{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"pygments_lexer":"ipython3","nbconvert_exporter":"python","version":"3.6.4","file_extension":".py","codemirror_mode":{"name":"ipython","version":3},"name":"python","mimetype":"text/x-python"}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"markdown","source":"# <center style=\"font-family: consolas; font-size: 32px; font-weight: bold;\"> 🤟GASFLFR:EDA,Visualization,Evaluation Metric🤟</center>\n<p><center style=\"color:#949494; font-family: consolas; font-size: 20px;\"> Let's contribute to the deaf community together🤗 </center></p>\n\n***\n![Capture.JPG](attachment:bb5f8f79-e14c-44e6-8b0b-2d80e13b8be8.JPG)","metadata":{},"attachments":{"bb5f8f79-e14c-44e6-8b0b-2d80e13b8be8.JPG":{"image/jpeg":"/9j/4AAQSkZJRgABAQEAeAB4AAD/4RDaRXhpZgAATU0AKgAAAAgABAE7AAIAAAAFAAAISodpAAQAAAABAAAIUJydAAEAAAAKAAAQyOocAAcAAAgMAAAAPgAAAAAc6gAAAAgAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAFVzZXIAAAAFkAMAAgAAABQAABCekAQAAgAAABQAABCykpEAAgAAAAMxMwAAkpIAAgAAAAMxMwAA6hwABwAACAwAAAiSAAAAABzqAAAACAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAMjAyMzowNToxOCAwMzozNzo1MAAyMDIzOjA1OjE4IDAzOjM3OjUwAAAAVQBzAGUAcgAAAP/hCxdodHRwOi8vbnMuYWRvYmUuY29tL3hhcC8xLjAvADw/eHBhY2tldCBiZWdpbj0n77u/JyBpZD0nVzVNME1wQ2VoaUh6cmVTek5UY3prYzlkJz8+DQo8eDp4bXBtZXRhIHhtbG5zOng9ImFkb2JlOm5zOm1ldGEvIj48cmRmOlJERiB4bWxuczpyZGY9Imh0dHA6Ly93d3cudzMub3JnLzE5OTkvMDIvMjItcmRmLXN5bnRheC1ucyMiPjxyZGY6RGVzY3JpcHRpb24gcmRmOmFib3V0PSJ1dWlkOmZhZjViZGQ1LWJhM2QtMTFkYS1hZDMxLWQzM2Q3NTE4MmYxYiIgeG1sbnM6ZGM9Imh0dHA6Ly9wdXJsLm9yZy9kYy9lbGVtZW50cy8xLjEvIi8+PHJkZjpEZXNjcmlwdGlvbiByZGY6YWJvdXQ9InV1aWQ6ZmFmNWJkZDUtYmEzZC0xMWRhLWFkMzEtZDMzZDc1MTgyZjFiIiB4bWxuczp4bXA9Imh0dHA6Ly9ucy5hZG9iZS5jb20veGFwLzEuMC8iPjx4bXA6Q3JlYXRlRGF0ZT4yMDIzLTA1LTE4VDAzOjM3OjUwLjEzMzwveG1wOkNyZWF0ZURhdGU+PC9yZGY6RGVzY3JpcHRpb24+PHJkZjpEZXNjcmlwdGlvbiByZGY6YWJvdXQ9InV1aWQ6ZmFmNWJkZDUtYmEzZC0xMWRhLWFkMzEtZDMzZDc1MTgyZjFiIiB4bWxuczpkYz0iaHR0cDovL3B1cmwub3JnL2RjL2VsZW1lbnRzLzEuMS8iPjxkYzpjcmVhdG9yPjxyZGY6U2VxIHhtbG5zOnJkZj0iaHR0cDovL3d3dy53My5vcmcvMTk5OS8wMi8yMi1yZGYtc3ludGF4LW5zIyI+PHJkZjpsaT5Vc2VyPC9yZGY6bGk+PC9yZGY6U2VxPg0KCQkJPC9kYzpjcmVhdG9yPjwvcmRmOkRlc2NyaXB0aW9uPjwvcmRmOlJERj48L3g6eG1wbWV0YT4NCiAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAKICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgIAogICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgCiAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAKICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgIAogICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgCiAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAKICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgIAogICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgCiAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAKICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgIAogICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgCiAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAKICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgIAogICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgCiAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAKICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgIAogICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgCiAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAKICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgIAogICAgICAgICAgICAgICAgICAgICAgICAgICAgPD94cGFja2V0IGVuZD0ndyc/Pv/bAEMABwUFBgUEBwYFBggHBwgKEQsKCQkKFQ8QDBEYFRoZGBUYFxseJyEbHSUdFxgiLiIlKCkrLCsaIC8zLyoyJyorKv/bAEMBBwgICgkKFAsLFCocGBwqKioqKioqKioqKioqKioqKioqKioqKioqKioqKioqKioqKioqKioqKioqKioqKioqKv/AABEIAKMEfAMBIgACEQEDEQH/xAAfAAABBQEBAQEBAQAAAAAAAAAAAQIDBAUGBwgJCgv/xAC1EAACAQMDAgQDBQUEBAAAAX0BAgMABBEFEiExQQYTUWEHInEUMoGRoQgjQrHBFVLR8CQzYnKCCQoWFxgZGiUmJygpKjQ1Njc4OTpDREVGR0hJSlNUVVZXWFlaY2RlZmdoaWpzdHV2d3h5eoOEhYaHiImKkpOUlZaXmJmaoqOkpaanqKmqsrO0tba3uLm6wsPExcbHyMnK0tPU1dbX2Nna4eLj5OXm5+jp6vHy8/T19vf4+fr/xAAfAQADAQEBAQEBAQEBAAAAAAAAAQIDBAUGBwgJCgv/xAC1EQACAQIEBAMEBwUEBAABAncAAQIDEQQFITEGEkFRB2FxEyIygQgUQpGhscEJIzNS8BVictEKFiQ04SXxFxgZGiYnKCkqNTY3ODk6Q0RFRkdISUpTVFVWV1hZWmNkZWZnaGlqc3R1dnd4eXqCg4SFhoeIiYqSk5SVlpeYmZqio6Slpqeoqaqys7S1tre4ubrCw8TFxsfIycrS09TV1tfY2dri4+Tl5ufo6ery8/T19vf4+fr/2gAMAwEAAhEDEQA/APpGiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAoopkufLbYcNg4PvQA/NFeV+H9VvdJutNvJtWvLyC/lijvI7qYyqXkwoZM/6vDkDauFwenGa9UrOnUjUV4lyg4OzCiiitCAooooAKKKKACiiigAorgvF63c3iZFh1e+s/JtUkgS1nKKrlnDOy/dk+6vDgj25rpPCmrtrfhu2vJipnBeGfaMAyRsUcgdhuUms41Iym4LdFuLUVLubNFFFaEBRRRQAUUUUAFFFFABRSEgVhWvjjwxearPpttrtjJeQP5bxCYD5sZ2g9GPsM4oA3qKAc0UAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAhOKUHNcL44cajrEOjyXdzDCtt58kdvM0RYl9qsWXDcbTgZxk55wK0fDGtvD8O4NU164eRrSGUXU5XLOImZWfC9SQmcDvWaqRc3DqinFqKl3OporD0HxTa69NNDFb3dpcQqrtBdRhWKNkBhgkEZBHXI74rcq001dCaa3CiiimIKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiisfxXq7aJ4Xv72HJuUhYWyBSxkmIwihR1JbHFG2oGxmivOPBetzT+Io0j1a5v7W+jmLpdn5o5kKdAQCnBbKYAHGAO/o46VFOaqR5kVKLi7MKKKKskKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAorL1zxJo/hqyW713UILGBnCK0rYLMTgADqT9Ks6fq2n6tb+fpl7b3kXQvBIHAPocdD7UAW6KKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAoorK8Qaw2i6X9pitjdTNIscUIfZuYnu3OABkk4PA4BOBSbSV2NK7sjVozXmes+KPFv2O4nsrzS7ERxNIENm8x4BONxkUduu2sHxLd6/9ksvM8U6qLma7gth9lMcCfO43nCpn7u7GSa5frlHozX2E+x7VmjNeIavp15aadLdT+MfFQWIfKkWo7WdicKowvJJIA9zVDUbvVfBvhmW8n8e6t/aUURZI7m7jlSSQnCrtkQkrkgZ9ieOyjjKUtrjdCaPfqK8Msvi1rljH5kuqWHiC1jGZbiPSLmFU+ssYdcZ9FrtPCvxf8PeIdSGk3c0emaqW2pbTSgiY/wCyeDn/AGWCt7V0RqRnsZSi47nf0UgOaWtCQooooAKKKKAAnFY2p+KtK03TxdvdLcK8hhijtSJXlkHVFA6sMHPpgk4rmPiD4gjaS30WG1vLxPOD6h9nQlI49hIVzkbskqSq5JA5BBweWvLO2gt4td0COBmt4WZY4mVY7mFsEqDwoYhVIbjoAeK4q+KVKXKkb06Lkrs7mz+JGktOItbhn0EspaN9SeJUkwQCA6Ow3fMPlJBxn0NdJp+rafqsbSaZe294ikBmglDgE884ryS8u7HW9Ht5rGcvNsTULRkBzleRn0zgqc9iwro9A1KODxVYvGXKagDblgMK/wAhkRvfARh/wOs6OMc5KMlqy50FFNpnolFFFeicoUUUUAFFFFABTX6UpOKzdT1/TdKjjN9dLGZmKxRqC7yEdQqqCTjvgcUm7AeORaQ0eiLfjWr5ooH+1iEmLygI5fM242ZwNvXOfevdVIZQQcg188w6JBP4F1KSG71GO+iguBJG93Kojfax2tETgAgg4xjBr33TZ1udLtZ15WSFHB+qg1x4V3c9ep0Vl8PoWqKKK7TnCiiigAooooAKKCcVFNcwwIzzSJGqqWYu2AAOp+lAHnfjzTm1DxhaiPUruyZNPbK2rIC4Mnfcp6Y7etbXw6t/sOi31ibmS5MF/Ixkk27j5irLztAHWQ9q47xXc6V4m8dWE8N3efZmsGihmgkmt0lk37tocbQ/ygkckcGuj+GsIsbrxBYiaWZUuoZEM0hkcK1vGACx5PKHrXDCS+sys+h0SX7lHe0UUV3HOFFFFABRRRQAUhOBQTXhniTxdfeM7ye3E91b6HHcPALSzykt+6Fh5W7qwO3LkbVVTgknJWJzjBXZUYuTsjf8deNrW9027SO9lttChDpcXVsxEl6+D+7hI52jkll6kYzgNniJdE0CDTzBZaRrGhiZFD+VYtPHKMcebGodGPuRu9xXTw2tro9q2qa9NaweTHsBztt7OPOAiZxjsC2AWOOgwoxPDM/jvx5a3moeHjpsOjW90Utrm6R0mvURuQMZUZAwWwOvtx5PNVxE24dPuOy0KS1L/gTxxqeiXMdtdXMOp6BOW+zTQyO/lqnDCMNlwV4JiYk7dxVvl217ejrIisjBlYZBB4IrwTxNPPFoj3vh0xWeqWksay6ZNGN8kikEQ7F5EnGAVzuUnqCCOq+G3xU03UfCVhHqq3FrLAY7R52hbygzBdgLdR95VyQAW78iu3D1ZST5zCrBJ+6ep0UgORS12GAUUUUAFFFFABRRRQAUUUUAFRXN1BZ27z3c0cEMYy8krhVUepJ4FSMdqknoK8o13xBp/iDxNvntLh9OjVIbO5uIx9nklJJLKDzk5VQxXBxweeca1VUoczLhBzdjtdc8aWOj3C2sEFxqd4yiQ29kEJRD0ZmZlVQe2Tk4OM4