# Training, Testing and Evaluating Machine Learning Models

Source: [https://docs.qualcomm.com/doc/80-63442-4/topic/training-testing-and-evaluating-machine-learning-models.html](https://docs.qualcomm.com/doc/80-63442-4/topic/training-testing-and-evaluating-machine-learning-models.html)

Training, evaluation, testing and accuracy

## Model training

Model training for deep learning includes splitting the dataset, tuning
                hyperparameters and performing batch normalization.

## Splitting the dataset

The data collected for training needs to be split into three different sets:
                training, validation and test.  
![](data:image/png;base64,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)

- Training – Up to 75 percent of the total dataset is used for training. The model
                    learns on the training set; in other words, the set is used to assign the
                    weights and biases that go into the model.
- Validation – Between 15 and 20 percent of the data is used while the model is
                    being trained, for evaluating initial accuracy, seeing how the model learns and
                    fine-tuning hyperparameters. The model sees validation data but does not use it
                    to learn weights and biases.
- Test – Between five and 10 percent of the data is used for final evaluation.
                    Having never seen this dataset, the model is free of any of its bias.

## Hyperparameter tuning

Hyperparameters can be imagined as settings for controlling the behavior of a
                training algorithm, as shown below.  
![](data:image/png;base64,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)

Based on human-adjustable hyperparameters, the algorithm learns parameters from the
                data during the training phase. They are set by the designer after theoretical
                deductions or adjusted automatically.

In the context of deep learning, examples of hyperparameters are:

- Learning rate
- Number of hidden units
- Convolution kernel width
- Regularization techniques

