\n\n    y_true_flat = labels.flatten()\n    y_pred_flat = preds.flatten()\n    \n    classes = ['Negative', 'Neutral', 'Positive']\n    cm = confusion_matrix(y_true_flat, y_pred_flat)\n    \n    plt.figure(figsize=(8, 6))\n    # استخدام اللون الأزرق 'Blues' كما طلبتِ\n    sns.heatmap(cm, annot=True, fmt='d', cmap='Blues',\n                xticklabels=classes, yticklabels=classes)\n    \n    plt.title('Confusion Matrix)', fontsize=14, fontweight='bold')\n    plt.ylabel('Actual ')\n    plt.xlabel('Predicted ')\n    plt.show()\n\nplot_global_confusion_matrix(all_labels, all_preds)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-04-16T12:38:45.316571Z","iopub.execute_input":"2026-04-16T12:38:45.316935Z","iopub.status.idle":"2026-04-16T12:38:45.484888Z","shell.execute_reply.started":"2026-04-16T12:38:45.316908Z","shell.execute_reply":"2026-04-16T12:38:45.484208Z"}},"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 800x600 with 2 Axes>","image/png":"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\n"},"metadata":{}}],"execution_count":16},{"cell_type":"code","source":"","metadata":{"trusted":true},"outputs":[],"execution_count":null}]}