There are two common approaches to tuning hyperparameters, as depicted in the diagram below.
![](data:image/png;base64,UklGRtByAABXRUJQVlA4TMRyAAAvEclaAVWTw7ZtA0ncf+y7T/KPHyAiJqA/7PCkVTl8NIkPfaa1aEmZ0aF+wbQyf4MojWnNQUy8d6hL+B6tz9jAotJax8yeuLDaQjSWi0ryxsW/1Fv/nyRJjqTSiHK1ACBBghyyySGHXBLkkCBBghwS5JCg9r7IuQ93c3Pzv5mb+d/cIo3J91qKAtvv5YoR4JZ0cnRwAXIVm3umaRcgJXAVIIMrsJrgCBVKAK9UIcl+pYC/EKDZEKFoFwKuSemwN4VTg+b3ciaIFqFz79MVSAVWAFCjCyRoEiQosFgOz/asJpscKuWYo1iXwSSALqVCPSjSsowQRaYM8V7QUMBptrFtRXmh5VsCIWVYggUA/vCFz519WFzbllfos89eag++GIbDMAzDMAzD4fCdrf+yaNsK2ug8HS1JHHji1aYGEPOlNdu2S5ZjyZTdULCgoGBDwYaCGzYcuGGzadhwww2bjWDDhg0bNmzYUHTi//7TlxGZsWJ9XUXkgAxIB9qNyjnDfd45Bw/KgGw72pR9KYgcUBkQSHCBYoXSgTkVHA/kxAJqKhYG1FouCM5JuK5CjYqWCQsXEZ6Two8CCWSD5loojVg8SZECkgf6QZI5FCkYJBaTAesaC+a6hIoVkg+NFhSek2Ao5hi0HdjFhORAgaSyIUHSGSfSAo0Ji8iOsaFgGTIwbajr+nA4sLBLSZJsq+57vv++cfHgwcWDBwcPDg42NjqNJElySkgFW//Ag2PCQsGDggcXCp4Ja3r/ZdG2E7TSQUeLrY1tzSBiCPi+fgU4gcis/1j/sf5j/cf6j/Uf6z/Wf6z/WP+x/mP9x/qP9R/rP9Z/rP9Y/7H+Y/3H+o/1H+s/1n+s/1j/sf5j/cf6j/Uf6z/Wf6z/WP+x/mP9x/qP9R/rP9Z/rP9Y/7H+Y/3H+o/1H+s/1n/sjvws3M9mx9XWJgNv8zXDh7+N629n4Wpvmwyc8rW14cNf7/zrkxzNcOCMp54ZQPyN4bmZ6QBHOXvd+o/1H+s/1n+s/1j/sf5j/cf6j/Uf6z/Wf6z/WP9ZeeWMMEMEJj0zsjg7JezMDIFNzwwszk4/Af1Jth+23z/7O6dnRghxevbjHP4D9n5YJdAzE4uXOfxLwdb+g/Z8PMfTM+OED9n0a+/fCwbTirPTj+n2LwV7Kdk2GALJNEo4O30r0LP12r9nYvFewI8dXzB8xwQhJqwJCH/F+wusp2dmFWenP51jOL4Ih59SOT0zSvhIzhQ9252eGVY8o9D8TA6nZyYJ75FEzgMwq2hIiCCDhLPT74VyFtGzt4woqM8mFZPyvro5Qk3Xs+BNb5OKZ3TeM0l4j84zs4olhCKZ1emZ2QH11aGc0fQsKGTNlOInz/4VncY8piOc1fExpWIDoJlQDkwKAahLRpIpLmk5JiMca/mYUHFBoJ1QDvRr1S0mTuiWaCZEhLMGfEGm2BDoVOa1Z39CODBaoraFRKO4ZMLHNIRfmfAFleJYQns5Ct1AYwVJr8+8WtVIE+qCl3ouURCumfEOjeIsxKSkAfngHTMueBMCxb8y5AsCwseG+uwsieJXAsODSOqBUGcmHNMmNAUvTUoaEA9+8uxVY3127dmfkCeuGnNMPtgw5mMCxTsQw5IGtINL5vrsrzz7E+rEx+b8inpwbM4X9IkLCoysQ9IMtNbVDA2KkCVU1cuacol2cM2Gd6gTZwU2B5GEg39lwwXqxBdWfEE00FhXM79WhDRxrMDUOiTdQKgzc47pEgq7kBYqc4NmoLGuZnGtiMHENUuOCQfvWPIxaeJfQaxKGhAM1NbVrK4VMZf4lS0f0w0u2PIFZQIqTMuSBvQCM+tqZj48mkn85Nmr9nxBNdiw5x1ziXcgtiUNSAaX7LlgLvExAv+KYKBcYnutiImEZomddUhqgbzElsVMQrHEtqQBvUC1xP5aESOJd1A4JhlcQuE7RhKKJQglDYgFRkvMfXg0j5CWIEzIBdISBJ4ZSCiWoJQ0oBUolyAYFDGPeAeLj+kFl7B4ZiLxKwHWhFhwQQLFoIhhBPTJCIcviAXPXsXTZ2+ZR1yDYJU0IBW8A0CbGEZ8gcgXxIJfIeqz2jjiWIA4oRPAQgBYLIYRsG17aPwrUsE1VH1WnZ4ZRWxAEEsa0Am+QGUxjLgkgXU5CqEAFgJAnRhEwGIGuBMqwV9B1meLUcQ1AOaETCAVAsCdmEN8IcCeEAl+hc1iEnEswJ1QCWAhAOyJKcT3aKAx0SckgrP4LOYQ39IPgDwhEoArZ/EnhhDHghZMCAQXgTrDnphC/GUQaEz0CYkA1rffgokRRAVoxYQ+8E0dLWExhPgQp6AFEwoBcFKtJZMIJtXPqr/cCwh1eNyCCQWj6Gc0/4xkqMPj1kxIjKKfMf0zIg5cgLSmZCSB8fNzaybUgaut02dvRTCOfkZrDghp4LcBLZsYPiwc+RhCKMxCIjoHavrwiayEYrZwQhi4JumzlpWMNHr4MAtXjkkD3+jZUjrDh4VCi67Q9SAN14m0MtPb5GEXVczWFb4kC5xtMe8ZPRQUeH1C11xg4a1nZ5+lO7RRweR58uXbJsPnRPi+6U136PnzHVOBgr+evUN3OETgevKlmUDbCKBn/4fqMIaCyZvFTKAXQc8+pjoUIdwl88OqZsJb3kLo2XfVROfZjRA0s6nwljd/ji/SnWcLIJmlgfGWN1hHFtEeO9ncX0IQsrxF0bPP6U6DGQ2U99UrYfTsEsGhDBFDMl+hDVnefEvMJjpOOwoimTcIQhWNgnDhIrnxP0QkHhGFNlUL8WERurNmw+TzKDNFzsT4sAitoSYV00AJKkiMrKBFMB7Rg6mJFdRHFNM8CaqF07Or9GZqngS1jXBcuEhrpoLx6CIpmBpZQV0MBkn5kG0jop6dJTUVhwgnmbfIwFQ4LnxObKamyakGbSOmnj2iNVPj5FSDWlQ9e4XUTMU71YAKTGV5HgBhqANStP+KAkzF5MJF+mIDSWCOL1KEqRHVRKM+KpjGyXNqofVsg8JMTRTJnBnkzJyqRaSmKaZhcvZlITgXbhAaf4zMkYLcRgY5E692IuLisqORIpmd+Hp2hqwwE4IpuGT+9nq/JRGRT40kLUYhjRTJXFKhZ9coi1FIE0Uya5qemVYLaJd4TJEys+cmJb/hXficrryFgfJbZJkaPXt0kcS8xgypRmQi9CxNw/fhu/BNEOG3gpiOvMY8+S2yXdLD6flv/wjqp8IT8BVQhK8GvwETulKRmtnRvqxNiqbzJ+s84cFe+MoVRRARNSGxkm2OFIc0pkTOLnuq7Mla3pwiBaYS4fHgaSAblMRoogyHtLzm9PwJHV9BlpcNvwZekBcxTx3lMIFkogx1apY4h7Hesn4sbfIPoJ7CCE4ndCPORBnG1IwPHqPgXBlrNttlKRxqshGXvmG9PC41469dLzkXscGGu76IZK6JRlwaBzoSd8IbGPuG2x6nTDEWzk2IiTJMqRkf1xCjRfaTK8DgKIZNyJQONKR+xq/2Xx4th5J+1KNmhJSaXabmSY8RzsFoM4CFNaFgIkYTZehTE/a4dpHMmlZ24fsgGNKJqHYJSfNwfGPd+r2a1EwfMkbppAxMaMiTTI5GibSH1Eyf6B0nkeWG7E1oxZNSPsyvrV83gygdvIWUxqpFem1JKZ6U9mF9ul8xqvSMX+1TJLF+EyDWL+GvB0TEAlJa6su+dNGOeJl5YREnJME7tRcEKGW1GyrxsD3EjHJa/CQrWBHSpnbrpQd4KeKEm7FMHC1Fii0hEBTEKInBtFuL6kLvelbaJOumB3hp4THikaHgwAJasQYZLOKMwbOWQxPfs5i579ZHD/BSg0hgxRVKEC/OaIX/slJYxPEt3dHklFleVpRlmSp8z0rXO48xN6kan0OkbCXr8n6GyuNZG+o3SRnGTOuJlDSL+0OGwvTpgKPodwQVHn3qMDYUZPKsRzcU65R8zMp0p+biEs6Ubp3To5kUgvXCkkRUGiKZQRlZM+e5YVGleAi90lq3HMYmjTjEB0TiDWUTKK1yAmMNCya4BgzNumaOd5p4gB1NI0YpBf72RZdybpX9VIMh63i4uM68sIwj2hzvdL1siUPqF9yyXw4ZtjQkW65L+3MtqcALTxB5tBzijsayGWraRBrNHuClqDq1PgqJUGXfcvtAHUw19kc4beJ1e4uQaVPXVxHcqMmAZB3s9VmqN4g5rKNFJotpwXIGzUSb4x003vAoUKgFDy7dai21ZJdNoxHKmMKMDx4KbLNjpqBiyc41pzKIKUsQKc64xVC9PzCn/URVBCCgR20JBlPO+a35fH7m89aRlPDDZhj80DLHSZ1lZVlm2TJZ1x1T1Bi/DKUO7bQzEUu2nkQoy7JQafwyiQbnstQog6ZJ9GUiA2YaMsOqPOTL+WlRPLJVM7Vsl23YZLEI2TRtqaXaZZOyKgz7k3XaMUVVJKGciVt0LG/8Mmku7S+d3CbSxouIDG2E16lxBk1mdpFIUFpEhNPAj0xw2/tkWjfTSLu+Jsk0kaThBX5ydsjMufZ8PGEYn9DESmMNOeeiA9FAzV7W+hMkUJ468oudyvmmZbbUZtCcILeDyqs4E5eKx+KDuQxmztxqlT7InP4UnjTTtCA5q0jPKmQYxut5o+R4A56HQzraTDrRuGvYGGQH86yiviEpGCYZNL7ALCXz9Q/FENis7DSpMIo7vgo+F2nm0bVW6IPCbOu+DwVstiNmcrZkVlQT2c+aUs6nWqerO7vtwlF2PPRo0C7jrTvTOHWVucaP4gaDGg5ht8cYVenDIKkT2EmBsZa34KeimHmq1iuIq87sbNG6KbJCmKSK0/Eq20ry+fhi0lKzaYsMOvTn1lkMQspwvI1n663gzmMwu7Y0Au6KqK+ZyNBcK2dWxZ3FbEAX22qgSSq4Az4U9cz80rWLOMRCH1iVUWpQFULNMDlbVyZ9BdFKtTLpF6y2m5asu4wyfKRCod95FiSpdVGMEhyFR3djwVYXFUyAVA+fC+ZKbj8Eg/X4JfxIlDMg4gxGhBvaJ/UaRIVQS+WgDCtFYpQBwZ8/qXzZcr2lln3pLZWLCs8/0c5CCRQ+wn+GWVRHH6YTbPeJt4IaZZmnO0lNP5HLeBEZRtMcC86RoibgExpOTBVp2qKG8cdDKUZZPDxfJ7FYORGjBPkSz1uIuOOwxDjATVJQejWaI5YzTmyHzwI6vAESOGViah4E5Jo9gB+IccSjV25Zm1xz9goBx3a/dT2Rk7QVmc+uzyzFpKWuT9f66yAOZIWjDNn36FcV2hMnnYQhyue4ANlX0RxDVe6LZTriuUAP6cIkAVnqrr+ziZhf2rh6sYXtkM2QJlPreiInKX7m86RQ+POXGPTpWpGse1ztH4ySBE8XCeJOc/Cf4fhSkboojhrjjFW1ELwR1Ld6AyZIhgtn5tfeY0Quam4RIS06PrNx9YZRZf2lT8yyLTmHWsN6SDBHU3/+EkX58PWOq/23L+RJg6SFBN/UuHrrRqegAWkspPRaHclNcfaKUjEJ5gvkTBrpsvdJ3QwlWfblpTYR8iIQHl3a2Lh663Nl8qlGYIGSszzEOHZcCErkkbkx+UEkU+b1jWFciKdJks+Qo+cwXFKEVzaU4dotKUgR8Sk2WP8llhEcS6SM01i5OIjv85a+XhNJjoefKcopoCySXqrSZ0+4sQzVOvDNYOTv9XFSsxO80eqywxoK5dYr1kEsX6oMEU8gzCXdiRWWjrCiJmw3hTZLZV83qbCOgV2kxiSbFoz2n3rJ3pkxT8cFVGHNX3jInBtUfvvGc6jlFC5fll+q0GcTtLzyKef8BNkQ4ZaxPisY5ctLjEooWhtGsy0fvl6xDvooyXKQHJodpP35aag7sYJPYB+wtyOsDq2aQhvTRiYnQHE0dRydkfShxyHAfWC0g6PbgR1Ozx+86j2BWRnz0JtrRTmFy3KQ3aVdVmiHSMb67MyGF36uzN+zt6nkMO0wIJQPX49o6fCBWL5EeQ7TVraq/OhlrMDjGG2h/gnQ5wm00mOKI2DKEZp2TRo5XbPrsUQ7ZfVCUI/7LovI+e0X9h/KcW+98YTmfC6Urg9c6k9QiGHhLRt99iVIYmHJX2PaYUCwupSsTzwtGGU6PEYw61PDr8GJxzGaZ7VH4IhTec7PJ0VmM7wkWxORkaUwRjzVA2unULw4kJGSnnhwR4TbcxDjKd2gmO6vpBpKjPUZ9Lt2iqaBnmST9NkN+5JOS5Ty4esN17YZPk2Kde47rLzqB/75VMGP0OwysjY2NnR61G8mDDKOypRrUqg8Dvk6iN863k6xH2BNp7lQbX4axFz2xwUmyBjqM8k3/xqtFMJblsd9NxTTKUbKOOd4fcE89nFgnybA++OZhK0F2xPuACh94AM0u3ws5TV3UR/oVcOXtZwfinmVRaBVnVIpXIYxAaT1p/2W4EoVBH22j7MdiaBZT4h9Za54mnx/TB2OcdoLwT0xgJs+wyoYhU1UNjKXMXYby2IlwV4kLwsmrT/tN5GKxdvpM6mYZ4NWPjzKJoG+t8STQMWmtqE1fYSDLssLjNtw+gybaGyEtXH4F8kWQoKM1QYUyBtgo8/kHsxKsJJ1ryqu4X1lQ23w5GLUi+62BpaLBDFWxhOMsooF0DAWOr/9ps9QWwDYj8RwqXT9VPuiLTVoP5HxUSYXQpUCBVr58DfmxhnkF86c/YErumXSsMb3FrZ2M+EqiG3v2i+d2GdC77d3la8/1nwL92X9yItSqgk1PDDB+8gfBAgh3ALtwO0xMre6MOX4lWg2IpnKy7Bcb/n8jD5WuguxHppgucUVmF6rw7WDtIi6WHCuK0E4xdspMvH/fhsGIclT437kPgLBw9EYr3z4g/lpwwxyadGjaxcjVxbQIAlLvrcM2dpF8o0LMLaN8+AtOFi6Qf1qByonkVYVVoGEOBKLm7TEkHvVeOx3FjDNFxWc8w0rXsBr2J6TCcJJRLBI8nUb+U204aXIFFmsaN1wYeN6y9VjH8amtoGi6cK0TZ9g/LLTKPw7aAv6wGPJSAe0n3ZmkMHh0ErLfNlhkExcYxqqte1PGvzTfpbpOp/kdIGI+Xz+259HqGMB55qJtqYjU9dbfuC6NJdiO8LAZWKYUbfbGou8qrj0kSHn/DWMwKZctKi9LQkwxdFpuI1aDdoDDx/0S9+Cc36oWBKI0lD/X3r6b38erY7FBMn1ls9/25sDsZ1gkBKgsUFb1/Pcamz3ZbYSJNN7r575PPpaHpuciioxyBT26zvLQ7Sq8qGhWttDrT4KU3HG07dYP/23L0amI3Nm5yK54r/vixv6ubb/AjAz7C8nUkOoZDc9Fa/oPMfADc704Xw+f+VihEUhCiu1qpmaQbBf374aKzuVEzy212wM8WrpxaFoBdZPnz9Va+Ni9EkD0lq1restTpq7AOY6hwacf2bseha/dRuMHWfYR8nOzyRfwE7PL5yNrq6nFT/tRYtqoViAjNMOlGv8GMFY8ZJzmKFkn6KLsbIGb3tgVHyLDsrMfH58Nepkyfku0AKWQ1+am3tznUQDnp4/t8+cG/ie9fCFuRg7EHcwbj7YRNGcidPz+aNbERW1YTbUFCUZ4dI2vlaMVM/QIbTw7o4tDx6iOYqtblySbflw3LpI8JT9/NLn0WZcKhWltB5iONdpDgpFauO27u9tgpkfI68+UH3AHWlxafkzVBz8RzXFDVrZIM0Y3is8rEbdd6hc48JZF6n40j4/6Q2cSYby5SkcaPNhnb/wPTYuRp1xfYx84sepufpNOkvk/MGrb0wUj/c/uz2HOXCdiQfWbSJ1hfKcwOXL0dSLbOrXbxCiFpzDacdSCIWzbTTkA3ueQ0nNJqn0apxquPtIlqnk9kyizhzfii5NZINdF6cs4ROabTZ5gqwVRLgDwnPA96zgxgbo1u82xro6KXRNALx8OaLKWi88MvN80ca+swTudK4T6bsW3eMwSVnVIyMYkwarnkgc77WGKKS4lqlU+uzMxYiSPk72bpKq7NN2phkz37M61/CqlcZfmij8iCrSXNct4WGC3QitQnYshWBuqTYDtldAESGsPK7a9mQEzErC+CSwHta4lqk0+uxqREiL2mzXV7TOqkPNGPme1cmGL+UOgzI2eqVSqurIiIyqL3UhjkU/xgTogDC4zVAz6oC0egg0xUOR2EEqG75rNzRArtsK/FE+XHUGNQqOXGIdfiQBcTNdMm3P6myzUuNAk8qoV+keKrwl1UVGUFilE4axcYU+2iI8nWflwFKtrqcaJ/1U2YlZmK3AXqDVE+mH8uG1uuVuUXBkH63OnKn8sMOGjqfxucxgm0iTZb+ABqCU9p+irMirVieUGjZ09SFpuFOMtuWx32JJI6RlP8uypmRWt1UW7e8IwyaDb1Loq0XyQflwl7Kq8yFaEvnbF6NGSo8Gr4VSgjXi8Ny+uUHML4VKjbgOI3/geuUY9veAdXeVFotE6UCftF+ccBuPKimp9qFOI8h22V4Vdtmw/9ZUUoNIlxlUi+SH8uHT1jbuNpK82A3PI5iJ4M4acpi/+hgxM4i5f0co/EjrMPLzs8fz+RwwtShAoPCHJi2bLOsr+26xrdprbV265IE8o26b42eSXTa84QZMzA6zxKi+cHWReLTy4Qh2GOQEcqsiN/TiUhE98FHUdL1GHLwZgzaR8g0xd/qy3WHkD1xRz+sfgYtIdQ5CGffAmVEGzb7UhunOVoYSWGU74WZleHjjVWiXrRWVIoFO2VvGSfV9iOu+lJ7lh0q7mojlw1HsMLRWCR5mTYbfeqqqYKnjC9aIA5h5Ttkm0sfIjVcfQMuftv9NRDhvdvWVYxhzGcV916RhbQaNnG96uTMeJM3nd174cl+Ez164I9llQ5f9jpT3bHA6fl9xpTJalOWQ+cIOw+XLoiSAjqVUsGacUbeJVKh7OIeg4s9cjGbO0Z2ZyzGXFe679i0zzaGffLu7MC0x+fK5uYjtjApB2XoX1beFncVl0VcQE1AOCttBEhtTsm5qWz4cNzdtk3/ScVW+ZAIKQ8VHMM+6eO23j1UGVxXecqD4IeRiYH3tMfzysyc0n4PHCZ1S6alb7wJjWnw+Xn06/oI6OYtePtwksk71lx22+lk5J51X5SOagIpYnvX5td9+pMpeuayvYj6zl7sR3FZI3llVgiq0+nz8G2/J1US+oTodjz7sK/p63dSfP8X/Rq1/lkbE4TCUWWsGM3UfET3r1tkzx3PVek07Lx/lutp9WfKQdj4ehNt3npsbno7HO/T6Ui4fPs3iBDSuNCtZkuUz1UnUKPhZt6698nVN5JfdWRvij8PIR9HGsy7eOvvbj+bKZZfNq+oqrS8igbpOxIS68/Hgpae1p+P9Uz58An4WPirLfmj1HTeubpyBylKwwMvEur1Wx+g/0gvXIotju41Lx+oCcZcvWxQURtB4DZiGPFWnaKlZq0/8mVhnAnPaPu7zybNGDbe8W2eFrlQoS2ijDdVWdkQQooic9Ru3Nl5R67PTCD/c29YGdqpabrnmWkHPPHhhoi0fPnlB2Kv03/b1uozPb93a2Dhz6YL6iqbTgM/AAsQ220UJYX7pYjSQANvVdUfRM+Tf7We2w0KqCXUNoWfm89uf7T9UmKt8Qfjz58/tK3QVF2954erGxsZvX7p0PLcJsFUYNV6d+9FCWKOvYp4C6+u6LkM942ziHXbNV+4aycLOGiQAs2l+12etRvpDnBfObuCHE4jkn10j+JbGhn/DP3iLq/2DOrPbWjljfQqtFtJ+U9/HiYXPdHEBRPrfGy0Pb/G1jfAetN//m4EJznTq66y0U672tsscFsG2YDgxv9eJrQ+nfO0ZrtV6fXApMFsb9C4ObP/ojKPlodhXJcs8rZ0ipi4lYfqVHzjjqWfhHi04meJ0g6nPeZ2bZQ4i3J5QGZZ9jcch6bLyBRzl7O+v6+OG9hZJYthVyjSFKilNPQ5pXEVeQEho1Qjd6U5UqKq4NPE4ZBpXFemhyixdJF+mMK4DUnvhh9CB+X+1F6IyvgN1XU/MnqC+8aySG7PtgZCnzyZ1XY8jN6YuGLcVOmMxFq4Tknkh6cSnSzIvRA3FrMWlzrIsNv71L7N+6YU+bIJBuyDk6bM4y7I6emvPUOany2/NXW+JC87h2BH502dudmhu3owYGI5PBOamB0LoQNy82aG5GbnhhsyJn45HzVxvqZYly+M6Ohf3j/evDS0y9pmtxj7DmXS6iCvtbQ2cumo9nX9/vDkOZhYJ7DkZX/6Zw8S0vYXWZZ75cnBqS/h4MgwwUIhou1YI+/1P6+bmSCvaPzd/GSzJOPXzy4AgGS+LOd/dBZfq+2zm08fjkR5ynzqXk5eDVEv6LK4CHwiGHaQS4sPCl//8179LK4+E+PqP5Oy4VKKgV9Sc74rg07umcfEboRFvq+fvZAp9tqBWuFRi19d/KA25PIlKnaWOWJFxvitTkhJKpT7r0yoSoTAVKnNJSIg1+iwhVUw1DAkJQw0lpSIWClOpMjMywkKrz2JCxVBL6ogILtUypFMshMLUqMyGiNAY6LOMSuFSgC5fnYQwMVFnqSNS9Dnf1VOSEEojfdbQKBKhMA1UZu03hmEYVIyyK5/XBXXuLw312cSHv3C5kgX1TJP9CmnI0PmN/V7N7PeKZr/XAgwNKf33C5csKNIzrfArpL7wJfkgM9ZnSwLFUIPasi3xIEmNGTryROaNpkyJB42FPsuoEy5VYfAOpIPfsFFnaUKcaDjfNacgHXzXSp81tIkJ56ogdKfGFeuacPAkoZ7FpIlS10CP3ZzVIXVkg+tPJ9SzgjQRq0PDORcaSQgHv0G3gnNudPosyonWblfOqK81JNp5oAHM7s+P6JRxRD31qekMop761LQe0U+d+/qC6KfOfes/p/zH+s8p/0EsPyrICg+s/1j4ISs1zD4wuGqY/cEhWfFC5IHB1QqR7++eQotP7HICEW9a5fje6kmI+O4aqsLflpxAxAODi/snIWKLrPAFg4sflmSFNw0u9q5QlfrUPr7xTQ1Wn5rGgqrUuf/Gm+ZWnftXyAof8zyBiI+uNKxI6n5ZMPOwnGbfXENVeNuoIlkUzKy66/ePQROTFB4YVCQNM+tv9n0ZB4e233QH7/3SFf2sUsaq6BbvfdfVu1alTHXXee+brqxGfLd/z5wik2K0fxAOXCbBoC99aC8FW699aIeqVSNzvcS4u9ypMqRo6N5eNjOyPzprSjEpWIop+tkyjhdZmcp/JBd1AOiXYL8c2na7Jf4D75u6VR9t4RncvahwZgqXG+dtX6B65g1DimXKYK6onVKYhqE0jd8ZD+GeeRvaixcviP+APChRHVXj10zu7mbF0ZZECN3bRECNaD7maEZRiySXUWGlRJpp/E3fEHMv6EYelGzkB/8jki2bu5tJaYyLX8e7mzwE7Vo8L8uIogbJrXRhfBZv1xv8zMb7yBzLQUnZqov2EB6RvGBzt/e9wugbv94yu3sZsfxKI4pJKuaLid3fsEzmX4rIEQnr3vnQSg7O45bZ3eu1woY+POp+wexu3yB5+JYJhRuC0ZnLlYEh9isVeUSSYO9cSA6+YxJvB+7eqerRJpG7ieNuFPP9TQpFH47Wgg2dP5nIXXOivXOlJoqIu5O8TaOmhbVNcnd3SH5ljzzVuR9zbn/LmKZCtL4vaZeQpMOoInYQ3u6URAnh7grFz3z6ee+RJ7+MSsES5Xxe4kfqcN+c/JhEQbQLiLt7BTGBeLtBcNB9AkXMOcb9W1wKGh8yE/vm5HvnSWZwPf4EOCZREAuMuwsEN2WWQNGAbCeUDCxOnf/YANGphwXI3b2aphDuxu/kXfqEA7lOWDlYtf9ooHbOo2roobx9UNbj6H/4HH2iFkyQbiE1T83he2IzCBvVUIK5e1QMLZi3D/htlNEBeaJhrD8hl7546DsKuJ2zamjAqBTDLgTG3TN2ryuGQFF49FGIQR6W31jgds6tWpjgvF0qhg0cO+yWC0mgAOWS8ARcyg4c+YVw3wzEJDiBxQgh3V5njxFPKjgWxdDBUWO3tft2HNUCLz3LAUkEMV4mViaRPh5OEy+0nAmQXmhOQzuj/2YYsUiEmGN0xpEa0N2KoRH0LbnrC8Jw9cifeT1s+KFFfne4uLh4XxHw07MsCxc9QzmxoMKhVFJmIGh+/XRz7DGK4eUcAKm5JM5e1EwdaMp4dhofIdYYnZ6FACcuetYWID8BAfYgv7yQmsQBevsDHLCJ+w0yvZh6xJIfDPthGJ4Ite1QoyLyBuoE07MQYeSBHivBNUgyiQvGPVviuOiZVuGm/tc+7GkWCy1CHOGSODNRzQWKIT5CjDGKAJu4KB0gD6E9yC0vpCNx0IAm7jeILQBTvGvPfj88Adt2eG9UYiVOJD2LkBIPKjhxHPRsNYZzVApHaB48I7wxaA9oG6hcf0iJB9WbOB56hnMc4lOGAp+9bNQWT8FtZWoC25cJbJJ9SgY07bagPNytqNf6C7piy43YSHNBOSAF59wE1jLkAW7n7NRSMuSo50uGdHGciLX2DrxAnXx37EdttYegvNL9SVZKf6gp6ZPmqfQbM71ejzyxBAvaXQ2ztWoowJi1cA2zSQaWKRprzne/dvitJT3qhEuRHi6D7YXILDE4HoAq+ZCkwHgrRJy6BwsUi77a/H4k5vG+tm2pWo3xlqrhujmzixTGsL4LC77eOtrEBMeb2VoI16/8RxtOZN+aWX4vhnq1gl6M6le7khnGTLM4dlUcL5pUmhlOoq9WEemvFu+nYhoiPd53erzF+5MhnCtrtTSx1FzztCZNTATpBCGxlvm4VUSJr5hMRUyAV0WkoAXjeckT2KUXRWxQxnsK+7agpl0tZ2z83HLGEYDeqQi3UfWVMy7J3V06NJKkILmVWfV/yjJQDMMwKBllV4AxDMOgeXCF7dD4u3XVvaKvrtp1ia+uelL21VUjU8rS2DiZBqWaBIr9Xp5kxnYZ6aEeeBiaLnRTMC782wMPfcIeeCicunvgQbnDFpdj1cASAPbHfoSJasIWeevx0P+9NNMn6qWZwqmul2a26u2lmZG9l2bWuBxjuUik7dAnQ1gVKRjWwezJq12CnryqncJwrvAM7t4G7lZe/w2NB9aevPK+RORYDcFlsxjlm5JIGnR+JIWHC5V2dPWUoTRZ5W/c3BG9vRcxWm/vTU5tuF0T7+71WoVHJO3G+zWLuxtclZGNZDUNp73mkTQoLcdtOM1EaArZtG26rAzxV2TYI6B088SuWXW4tg5m4rztu8mpz8bOM7i7CEJxOeXUWQtXgkkkDTLL+meNrf8cStG4KgA4N5WebmXvnAJxrq0OnmpNMTk12rxpPNWWCl1zKNlOGZL5tT5xonJZGuyeRXdTXXbvBEZ2sbkkEeOopK9Jbxf15FRqY1WQ7t70+Lx42EaaYkx4SJZQllgqWB3+Xj+pKlN8b9RLmuQUJbgBLx7QsrAmxhNVdf/b/7z065kny9dBeCr9jelvLkU07v3iBNTPiOYf8cKA4t2eOhANFngNrSo556n5xHtPj+o9/byrgoA4PGi1Wso8Ck1cnxp6r1pGEWzxc5AquUWrZrhBNJekr+rc10ecoxu0JkEVkyi2vPFkQ1w5+RDPCG8kf8y7uu66rq53sxnAu0SEBvdlqVyYa3pdPstmxhs6acn0Agu1NERMssmEuYir+Nh7X055P3veSEJYgobnoVXN349HJo1jIQKKFge0U2g6jrepCA7UCoLW9ZbLeGRySyLJ1mE1Y+Znrrom+vfTsm/1Gh95Iw2hKjjnAumsXs5OFvTeJ2lacLanJ84oo/391JdabbHn6AgLrE69SDnnaSUJ+iA00eUnZHimSAMGKX8/Je8os1abUhESQYk0rVuMkFdmxRjYZ3XaMtraY8rfTyPNftd20z8mGzXMVtWfoEycmJ4IKgNSmcPHhL1esA1fkM1rmB0Zr5KGmClbAD3TRVM91MELEYy7Z4pwOymfCExIZQ4/AFzP8NLKvBUisr9+qgRXlJdYz7QQVygJLgVNy5tal3Ha3c05DhR8ewGjzq5wesOQGPQXW95IR6jFuJtzgyJaoODcUzxz5qZdEv8FtdRfXKEWYJyIA4PlKKRLAwXn3gTbqJuxr7vA0nc9Dh3AX1DrpOihKSEBvBKktUrTF/SBUE8GoNy3BGjAutxNzcybJaWXquV6kL+gzpqLK3QCxP5lhh2Zki4t/t43CvDiLE3w7pXUXt4IdoWbS1pm2gamh8f0Fu97IyEhY3AWrpG6Ni027cGCwomzelymnwKMQt5wvwpy0S95nfNfUCetxTlaQl9SSwuGM8NsoohZlJIwjZM83k8/CxiNvBHqu2eTonIZGyBKncVH3khJKOWq9OOhorXmZdnPyjJV9GDBQlE5f+ppAe+AxaxY+iYlXefD+y+orcbiHEEhk5prLvdaga4HC6bQR5G4LTBB3hMUtdJ7n+ruFSvgzilLhAaa0hMUpZW0XVjA3lxcVW2ZnVIZwxp3k53MJToNGKW24r23ZSK5+tTwiqbzm5ShEP3iFEiGgZUK8t7CSbs+NUG/ezYVHILfwp0lkZ7poiY76rk69/EarA5xnfXLsskWsWs/Hu/XskIGXQCf6m+FwiTEjY/TPVPNB1pfhugIdurZAVIplB6mECJJHB1PGuvCPwwTMo8UBWgHKU3I9Sky/O7Z9BqbnxWC0agA9Z4IIjnvmKWeMYvQAxVCdNDYvGGgUKF5vD9pEx7vQ54ro04OkVWR/GFtzVYwmidMsRqK1oCzeumnA/uEeadOZrjqVLQ1140UaqTOTXOpR9MGqMCo1UkFxiKFs+xbXdXMfQMDpDRWgeTybNwGmMGYU4Yc7/X2LC4TXMobPfVf0BSqDCxGWV5UBqT0yAPmZ3XSUEjxDeeCo9NRvOyNRgruLsM7BWd/Qz7itsAExKRQIIvKFJgS7P3CSlP9F2YK75+HdwrO/qYOlZURKZ1AvAuiTkYp0MPdZG8d9V9QFd4713tcwsH6LlclbYIRhFGhTFKghSp34KCjfoVUjWZJDbPVXq/3yV3LAbo6W1cBYhiGIeGnzEEOOqRdw6zIiN4MEQj6ZHqma36FVGGWtELkbe8ykd5LIK01NHyBK4FQ/SpQ7PdD0vNlkp45LG/2e0kzB0AUO5BQz/RNXfs0hZ/2DjyVtc/AC6O+SWSdSk0mChwQN2IywYWrKhWXMvhezeSUEG3zGprC+0KfgQHMpJluph5CeZoqeADciMkXCW5JpJLh98eyFN1b8NLpJ65EheD1HICD3u/fhZE8XUirTuomZSjMomovuLbwYeD2BeMtiZQ8PEuGTEXTjHj2isShTYqueQ1NYa93IGutDz5llp27TstyyMzS3ONmUrUbnKsav15TA6i3JFK11GAfuW9S0ZVICbuJRDvx3tOjIkMwew7AQa/30gmr+nOWUUlw9/xeFUjAOgSU3vuNvpidehnBKJz4A7MDtmS/2egnXvZGmsIVpT478Gb2X5TWp1jd/fMf9AILwEwT07N4TTinNpJcOZtolYeIdWFG0XvKrJ/Y8kaawkdAn2lmfv/866wILz6VvtTr9YIKUHXvm86T1m3CQw0lIOyFvC4O0gS6eMfI9cx2rdNQXI8QQek5APoZb9GHX99z/vz5Dz8UMhwcHAQWOJsnsloK9cG522rqdOwT6ujfT+l/Qm165zQUe95IU3gX6DP9DFwkLfMECy6gphRELYRoSWlvr/tgrm5d+sFpabZRIkg+B+BACkKfrYVwi1S+ezZtX9LvsXeYXi3FFcVITOpTe++cvT4zAif3fnEEKJaxcOmycVeHNs2geqZxGuxIS+rcf9nTZ35AYFMi6fmi4Rxa3dxIiLepCj+SnjpbM+I2Ccpl3GaAEas1QNIUrq8pcRvPGrhd6+jxfU9h0hT2PH0WqklHuYxr9grmtHNYuCRvCFtb4saDJwtmJJ8qie+ejca6vHyawvW1J871S6QMIuIxSgVz2pktT2HSFN709FmoRDL3kMpHHvPNZtbSjVGnKjwjnh61RsW5uS/IG413KOtJXxdVQlO44umzkIm8TE+HhUxyhTCixXvnwif5QdjaGH0dFjbJE8LCJrlaWOgkRwgLneRyYaSkzv37anUWGuvcz4J6pmPqUaMkfhldxyRE+WWUBfVMm9SjFlbgZ3kBe54+Cylw1LPcjPfeDi/wtGwIDsNCClz1LLfhVwqgTcS1AsYkEHxfVAv5frrsdOZqrVjelqqpyN0ZTs+utwjDTSBQRw6f+doakXqWqfnZDGgT/jq2gp/2ZEzD/wm5FdVCbtNlfznzL4tc7Y/F8rZUTUXu/n5m07OXW6TPXjbchpD/o44c/km+dopIPcvW/Oyzgtq9+0//q1WwpS7ClZhGcpUm8/iLFzlagFjelqqpx92ZAuSvSxgSaRFGGxFBFTn8lquc/fknT3/2ARr1LGvzs+8KaK/d+5mnMVsFQyIovCKiRXklPQb4ize5GYFY3paqqcbdGQPUb+v+uRZisBUU1JDDQ/47Pz0LwaKe4YXX7n0/4ylMmoJw8opk8dq9HzV/k6bw2r0fHnrqjBYBHUYpQI0jRwCHtVGGQc0Mg6IZBtzxvRlPn7VPhkG2ZD890xrA61DaFkyJCuzqFakilLflakK5G2u8BgyJ+BORc7hUyYR6pjO43lZ55SqwV+SKSN6WrInkbqwhDIn4FoFzuFzJgnqmL3gN2G5rTyTZVGRL5jPE8d0ZT5/5EpH1TLJkQD3DOZBh1ALMOJLFa/fuXycYoBkSIUkAlt1GLkA0JEKQgCu7jV6AZ0iEIgFUdhvJACmOKAEWRjBAiyNJQIVRDLDiaBIgYUQDlDiyBGgYsbCVjRpHjmxlg4YRi/rUUOOizfrUtATqOFpR5z5uHDWqcx+5HjVS4ZcRcj1qxMgvI4x61KgGGPWoUS1+Rjiw9j2ENIFQ7JJsgHDdLGHC9hCScmBfBoAyYe96MU3hOnHC2gQSReG1e6LPrqgWP6Md2JUBoE7YHkJSDmxNq1EtztEUXrt3y1NnxAnLt6Ap/Jw+gVLKgGjUp4YyJUv1qaFMiUad+yhTslTnPs5ymuGXEc6UKvllhLQWkuG7I9JaIlDfHbUCPyMrnKNQ2GSqUw9svA+hWvxXssLPCRf/nqxwjnDxN2kKr90TXapPtvA0Jk3h75+CjkjVMiSuiFYtQ2oHotXaQ3AFeao9xM7fUHpSw+xr9wDrlDXMhiV7JesxXoiEpSO8CNULEa3AR2SFl2kUFoaQaAq3KBc/X68hHNVcTba4RVZ4k0RhcxBJPrCsUe1AHagUW/QDm4mY66mDF0Gj2gDyc7LSBpDrPTH3wUvnP339ReYXX3/+/IcggkJxy1sJTeHv93q9/U8ZhDxnEe4+/6GYWecjJLX5k6jw2r3vzJy/y5yrgxdxwu57M6QIC8O29ITX7t37Nj4OcI6lIEvx+u97kWQIq5dQjvrUjL+ZH9qngBnGDKdZFi+zrExZEuW8F0mE6lOzjCIcde6bfjE/EF8FmOga9icqIZdNCmc+/aB3QIPq3LeNIiZ+GX3PpCDFNay1Yros5Xw35xe9gQT5ZWQX9Td7B2TDd0eLkbnvTARNSsVAh3x3NG4dEeHA7BViKadLU1EzMDy/fke46gFwQsIvwDgxl7UGwy97B+QHlDs2RER47d5vGXm0HT7pHRAgEJ5DMzCJSwWxnbSZkOx1SoS53xBkhP8uklmZrbilkO3D3gEVwq6qWZKB2ZSHlS0JmBAh7K0hURC+ZUQiq7aXtxHS/X7vgAxh1xWOkadlSAzxVYBTnMlT6Z+FCFmGtK1YjWDUHqKvNkQksRYYAk855xO2d0B/ag/BWQG9qGHWbM0JBkvOwYppUA2z9redKdr0QgRD/AnnXFQYOMF+74ASeSFi5QA/8aD06ONIXHDO52kU972V0w5SBkuwFnxKhMCxmU06AHayccg8XidImPawykl2wFv5yXZAqkbkJDpgrvzHep5yFTFNBurKCf+K2sCy24iFD2kwcFf+YzmWIcnHfkb6cYWfYn711Y8wv/+H5ENaC+whqgRNtYc4vNpD1iYEfG5892XrPmpN+VGlPw5nNBfIK48q61NDZw2zUUJ9asHc+2HIS8GiFj70bAxqf9Vb9alhrzyarHMfma0QiRbq3L+6HV8jQrZRIx969se3V1qrzn30lUeRfhlhs6VqaxK/jIL8tHe9J0JeRI18yDfv316lQc/wDvzCf3zGm9mr29tnwRwlhPpQEPhuGvQa8XBY+I/LeLx/dfvH7/lwjsnCwGe3V5oL3JrVSEuriB5D0atX0QGv+XV0ji3wv9xe6TMQ2lxNV1rOuJszf+qtKBrg6va/UEbWQP/t2ytdBZeFk5XWVYNevbq73zs4iAC4uu3DMSHNeHulp+C0AKLSAw9yp1497z1rzc/V7R83IUmH17Qblt0FSlN6aaaYNGIGdtfM2p9XiTHx0N9eaSn4LCBqBI09eVW4qpoUnghwCJNcUU2mUxgWTQW3BVAHTAe5t/eyiYqJ3Nt7U5hp1QfDp2ESsr56ap31Up0CDEEBkToKfguIFm1l46xHQBdxrO4R0EzXXTPnewctRdwcLluic3/cv1q+Flr56vtEAPyGnUoWGIjCkETWM7zDcQGRYn1qmOo14iHnu0a9RlxMdJe7gVpEWonAOVyykHPfLj3dyj6ckd2GfXX7h+EIQnN7JbSeoZ5W2cymDHXumx/we5H//Sthvrur7Vn0YW1iI/LF3kEIrHM/nFu8X9PN++XbwUEG+IadPt4lgPpRaD3DPq3wSIRK+GVk4pHILpxKL0+b2LAekf3eQejzyyh4i+c9ooZWuoUzr423V6Abdiop4XhX5J2XVgCxzdU0gz8BNatlZTlkLsppFhsJnqTCQaPwB6XG+vh6N9a+CYqZlhyvwfFfNBpYPVpEIjBzWm1qJLlJ1Eu9g5BHtMZ6xirrX729khvhCMRrepFb3nMIBv9OMLHHicmnYY9w3G5fJKqyPt/B+F5Tkwy+CkLhYIjeFwm2D3oHoY5kFbRuyfpZcwxaaw2JKmD7uMZgAh6GO8IKWgE+HpbHgNLP1tSCv4HWp1fVkHM+H+L480++3wCwBJKXQ8Dd7m3hYCKwAl+GKCkyAD8LEcvbUjXluDsgc+iZub4ZVig0Hq/bYrJdUVSRw2/5ytnL7H69hfh8mLQ2bOCXKU/PsjYi2Mu9AwSMgoRPA0mqzIOr/bFY3paqKcjdmQKb8AuPKV7L81+0xSiQqCOH81QzIgziMGaU14YNWg6SAvUsW/OLHT8LJ4J5b2WNobiZ4LRfQ8g6XXb697naqVjelqqpyN2/MpCJ9Kz0yHBYcp6zHXs226WOHM5VzTognklsw/6IMhyfkKpQzxy27Bnx9KjegSVXLA49y8qvYQjMpcvaU672tljelqqJ5W5MGSIxAjbbpY4czlPMvgdVP2vTymvDHuHqUxNcz3SNvdlSOkZo/+iMowVkQNMQKMNOOYpZDXYg0zt5GVid+4WeFNcrvYMwwN8YnpsZD/ATs++AUUiMCoqPIyoSlAURC/o4LEIecGf/LDLTcqC2O3a6Ul33SAV9PH2TiRWFNc5O/y5gMUKtk5fVMEzaUl69FdIAELOc8DKxvlaGN3aATDIT8/BhgKm25D6lIEbrCkcn5PtmE954Swm4CaCw72bWmPr6cu+ATFClWAtqUDX/ld5BGEZyQQnZOY3Je95LyARTjxTr8bC6v8bHFWEzSpI0GqDXm/r6EaFgCSrdRzqpl4FM7/CL9AZWB27XtFFnvoRKUA3FxGFMc05gpnfofYr0BqbAsJHaXn+O1RYZAK1kUh/lSrgGZHqHXlonO+ubSIv96SFB6DJqUIH9ae8gvKHjJzkvLd1US1XeG53rHYRay5CNk5/NYWC0AxwjId7XrdOgPCMIBUv7HixyBViuyPQOtbWHlE6GNgUz5ENrMqTZzJpUYN/tHbSVOvclSjascx9fC4GD7Vgoq2tsJ3XuyxLQGmYluWG3fdn4iC1F36ZHz7RQtWptg/1ezez3ima/xxeusBsSrVRbbYT9XqJ8D4xWohv2PBHWpkrP9A/VHjEAo3ySaYppmXIOyzcpJ50cvj+CVLkvYz3Tq7znTSI7kDbkXMQGsVPY6Oq6UrEVztgBMeddVO8SCqrJEAxc1hp7kA1zLncJn2bSweE8gWiVUe7FM4JSULmS4Uw6zeLYe268kHr0Koc9NqOehDO+10BU0NpmH/TQhE5QVYsUprtUQZKj74wM8XdueOrZR3MCYOdyMN6mFVRJAxJf2i4sKo0uffsonPGzcEUwJBtrl1ehL2lAKaiqpD9kxSLYhUWmpgDO9Q7CGImqrN8STbvOx3g/sgNzNsnKFC5Ny/7SwqBISMMVCXppJmzrfg6F+UEkDFEb6DMXx4ltMYCwhqvZe/LKVy43Y68nBS8iSgPJ9VeHM9y0UHp7j6yvnpw7jC43Y/Jte16E/Q8/ABGRWX1qKOb99sAwSBjX1k2kdlayvvpwrqmykp7pH9xiyqrw4lPp74uZkFTnfj5inh3+9lDnvlyhVM968JFK631oh97lZBZrL+EVEfveTAAZd3Vou0m1TH0d2jQrlnmqQ+unFr+8HMagbp79plvC53abfs7L5pKp8tpdlQgn7Eu9XsDoy4behd6tKpmrzlNsKXbKZNosnmKHekYuP5KhDCwa/oq9LUovqOzefvpBkGjrMITehd6+mFXIWMR7e+mVSL/Eu7ubcKtt53oHhAoHY71li1jOq6qnDIU5YX8/OFTkqYvbbUy7N+tWdbQFm7cX9T1hOrC5u0TtE96lVLgCzjUT/a1pAvLc/WYmILQd5fZrx7SN3B9GtTE1rN72Ras2NsyZrOkx+2JChSukmxRnIlFfRDIvA8FIHpGwlZLf9Cqj957V22t/UNnQdgkymd8gVuBCGPneaK784hS6LRoAxoa96iDiiqJkh5gXlkfcPFN1jYckmWztC9TW1UikaMBY23py7Izx2Zigcum171VF5dfbBO4OZxU2JvD2FrPDHpViCca4qiyHqTEB6c8T4m0aNRFpsGiiQVEUyd3dY5UtIoUDp+NqBMmWPqcEaJPVoVURLUTj/GolUQXeTt4Xvlh9gEiRYbSUOBYM/c0OpCXDtYooQdw9qiisQXz/Pl3vHYQs8rspDx3KvUapfc0SAHG1C+phAuoeQoWFoTzo3cAzDMOgDDJTDbPaZrpESq75GahehS/UQwfk7kk5zEDeXrD6xOCz36uE3KoVIkPOucBKr018zAGIplUNo6q7mXkD5e4dUp9IoZhg3ZEmBwo5mRGsTnnvVcMGzN0tJzRlf4M/IygUmcBh3X6doX+p4XrfQjUsUjgAldo9KUDqCykUJdr916lBis23dHBdtKMYZjhvF4qhhnP3hNbWqpEnUo8FCokg9i2Avfg1Ck5gMUJIt9fZY8STHRwHxdDBUSO1/pBODC89y/EAXeKMmGTTwz5hBmQn/j0bOqP/ZhixSISYY3TGkRrQ3YrhAMcG83BHgZ+e5XbEZje20uaAUU/R9GsvaIShm2OPUQwPAam5JM5e1EwdaMp4dhofIdYYnZ6FACeOA3w8yCEvpCdxgN5+BzZx//wJYWjkF01W1/XYV/sOFZE3UCeYnoWIIg8xCK5Bskncfr9nSxwXPUM3VQspjX7Te5rFQosQR7gkzkxUc4FiiI8QY4wiwCYuSgfIywg8PMglL6QjcdCAJu43SEHGUEFISDTZzSzLaj/tO3QE3kAHJ5KeRRBEHmIRW4NkmTgeeqZTAL/gEMzDbESYXlRy82ZkCBF7/fJPWHtQAvITYEb5AcP/RFYPuOgZuguGnLSQF8om9kLLN36CBIFPAaWDTc3FbxCGRooyi+N44ssyvsMgz6dk2frUuOiZnngZsKnt15e1kL0KL6UFpQKtvKA9xJf5OEh2QXmcDdFsjkz8bQepgds5z2opGXKKeFu5te4XcPRohTx5ITLlnAvEuxvnW0rACiXUQqvqAoEqOMZVAJKUQfgGWPA9eAIYhG+Bj1igGUIqRYrNv1RgFKqpT+0A5u6dYpjBvL1kcz0zjjiSmzT9W2Be8/0hS4L5SUsLMizxFv5lhjskUQ1VgJKrWj1AuXuTzfXMPPphwPnu7m7OaWLKnwJJrq8FvtrWAumPj6MBWViGfhn55yHMuxWqoVX1IUkFxZjJ9cxECn4SLFXit0NZoTpfUYMJzrSsfMwERK2etjsWQO6eeSH/e+2dzfXMROIIsJsbFiRwqcS2H+9aXGJLCYwDGPp37adLwSGa7qgeZr/eKrnFZjUMUybXs3Q03jF8vSxHCDoy4dcDSaYdb1P9Nykxin2Xla8ZQeildg0gXI9JII4/FWQLhLuLbK5nOCFHCOpa5Kkf+i9PHTSMsbG1geRt7wSrnWhcj0mS374uqcH37GZxmzkt+Xvt3WZyPUMKeUZQkkqMq9aAkNBqrO/OTlaZ46eohK9p5Qbna8oG+aSXWl+yDV+TzfXMVKIayCrzSMeKc5Um9V2U5bAEY1kZE4wmyJN90aAm3OQDd4t64fXyvtfe67Xvs7meCctkkZUgZPWETmCqGrvOrABC189d8FAmZmQsxsKZ47tu6kjSOR+qwvUNu7vXa1+2TmXDmj2T+R6NelZVcZOyMgz7321Ev+sGmWEYBmmSGWuYxRnm2eczrTBSiUw/D+nCRJMWsCuvJpU5/u/KoGhXXqkLNzbM7vaKO625JrrraMuUyZodHvWsHjKz9l59vP6hNxNk9nt5khlbM4vMoHGQ8tQbBnpyuHCaV5UM5HnxO2uqIEDpVnja/nlL6+Qz6SFYl65tFdnJZ7vGr2PcHclkhxGNejYppRj17RzKvZlPvEhiQbUyuOpkEABjtguGM1xmcQJeXzepJMunH1wJCm7uYm6/NtlNnMvknPwQoWc8Q2/HuLuYneqs3fh4d3vfVHjUs4w5l1RZk3mhKViW5JdeJLEgSbWWSY4CYcx2UrDpPfq6+8tg9RrxVFJvv3YYha53ToYI1luw3hNzKrS5aOLcvdQtGvXMldKRXbNUybJsJF32/Ae9A9qAoWkStfXabkCM2S5gXpSuvAFQmnb48/ZrL560pqtm59SIc2Pdxdx0bmXarjj4iHWbEZF65gpYIjJzBvcBDQj0ojfQCqpNxUGjOk990/db6+qCmVkZc/ep9BO0vtb332ofBkYcjsgSYd6rCb2t4mozxtDdMy71zBVgfpqYiJSUiqEdMfZ12XVdUe/m3BGXKpXkUaCM2U4WZSrHvf5U+lIP4358+PuCTZmSBT+di0w9K72RzdsonIHh+TbEVDQ+puP/0Qg6g0aVO1QYsw1MiEX49hbFfawKKvu9XMmCiolLPeuLMa1tLdH+stdm2B3CU+li7/IAEsFnUDKTJJgFrtGfUkx0Bir1LAYprbqqbIff7wWVRB+u3cZ2/P+mzRFRHTnKb9ANZjOIIzVQqWdDhL5wyoRkr7cT+ujtDmC5ywM5ItVYyldXVIMUeYNJPatZrA2lgNNL7YMiCN1uWastmnJEqhnUlZLCHAT5JnxFZU/BpJ4NGcOyUiJ4MZAMwwB5uwNeEJX74a8+NXwG7UqeFwePPTz8HOSK8gy/eiZo84gRoj5sE3Xub4gqnpM0AGFEX537+AySyl+uecgV9GwKpkgdavl8y/FR/+wmbjUd9kBnkBy35kGj6xlY36Fidajl3bbADHG7A7ockU0Fh27NQ56gZ0vQpxdo8j192A4oYVqWnhsyUJYzmK150PR6lkCMhnhdevGN0lcIfbsDlryQoxQi1Qm51iFP0LOSc57iCfgf2gAdAcAkJ6SruTo3dYFjK2Sh6fWM92FU3+NJEXzmYITppdx8kIHWgtvmGoc8Qc/AvRvAIWsPbEJAhjEPZCxrypUswGhtQ56gZ5znHLdRDmDUOSBOru5opnAnK03WNGhePVM5LWBneuWAbHIuX5Z7lEoyrNY0aF49E5Bl+2SCo8n/2Jaz0sfqEgeDtQx5gp4VIhnVPqkAu6Cy1SiE4yZdrYJZ7/6PZMgiT9Cz0qPBE/DXm8BTAzLlfax0PtM4WYF2g/VWkVud+3jUsz6eZcjU4zeKYNa5z4ucr879geKYUZ/JPtMz2kyZeTVwfiRy88sIj3q2ROtQy4mQ79tJANnvJQvW9CwUdx/GpqLcgYbtlOUGLw6iX/CoZ06wwMrDurtHJHCKLuYx80HUabqtkFUtOuu6qBdE6lmB9NClHp+2U+Zcj5mZfZKxkWnbmXT4yLAxerOoF0TqWY1kLTsD7swGnx3k7Q7I89g2vRZ3JouzrVSiQjpPrx5tAhHGES+I1DOXigmOcCdsL6DwuSGlhzwPZ2yvzcS07SHnCtP/AyHDjt841zuI1sCknmWcI6SkpkK2/aASf98rYNjkeWwqbPqbl8dcKQ4hlceZh5zzoe8aARK1gUk9cylCDSI152LaDijA2OV4DBSZ5zYiDaTcd+bUKS2c5Bzxgko9WwrSSWXdo5YftgV2cLc7IL8jUbUHXRPMTNu60eaoipLBpZ5N4WA7PpX2nn7egwrDvf4H+TlPw7wwgbNTp12VTOpc9VXECy71zBVgMF5Yg/HFD9oEPdTtDsAciAwzCyNZrIGRRJ40PiRaqz0EaXqW9N0nw+SU6zN4kn0vzfj6F84wgZnmdLWHjKFGNLiqxMi0rcnRJafOh0QJNcxiX8+4DTkPa6NOAs3F+Pu9g3bRa8Q7sNsdkKfVMGtouNbi0l1DyVzXl9NqL1yRQ+jZZCj3MWqiTNkthsy5tPSP2wWuALndAbkdM5a6FG1sdZQpyTYzUJg7Vbgnj9AzN5WrZCv6i9gLii4BZdj9je2D9gBwuwPyPEBIR1ayjUxu4+ARMOXWrUI++YSe1amuj1FlKe6/1PNUGgLtYdgGY+HSTS5jR6Dw5MplfXIKPXOZN6PvYtT066VY/x9jkLphyzgXjrkrI5aWZn1yCz1zi0K1XhGK2r0r9Jk/EPAuD2xZ5nztclcGLFTcoAr/5Bl6ltR92Mmow7JfO6HP0EiZ1ZTO3ou5yQPL5HJXdoTCTEeVP/kf2Yr8RM/aF27uIh0CSr3bAU3durwV1wUK88i3jQDJVuhqPVMqzo1FQ7nNAz7a7/95KwlQmF1XrQnQ1HqWdlJpu03XeOKxrh6DtOSuHKVAYY4V4XmmR3PrWVpJv+WxAAu4OSvDeZYnV9GzNwkLsA4kpY2RbE++omekhTzPcz/ztyW6rc59dOiZBJBMDkddnfuopsvaIF4/hem2dhnhRs+Uwn4vWTKhniHfNC+oMes/J4Pz3kmIqCgXb3zq69Z/rP+ceML13gHlQqvVp4ZsPSMj9akF8g7BarU695FNpFbnvl7iTes/tnHePQkRe4SLb51AxAPrP7Zx3jwJEVuEi58VjJUaZrcIVw2zzlhphchWtOaFiPWfk2S4fxIiKsrFR51+6usnDvGG9R/rP0qi3W26xXvfdEU1mupgW89CH33pI72g7slOUJ/zMv7ep76u0epT02cMgwrol0iX8BP3yYF8atFmFj3TTw2oMUDq3J8s+mVZpGU5zZZORairc3/uwrlt5G7KRG8fiG+qrNLk/VUjJv0hq8O0diGQXRPMMdxwq23whLJdfSYumVl/n/MyF/boPfNd2QsCD+2KM8lUitHeC8+0Dnf0YSjzjfsMh5Vm6pSlmLKfZXWWTYcwgrn8Xhri6MPQZF+9mdyLL5gc9KVU13SpzK3KhlCau7/fOwhpjEm/gjPbtd/kXTwwOGjgXJMY/N0VKIZwxgHiW2IwrSrTcL6bczEx1FUsDWGMGuSLhLGsKNMHY+Os/pRB7n7SOwhftERY8qFfSWbpjVZJqaQQw4thjAro6zezrCLjQOZT316w872D0MUSAjHsci3+2NCgEemsRpLMauEnvbDFGI4gFPnWxMyoYTYRqayhs5RtIXi+F7ZqmK3BaKS0YbdTXXddV9b9mD49002YGa0QaQSxtXAlnIQsL0S6EJhhlM6G3VYHT7OmmNKlZ6tAOM5RlFEseCoNWzRw9JKhLRrv1zTz3i/9ah0LaYox4bthi3AEopYLFTG3pRkRcBhX6Sg55wLJMkDOH4YrRjnTduRcXItJwplK1/J274BeIFJYCxT5UvZWHK6YACklwkhpQxLL96en0LRcIRjEggkKU4/XwyoFK2kdt1v2JpMUq3IsBEg/NbFgrftgTnYOejisnpFEihAzsda4T2kPyZtMUq3G0Ufmk5C6op0s2EA0mWRcNYP7QRR44ZrWMuQE0mSSg4blp7SYCrD2EEnQATWZRL+y1zuIzupT02c/DMOgvhpmJeHuSQ5NJhmGQUcRodW5j/OnhCEvRNjaMivJ9syW8mgyyeoPC48UT8C7YYteqi1Va+GaTGI0gvOSIUOXf9eLTFsz28OdbGU0gvPaQ0IXPdDHxiRBAfiBPO16Z88iNXBew2z4ahVRB1K0kCzo4Ki0awOvKQYL8BDp8d1e+GIGuHLG2qUaTgV7r8CRCKZI0j0fwnBT4qurLp18KU1GcN6a2TDaAw8SpjMZwXWL92vo8X6o66UZxqYOla3LxSC/rSJK0PqzLcNc2PsOORjPCJqBS63fwRVhquWMjE3IJ1qQv41pM7X3y+TkzEbDXiga4WhddW0xhrHWVccvvvNBQMx1dTR16+TCAregejWOqg+GvvHSIeyUkFAF04uKxlPt0LdONlbCMWlXIrU691HaSzOJmZ0kTS/NhKM699kDqqIjrN61MnJ3BUcrvp5pJyI1v4xw2pNXmf4d4wII9Fj4ElBoB2HIm1nJuptLk0kO4uuZ2QU6B6l309S9vbcYQmnSiafP1qRw6UCRSr9yvXdANagWqdwnsmW2iONl1hSKHgFNqrUwFRStfuUK4aBKSoYz+l4jXlTVmhUeHbFWuJU6qmpZGPUseuOqNSwcOuuxmVftqKq4SeXXswjFwlXVWhYOHTpbO+MRpAZk03LIzGXZrxN1ucg1MGNA8ppDDEewrWdrYlwfDAnDWi3LHiknrU2a3SYZR3fyBykdGGe25KhticxqD9FnDINkkf6G3fvIDEObSQ5jWvRMSxGd1TCLcUJSDbPSt/EQ7ey96GORNiS16VHMlSn21s44V0c6BDTa+43hXDc5d5IPrq0WT2lHUhiDMpjTuGxRYRSwmKpcPGFd0QfLOeEBJZm1h/yFN09CxBbh4u2TEFGd8h/bOG9TLshKnftVeKpzPwvqmX7ieu+ApvhlVIUnv4wyoJ5pKK6c8h/bONd7Bycd4grhYu+U/1j/OQGFPcLFm72DqKw+NYxDtupT2+odRGZ17iOcyKzOfes/J8zw0UmI2CJcvHcSIqpT/mMb5z3KxdOjegc0hYpycYUC0u7qrmu8P3SbfjYEoB01zCZxljVlk2XLBB155PBd6Qkj+27RH6oWWzXM6ikIhxcik/6QVWHYxLjIIof3S7QvO7db2s0JRlUrRFaLqIe6m1glImp1uch2yNyRvUBN9gM17QaW705mIB5KMcrbnJd7M0tZn7VF+obsKXymLzlBrs+7RAPXZ5bzo8pSupFVDASZuvZJleiLcNG1OT57NINJwTCmqZ2cNls2qXy/hNsldeIvS8rJCkxSmOTS3lRzt5DkidsjfeIv1E6xAgM6R+bGGUjkMgYixQFHmDuXWJ+gQALGdGla0DtluPDldke7QNzd1vHkBAqY1rL8+00Zur0gI84dkO7yerboBZk8Wg7TQCPQXbLvT0gggSPC2bx/EmTApiBfto+8HnJRn1opWFiKFgv+ZWAZBhii0+QP5/P1qW1Rizr3Y84xdFEjhNvvHbSxOvd3YN/ImA2K6txf/aERJNY4L4pfb2tswFjyeZ4RxAInxgar1PcnvTbGIQRmaHN5KmJBLZhgFfv+ZTsjHIGYTkCg4ZyHKPJNxcM2xgxIdQICQ48+CgsxaWNMgNQnICASWEucKtcEv98jVqid7CMUxODVeEm2D8kWV+gFiKf1zp9kiBLQCZ5y8gSdYEUnT9AJLEPm9kRkNcxiG1ijWlBrmBWq9pB8voZZ4WQfofBCJBHEeLWMfBhM9vtBqBpm8/lWiFR7pIIKuCCCwlLwQRvDLWCtEDkJgZJznqLQ55xP2F7791StcycAAFWZvsMyg/R8r/17M7s7EYGJoMZ6vN+jWh7vV2/SCqohUqX7Jed8t3fQ1nAHg+GqiKotYsECZxIL4f5Su2MMSF5O9kkBwDRUNrVd7IaC72R6B0SrddVxstXr9aKyVhFh8sklhg/Y/b22h+uT9sCDy99xy37JMLz4/C//bzRmKxuRD+fc2Hs8m7r32x+uT9RLM2WbuxM3rOqYMCFCUbug8X8e/h81842HiubhN0LwuJBdO/OQNZBrWW21QdzOIOjJq7hULMsVi9IsOEx113jCus2uVSdtXxw8YU1XjwplrMqFdPeh6Gc0vmTXm7WUql8Fl2Hg19t7W1pv700uQl5en5qbwjmWgyzGMA4E8yZ4LNoLl+FhSTGqkamM8bZf6laJtNUS5+6uR6FwNSCxVSaGOg3KVKEgCdvRegR0ndIeAc0FbZnKCa9mEccujpdZIffUXN//tJvIjpn8RZDcP3ez+pg6urcpgq8+qobF3cuExyHNnEFsCiQqXMDh1mvEkXdQ+9a5nJ1aSnQ1E3Vd2oshFKd0PmfX+MiOmTxZIBpQK452432st7dKPCgZD6zuLlsUDtIJ/sRMeTVAef1I2iMTm+qyW3zT5fQ9i95Ic84ghSYfVvqa8IhkG38GZaQDLlXGfCDPG2U6cbRSGn3D7u5lROTvc1z0l7GrnDKDW5HGskdWlmNjWd5yOLEpelA4/9KWYcEjCTrgUl2MTbQIOqaDkkJlo0/gbt9gcJgUNoUgaRVLkNiaOrvrdRv/UiYpnJYYJAfnkd3bWykPAhS79wKLQ1UPjS+UpVUkqWzVzW5Y+JWCrMAo0aAm2iapuyX8IHDVDMgdqro0MqZGrSjFOEUwc5v4kx6gRseNmugA3D3JDO5vyCU97kbir3Z1U8h342O6SDyhqBVLkc4qnLXPfTlPfcnchCQ+JlERFYS3l1ZFdBDurlFZ0GwcO8OmNBAohihdrFIAn9bV+CkO4IobpAbfMAUfk/QwDSOckSlzlAn7U3EZjnN+jQ+ZgdpT3KuHGuaQpFHQZIGhWAViCjyqwerOOf9RALHIDL5TgImiphCT1R8SzkGf8+BMFr6jje6bkz8sMcQ+JFmUwwGKzeoPtaKHVRAWc+k7ejAK1VCCuXtUDDPchZoi+FIaAlvnvhDk5HXuN5xzgefBsQ2CH5IsqoH0tnp/7OHcPesp9ntBwFS7jHRYR81NwIv9xgK4c1YLExydYijg6Fd+4DznJZ6AtQRzGbReURGcwGKEkG6vs8eIJxUcjWoKQQJ8Sw7zHNUCLz1DHojtPAObI0eJ9IVp4oUA9SVYJTSnoZ3RfzOMWCRCzDE640gN6G7FAOjtDpbrC8LQyNd+d7i4uPjLgCLgp2dZFi56hvgu58GnzDJ9qbmBbo49RjE8hOxdOS6JSx7VgW40ZTw7jY8Qa4xOz0KAExc9awuQbwB7kF9eSEniQHkHNnG/QaYXVbxvxQ+G/TAMT4TadqiIvIE6wfQsRBR5iEFwDZJN4vb7PVviuOiZVuGm/tc+7GkWCy1CHOGSODNRzQXSoBfIUCIJ3I2L0gHyENqD3PJCOhIHDWjifoPYAnDFli37/fBEpG1HjY930Kzykz6T8Il6iEVsDZJl4jjo2YnCuJD9YAPIoKd/2CJPDxziU4YCn71sBGSHwqcskZfhv0a1LYyXKf4p64QrWmAn2TLOw8OKfG4i8aCpF/SMsLODZNwsyEL9NXjeGyUqAvafs1NLyZCjqt+kqeGY0Kt09Kn2kL5HinhbdnARqU9SO9XUHrKAubtXDBPgYTd6oU81zMZoxrJTkYflO0YwKtVQgCHXhhmJdPNVy5UXsJJYJVbV/Etkv4xE6pPUWTVMqj4kKaDoV4Dogxdjdeec/6gCQM7jUU/bHQ9A7p6Uwwzk7cXhl3M98kQiKKzJgMuz2AjUJ6mzetjBeLtz6rESxt0VgrlCn6gaDAevJ7Az5/zo8X4fAPBRLxVdA0gH4u7JOSVNIKYrQSSpfZQrgLUku3aiiXzBZE2rIsbwPTkVN+q1DobkS14Nolog/alxlJU/mQEupnByKsL1qr6ExENyd2/cihBVifLn6JImPsVNxBXIJ7ny+t6pCVf4tZIvM7pdkrq7dCjmXVqEbVqrNu5mDKnnCv2K6z373pnsnQ/Z0eIhcLdye+djbJK5+9DimD0SRTUBz+GpM2EhRMq5rvyL27HunbdbogeHpEeLByZ3b7drxXVEw3xI4u6ydTkC64UD6Kk5sz5XYa4rP+NGcu+8ZejIq2bnVIarPJO712vvO7UdkbRl6G6mTBbWLLtSROUKMDA3S0VE3Jf6aS5dVgiI0O956y3N1pH+v1MbbuxY3O19o76SgPrGe5ZMdpjcqhGV6zPnMGJY9rOslHufh7mYVH7HubmI9pkrpeNcu8k51eFcv8S7uylU+Certm7iM9nSY1rhPupFZHXuayGLCziTS900p+xIa0N8N1MdPM2WzeicCnFuKhpPs0OlylN42r70NGuKHa71bSuo936xbRIi/TLSRFYXzKzppjnQB7yG+PP/56mu33nnnQ+x1LvwclLlCu9F9/UH1QJ31/Wk2DOap6r+INs77/zPuh8DdVtdoqom/SGrwt0npTAEYIrfM1rlSUjMVV9xoqqqZCmeVDZZ/fuR2D0bMlnQU9ajKrx9EiIqysWVU/5jG+c64eKnZIUrhIs9j+jLMiTSicTqU9Nz+DnIlfWZcCUaq3Pf+s4e4eKjkxCxdcp/Vnt48yREbBEu3jsJEdUp/1nt4T3KxdOjqEoNsxXhqmG2ukJVvBCpIjEvRCzwvH0SIq4QLl4+CRF7p/xntYeXCRcfkRX2CBdbp/xHk0BW6lP7iHDVp3afqtS5X20Rrjr3q1P+s9rD/ZMQUVEu9k75z2oP3zK4+IKxwgODiwcnEPGzgrHCPza4cMYKzqRkUtf1OHjwM5F4Etj7Cub9J4HpMsZ1XU8CB1c9WytTz4Teeelj4izL6qBxytMEq3P/QsX3S2sYhkGXUWdZFgcNvrbOeHr2G8MwPNHJ3Awc67e5mlDtp1E0+71G42bgeJuvuYynZ78x7LUzbfWbzFlpp/osn+XTg6h209MdnTjsMeDu38AC39VGejZIRM/QzM1RUAOKuOEOHIbAfveoXvvdITCJrHoUILKeNVmWlWsQfabGndeAZoQ1Lpqm+QXnHTccfxDY19Tr7uPXfhCYNLyNBETWs180TVOsOfSZIndeeEZHGxLQ0nqGZZ58V0vbEwzw67paz04gMus/1n+s/1j/sf5j/cf6j/Uf6z/Wf6z/WP+x/mP9x/qP9R/rP9Z/rP9Y/7H+Y/3H+o/1H+s/1n+s/1j/sf5j/cf6j/Uf6z/Wf6z/WP+x/qMqkpRBmFlwdMggDOiKPiNHHHEOAo+McYcsCUZY9Bk5CgPOd3d3c04TU3YkuVaERZ8RpNC1VInbskJ1pEWf0aMowG5uWJDApRLbtEWfEaaoIxM2JZl2aIs+o0xRXYs89UNHW/QZqYpKUokxbdFndCkKaMYjHStJYQ4Iiz4jRcFUNXadWQGELmHRZ6QI8+zzmVYYWCKTtugzAhU1NspTpy36jEJFHTp9nvqKtugzGhW1o7lAJefUERZ9Rp7Ctpyvvq0zYrJNWPQZgQqbioNGdZ76JmHRZxQKlyqV5JHCmC1t0WeEKmpUuUPZmC1t0WekKiqZSRLMCIs+I1KojhzlN+gSFn1Gp8JYyldXVINEWPQZoQozqCtlF7Noiz6jVlFQYa5oiz4jV1GSq+q0RZ8RqyiJAWHRZ9SKTQWHjq7oM2LFQFnOYEZX9Bmt4iiFSHVC0hV9Rqroaq7OTR1d0Wf0Ki6X17RJuPQZPTFpu5IFGNEtfUZOqj2aKdzJShO6pc+ISdXYh04ugZnzinDpM1riPs1YXeJgQMD0GR1p0pWiqMER/dJnFKTZn1ofRLuOfukzAtI0dH0m+0zLaLUaUSt9Rke6D2NTUe5Al/vOzCu6os9I0XspupjHzAdRp5WQV1RFn9EiZmb2ScYGpm0Hg1XKuQdR0We0iG3Ta3FnsjjbEuyFfNeDqOgzUoQzttdmYNqW8zwHkCZ9Rk7acWVeHnMFScFBpAdd0Wf06L22bUQaKO2aeNAVfUaOjGRvVpqgNW2rgZToMxqEeWECZ6dOuxroij6jR3FmYaTIV1dDS/QZPWIsa8Qdy2KZuzmPlRATfUaDMDRca3/p7qFTQU30mWHETLq5Xjy2OsqUZJupoCb6zDgChNSutusRvI2DagiLPjOujK5oTOs/1n9O+Y/1H+s/qzIMmHlGWPTZCURm/cf6zyn/sf5j/cf6j/Uf6z/Wf6z/WP+x/mP9x/qP9R/rP9Z/rP9Y/znlP9Z/rP9Y/7H+Y/3H+o/1H+s/p/zH+o/1H+s/p/zH+o/1H+s/p/zH+o/1H+s/1n+s/1j/sf5j/eeU/1j/OeU/1n9O+c+JrhyNx+MEhWTsBZKh0eAkzRzCrdgLKQpAwMMjgqHRICXNFGKT8zznTRQGQMKuQzwyDSn/PSUBzyzQc4oGKmmZ6O2+HIwB57u7OeOkD1zKuSfiDoaA2zsrhmG1M0raAhdfJCzlaToabHYZhHQ123aoibcvfhEo5eBZQSbZAbuKRs1gfCUtGZsF12F2FIkrh14+wwCqyx1c3bpty3iHvZDDIETojlzwuX8k7GmK05TMDlXfBpBzc4SmgcPVeP8cJuUcsoI8sgNuFS0mg3GVtJQNQ7o5cp1kR5G4cuhkVkZT+yPI1NnJtgkVxi4MUHGkAxd0jqSC3aU2TckOq78NSZ0ejnC4dyQieQ6Sci5ZQRbZAbeKFpPBeEramHPDwF7YcZ1jR5G4cmhkRlBdrpAz4O2ORMYpewpDF0CqYxx4/hnh37SmaZBybvhtdI+wfhnE4S9EyvlkBUlkB9QqWty2CSFpidg1CyAhkI46xo4ia+XQyrhDyBiNRNrxEhv9a6gwtZkrAxg6JslKfBnm3wZKsVgijsdLCDhkBXmAWkWLA0TSwIE7Jc86BRyUI68B4YDwEFHQHdviTmPFMVQuB5XaGKg0CwzjDskROLA0/jYwSo7dRTmHgEdWkAaYVbTYDAYiaclRL91V7kKdAE5injfhUsgAkW35CNLigFQ+bJKCSotwqrnEN887DxajrEeNvo1dhOFplOcQ8MgK0gCzihYLjKTZkqtfo0yU8LgzIHHl0MqPg2fjXtOQ2yQfBiptubMtNmY86EK1AWTVideBsB+ZuTsbDEaDwSoVyhRruKCcMwkBj6wgDTCraLEZDEbSbICyq8IhSJRIHHYC+Ih53oRUxmCzwmQGSZ2FLhVidRNVUmSToahd15FxXdWBZTpQfRvbcpkmm6zBuL9t3rsAgUdWkAaYVbRYgCTNFt2Bzg4rEiqjTgAXMc+ZGEu70cA3sm4rdIbTiHrIOZClIzNT/D/BM53SHHcVIqWJtQmlu7vzawcCj6ygV+CjTvEAyaAtRpm/kM1OABcxz5nYkR+jHyqZ540PjC/HcpssdEZHZqxKbW2bWHxUCLXpmy3nkBXkAWIVLT6DAUkaMtWRYvftCEhcOTTzs7sVLpvSvndkdYG6Ud3brurMHHIujWblsVXDkV+AzwoSAbGKFg+UpFlitvtCxp0AiSuHZn72DvaxksTM6gJ1M9E7M9uashVWX9SOXwDPCjIBr4rGAJSkYbPdUQBXjtyJTYmRQU2o4uKRw9UgMeNI4tCKbVwsttqWZLDqwsoWnSFffHp3n4w/HV5pHA8GA4ewTVZfxMziGF6qR8RqWxN7FYxLoPcUHj+6O4bJeXoZRZQVdX0Z8dPTx9dsxGYkRCoaI2QevPeZ8wt2ScNmrCCY+kv+JDHqjX5+TvqbmsDrx58hEv2Z82sALh24dKRbOXQtqeGSkVQTqnjdKhHVeYEwUO97xqWeBjJdZ842M7OBRZMu0larpRivgHxAwHTmZAzlf3QRA3Aa00SWb8d4m6xIFHHOIk0KNnqQGG9rnLMhUp4sKwDwxc9Q1iWtsiYxVtTFoyMtwsdHz1lgyEhoVDQmLumZl/xdjEXSWkcw9Veyk6pkS3cSCfYUUquLvn5E3dDJCCdVDg7SkWrl0J/gqbajLtiZclU9RQZmeFacG5+AGrHCJsnImBXcKq3pV8baapUUA5blB3t+qtXoz78ZF/9LOtBpHLsVVC8rhG2yUaIDm6P41QB8ZsbbSnc2jFcTZoWkXD+ir8ubY+QcdBFW1PXT4/EmEmEyRY+uY2DMSFhUNBYeUTNv9IOfLJKGzEjmqAqk/hqw7pNkKf0Tm8Kn1NgdyaOqS/Kd6Zg64RIrBwfpSLVy6FgGRv9/j8TupKmnaNtAuh2JHRNNwLlC3sPZtnHes1l1SGhbLUuxo5MvZ093I3560Gkcr4B8HgjbZFMYE+NAzWpbacmCSDlsVmDY6dzcxFXWROQsAVbU83thguJS9M3nDiAjoVHRkmewe9dUHkXYscOuhE8A9Zfrcq6RXXyUz1n9Hevux8cjfUMntzUI5QCXjlQrh8Z9ZFNzHJRr00ipAQOJrs1FW7tSfvBgjIbFVuuRVEhuUNfiGPHTg01jzkDNAyy+GQtUEiFgsq1MQKQcJiskLdKbVl/TzU2o41xXFD1B8VF+DpGR0KhoiTPY8f61Y5E0O8wEkjczePrr6NBir0qYwpvjZ1hSFZ6iBKUcsNKRauXQr3QlBiaVOO4qKjLMFXulySUlJiSpRnLgR9bgCOUB2lYrVI1KZlkzHCoU+nZi74GmEUojgbBNZkeVeJY39dvKCETK4bICe+xubiL1NVHOROe+oq7vH2/o8aUVhgmRkbCoaAyK8c9iM9i9axZJQ2WcyoyqKnD6y3V1e5UimXLB5O/YBB6ZVsH9a0jlAJSOVCuHdoXzXRWKfU69Mwsx8l09Y5kjeVvMD+dyuAGHs0TBDGyPtDEq//tsttoEncyK/X6A9+lBpzGXP8AVwjdjc+0NBmbbygBIyiGzQvy+M7I2P3N++fj83pHiMt4ryt0xxZd8+wEiI2FR0RgU4yY+gz1lkjTs2kMUV8YFSn+Z7VW5cq+KppDd3TdEqmPjew6sHGDSkWbl0K04mW0DQ4UgS3g82AEHRTaMDbWGpDZ0R1ObieKi1MFgIBfc3BlIIUHbaqUk6c5oPN5WCJdzivfpcUgj1EGbFt+MPzDe1jhAUg6bFdg+R3NBq5uA3L3xXlHntPh+Jia+N8cvAmQkLCoaRTHoZ+F88fHTb9KOP56zbKlWrJQBpmJUsgZNf420e9WmKplyHsvN8XjvnJJAaqovg/jeHaM+vQ+uHEDSkWbl0KVYaL1EsQd1JVGSTc6N/peXGRluzqFKbaj365m5HSTcraZF5NE1QxUclzDx55PG4/He3d1nnrOvGQicVh6DkFTG2xoHSMrhs4L6uFVd2OE6+heXxzxXlKZ2zCP5chlpe1KH8UNCoKKZoLyUa6BIf+yYqDR8L0RyngVQfznFXrU6UommO+9Gif39y5g/8MUdNT2mpOASXDmApCPNyqF7mSn0ilMf8ZhhlxnrBqlQGzrFYVXDLO5WU+LxiKnugHOY+HNJ4/3z66RrBhqXsknYNt5WOkAp55QVRpwbFDh4DppcNqepD94NioYLvY7yQ0ImitLORnXDJgYqDdtTtTznnSDqrxHnVjsNBco6iB1ujvdYTtw656AcINKRZuXQvcDcXZPrcPFQG4Bl1igOaXsMsd7qeBg+A3MdjfFd0viDxI2V5wC+Sc5YJ42B1+6G28oAUMo5ZQVVPpTB1twcr7muqK6ZvCNm3nQ4PyScclRVRukPRCycAg6c/tLsVWamnL4Zl8KnbG+yXbBE946HcoBIR4qVQ/OSqNMNmlxga4xt7nclxaHSGjuG2G81G48ZS12BiT+PND4F8U1y7P5HMd1WJoBSzicrOM6Niht8McIlzxXlDOWtRqtBVdn9kJCKrHvMki2bLflH+SreSYKnvwz3Ko3AFwlK/2Y7TnlM2dY4iDmIdKRQOTQxY1MbJDNsdDXwdIHeUGiNsRm4W836MZcoMPHnkcYLEN8kXxA+F2wApZxDVlBuzeFAEyKci7GiTILdDwm1JIYOzB6ikiuWy0IeJgHVX9uyaHZ1yUVriGOL2TlTe//vuCgHhHSkUDk0MQOJmY5tC3CMoSajFXMus2kG7lazbuBRYOLPIY33IXwDUDLkpg5dAQzDbWUDKOV8ssJA/iHpws0/IxFiRRlh90PCKzkfGu+ZmE/JWbEWUfwp0PprIDHQMZLFSZDC8wiXbNsaF+WAkI4UK4e2XZBWP2jFkzFStFh6eaXSGmbYbjUsMPHnkUYI30DXHpLodKUZd4wApZxDVlCxm+e5WSJRiBXVMhC4IISf0krCQKVZYSLJoQuQ/jJibPb53Bw/A8AX2bY1LsoBIR1pVg7NasrMh8pGsVQ2HKFtoOoq0LERtlsNC0z8eaUReM3ERdTk8DUZqMKKc0SAUs4jK2gwK2wuxIpac+txxe6GiGuRZchNxQYGSH9hcscLPsoBIB0pVg4NW3uIGSMdiYyRiX+8cMi5DbZbDQtM/HmkEcI3XFshcogKTMphs4KenA3DUyFW1JqbLufotKr2kERlNiA4+isA//ARcwDpEEE5NDk7ytI1WGVjdAJ2EZkpssHacfQdhG/gW6qm2xIz4IN5MDMqnrRj2ukbQqyoloFARUOIln/9uDXMGu63bSNaLQ4v+Ig5gHSIoBy6G1O1NzZVljsySK5CVIPDmVWCY9sI262GBSb+PNII7RsmpbM5uDvyxMIEJuWcssLYTm8KsqLWztgkqIxOw6kvE8P1QsSwErFREPXX2DTZuJJfxgs+Yk6VDnGVQ5NzpFNtTqHUDE84rUxUjXFnhm06i8OosRH2Ww0JTPx5pBHSN+xvqZuXADhKkYFJOaes4EzdRxIiUyjk1W7zDAsEKhqeHaQdvUqzw1CY1AVNf2lsBBl+Ehe84CPmdOlIj3LoYowvuT3SlTo4dEo2FaC4HHp0CO3KayVyqVpPGORz2201MDDx55ZG2DUTr8vms2MxogKTcl5ZoWt43uHyQoBMsWkm74yF6V0cPY5BRVOjTLboq6YY6VWaLWbmnneCpr/M7Ug7hbyOF3zEPJF0CK0cWpodiZE2+7frjK6kWGG4pT5mSaTU0wemO7yEZke13WpQYOIPmUYuvmEyVTWOchXZag9Bd3EBSTm3rDAwO5fy+Hg8558pRkp51duQczpe27IjMbKsT21bdYSG6/G+cYJkHDD9ZbxXOUXK8JwbXMQ8oXQIrRw6mm2JHb1SOxwruhcm35XBKOS3qdiz+XDkjNrMOlKpBZUgbrCa2Ww1NDDx55TGhGsmGe6QlZUcbMsczVIWZkryHBOQlHPJCppaBNJttbq9IRrF7vhmikNjeXPmBAMMKpoaOKjcvR8orIDt6MsPo6BWg4piREHSX6Z71birqnOfG1zEPLF0iKwcmhbctnsOVDt9dzDahm7GGLApkRgw4FxV5SOvZoPRaLBzCITXFX0cq0SZbabMPDbfanhg4s8njUBrhrlLQnTfxs5gAJoXI6uzHBGQlPPLCgPONYnF68eEAN8c7z/mvaK2VfIezoDdJ4UgYww9jkFF06D8BUL/fqQfe5roVRoqVVd9AW3Q9NdIKdrM0ydd1V517rjBQ8wBpCPVyqF/r65accCTq1zKy2W0SrZbKTFxOSbPTToUZWxypAoF40OLrYYFzHuc0gi0ZpgHo28jz+FWpYec4wGSco5ZQRIPiNYlDxRFWVGbRvJKZmIwQKGiaclVv0BW/PgHBioNlzErE3QB0l/We9Wd4wcP5QCQDoGVQ0czlhgY5aPkcmIh17GtsP5hOshqY1dbMbamSPOOevvAVlhsNShw3uOTRqA0JRq034b8XYxXmICknENWUH82Sm0fxC0Ss0vuK8p1DeSVSoiigEJFU6L4qWprzF9VepWGgMFLMPZMWFDkt92r7l1zhIOYQ0iHuMqhm1Ecr+hOo6kSDYodjnMNMyhq6mw6rtK0O+ZGustNc41Uqc1WQ8Mc/3v0uPFJI1Ca2AehS42/C1RAUg6fFcxz0Kgx47+inPjQzVoQgaLHUaho2p+SwY8XpFr0Kg0bJ20MSHMER3/Z7FV5HvaQOkfgxRxCOgRWDj3NSM5UV+/NuVqppEdaUsjMfEtSxWGROj+G020Tj4WVYs3MtxoYMO/xSSNQmtjNDQy+jBwqbFxAUs41KxwdaoWLXMufACsKNsNPt8ldV+GAQ0XT/pRmuh8vNPCjV2nYVAOZQ1cFR3/Z7FUs3GPkCbiYQ0iHwMqhmdErv4HmQEhVNaE4AgFiKNhWFbq08ruT9f26u+OMlItKrAHWVpOxoMIem+u7mPjfi4k/pzQmShNgrxFrv4uZqwRAik3DbY2LDEjK+WYFN9MJdyRbrSzCiqokT0x1PzOcHxIqFY1zaf8a6H684Mwckko7YklYo4wwSaRA6S+bvSodJN+EziOR+SLbtgYt5hDSkXrl0LETyKFBk8WZ5VaDrxiEbUX5PlhziFXY3kxZHbqDxCxtIguX7iR4W33vSNjjOJ4SsYn7nfaLd7T4n8fEn1caodKUbGZwyMpwKH0XIyi+zbbSkgWSchGygip8htxRiLGinPozT+Em4/yQUKloXcVP9Ghl8BO0Vmka0c3kOpJ/1qMA6S/zvepQ0cb7JCn84n0yihdxZ2ERdg6rHBDSkYWVA6+9Mret84RqZ+WFmVHPUiSKqW04GgxWIGwOts1ldOOBCEJbGG81tDHH//IaIm6Pr/mlCTq20pcxEw3f8+GnKkRWqMaDTUm0a8FWVHUkb9DR2lujGR2UwN/gYNtYpbUsBFp/HQ12WrpX8VIOEOmQgXJoWbYFmNmqMyjmoAqHITLXaPgqjf6EFVSAYyRc6pHzoSMWGg1bpREgEqgBV0gMJOVKLTQatkqjLQFbBWIeP87ohUbDVWk0KOwwuB4XR7UCCbuOXmg0VJVGs+6GZ9cRlrvh2XX06o4JO7Jxx4QRVdpJiMIp/znlP6f8x/qP9R/rP9Z/rP9Y/7H+Y/3H+o/1H+s/1n+s/1j/OeU/1n+s/1j/sf5j/cf6j/Uf6z/Wf6z/WP+x/mP9x/qP9R/rP9Z/rP9Y/7H+Y/3H+o/1H+s/1n+s/1j/sf5j/cf6j/Uf6z/Wf6z/WP+x/mP9x/qP9R/rP9Z/rP9Y/7H+Y/3H+o/1H+s/1n+s/1j/sf5j/cf6j/Uf6z/Wf6z/WP+x/mP9x/qPjsQB)

The first, standard grid search optimization, is a brute-force approach through a
                predetermined list of combinations of hyperparameters. All possible values of
                hyperparameters are listed and looped through in an iterative fashion to obtain the
                best values. Grid search optimization takes relatively little time to program and
                works well if the dimensionality in the feature-vector is low. But as the number of
                dimensions increases, tuning takes longer and longer.

The other common approach, random search optimization, involves randomly sampling
                values instead of exhaustively searching through every combination of
                hyperparameters. Generally, it produces better results in less time than grid search
                optimization.

## Batch normalization

Two techniques, normalization and standardization, both have the objective of
                transforming the data by putting all the data points on the same scale in
                preparation for training.

The normalization process usually consists of scaling the numerical data down to a
                scale from zero to one. Standardization, on the other hand, usually consists of
                subtracting the mean of the dataset from each data point and then dividing the
                difference by the standard deviation of the datasets. That forces the standardized
                data to take on a mean of zero and a standard deviation of one. Standardization is
                often referred to as normalization; both boil down to putting data on some known or
                standard scale.

## Model evaluation and testing

Once a model has been trained, performance is gauged according to a confusion matrix
                and precision/accuracy metrics.

## Confusion matrix

A confusion matrix describes the performance of a classifier model, as in the 2x2
                matrix depicted below.  
![](data:image/png;base64,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)

Consider a simple classifier that predicts whether a patient has cancer or not. There
                are four possible results:

- True positives (TP) — Prediction was yes and the patient does have cancer.
- True negatives (TN) — Prediction was no and the patient does not have
                    cancer.
- False positives (FP) — Prediction was yes, but the patient does not have cancer
                    (also known as a "Type I error").
- False negatives (FN) — Prediction was no, but the patient does have cancer (also
                    known as a "Type II error")

A confusion matrix can hold more than 2 classes per axis, as shown here:  
![](data:image/png;base64,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)

## Precision / Accuracy

It is also useful to calculate the precision and accuracy based on classifier
                prediction and actual value.  
![](data:image/png;base64,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)

Accuracy is a measure of how often, over all observations, the classifier is correct.
                The calculation, based on the grid above, is (TP+TN)/total = (100+50)/(60+105) =
                0.91.

Precision is a measure of how often the actual value is Yes when the prediction is
                Yes. In this case, that calculation is TP/predicted yes = 100/(100+10) = 0.91.

**Parent Topic:** [Training and Testing Machine Learning Models](https://docs.qualcomm.com/doc/80-63442-4/topic/training-and-testing-machine-learning-models.html)

Last Published: Jun 24, 2024

[Previous Topic
Functional Testing of Machine Learning Models](https://docs.qualcomm.com/bundle/publicresource/80-63442-4/topics/functional-testing-of-machine-learning-models.md) [Next Topic
Image Segmentation using DeepLab-v3 and Qualcomm Neural Processing SDK for AI](https://docs.qualcomm.com/bundle/publicresource/80-63442-4/topics/image-segmentation-deeplabv3.md)