qLT/iF4fu90d7ex6VdpJ5b2d/IkcoOAcgbiGBBHzKSPfg157O0PhjUI5Aypp2o3CxupIHkTEBFYZ/gbaq4/hIBHGcWna3GvwX1s7FxusbggEL3Zc5HO1gRx/fNcH1+V78uh0/V1bfU9bt7mG7t0ntZUmhkUMkkbBlYeoI61LXIeBr0GTUNOVSqxMlzGCMYWTcCMf78bt/wKuvr0qc1OCkupyyjyysFFFFWSFFFFABRRQTgUAeUeJrW51P4m6lLBqdxZpaWdta7YFjO4kvISd6t/fHSph9r039n2KCPNxfXMIgQMwUyyTTY5PQZ3kmsbWbDSNb8eeJQ1/cvdQzxs8VrqEkW1fIjAO1GGcMCM+vFEMt9qPgXwHY2csDT2+nx6jOlwWw+Y/LTJHqXY5/2ehrgU1GdWT6WOnluoI3PBEr6h8Sb2d7W5szY6WIXjn24fzJiyspUkMMRnnPtXp1ed/Ddb+48Qa7e39vHBsjtrTbHKZF3qHkbDEDIxMvavRK6MOrUooyqu82FFFFbmYUUUUAFFFVdR1C30uwlvL1ykMQyxClj1wAAASSTgADkk0APnvLa3kijnuIonmbZEruFMjei56n2FczffEGzt9Qe20/TdQ1RYWKSz2ixiNGBwVBd13EEEHbnB468VxFvdWuu6ldTavp8sGoXMskkaXyjzDCH+TYQTgKpQEDBB5IBOai066j0S9/sC6kVYobYzWcrMMtCuAVbnO5OOe4IPUMa8ypjmm1BbHXHD7OTPSNP8d+GdSjia21yxDyqp8mSdUkQt0VlJyrdsHnPFdDXj1iltFcX6R7ntL4C6EZQ8sflkwMZwSFP1YnvXo/hK+a/wDDVs8js8kJe3kdzlmaNyhY+525/GujD4n2zatYyqUuRXNqiiiusxCiiigAooooAK5jx8pi8Ly6gsfmf2c63TDPSNeJD9RGzn8K6SaaO3haWZ1jjRSzO7YCgdSTXFeL/EOla/4H1ux0jUoLiSWBLeTymDFUmcR7vcfMcHpUzScWmON73RyNg1rp/wAVdF1KK8jW3unktrqOPBBnaPETtjoSBs56/IK9mHSvn2wN6/hV/DenaRN5unIbX7YjRwxiaP8A1ciZOS2Qj9OCete3+G9WXXfDOnamuAbq3SR1H8DkfMv1ByPwrkwcvc5H0Nq696/c06KKK7TAKKKKACjOa47x5Pfefp1jb3k9laXQl82W2k8uVnXaUQMOQCN5OOfl9M1c8M+JEvZn0q+kA1K3jDjdgfaIugkUd+eG9D7EVl7WPtPZ9S+R8vN0OlopM8VXutRs7LYLy6gtzI21PNkCbj6DJ5NakFmiuT1Lx/a2F3JDBpWp6hHExWWe0iQopBwQNzqz45+6G6cZPFdFpupWmr6dDfadOlxbTrujkQ8Ef0PbHY1KlGTsmNprctUUZFZaeJdEl1Q6bHrGnvfjg2q3SGX/AL4zn9KoRqUUUUAFFFFABWL4j8SW3h+zRpAZ7udtlraIfnnf09lHUt0A5NZPxE8XT+FdLtF0+Iy32oTGC2HlmT5gpbhcjcTjAyQOck4Fea6VYX2qX1zc3V/NdXM5KXuptIWIGT/o9sQAFUEfM6gDOcDP3OevXjSXma06bmyDVG0nxF4juG8QWV34i1O3hBnaBN0FmCxIjjGRggrjK5b5eTnNZzyTabrMVx4X1S8g1WYhYYNShkt55FXkosrqFmH+xJu9mWtPxP4lfQ7rTfD/AINjtbnW7idYY9M25CoRnLYIK44IOehJ561tPpWuaKzjx7/Z9xBeMFhurQEQ27EAeUS3zKSRkNnBPGQdoPnRdZr2yvb+uh1PkvyHeeB/F/8AwlGlBrqJYbxFDSKgYK6kkbgD0OQQy5JVgRkjBPU187aT43v/AAV8TPs2pt/aOl3DMkdzbwlnYsY1MZI+XepEbMcjjPQsBXuuieIdP15JzYO/mW0nlTwyoUeJsAgEH1BBB6EGvTpT54pvc45x5XoalFFFbEBRRRQAUUVyniHxsulaoumabp0mqXoUPOqSrHHbA/d3sehPOFAJwM46ZUpKKuxpNuyOrzTJZo4ImlmkWONBuZ3OAo9Sa8w1O+8Sa7dh7nVG0W1jwYrfSZdzs3rJK6cj/ZCgc85rAjmvtb8RXuj+NbuTU2tgJrGKRVSC4gPG9o1ADSK3DbsgZBAGa5JYykk7a2NlQnpc9tguIbmFJreVJYnG5HjYMrD1BHWpK8dkh1nw4pvPAzwQsgzJpUwxbXAHPAH+rf8A2hgH+LNaPg/4uS6nPb23iaxgsZJ5hbGW3dsW9wekEyMAUY9myVPqK0o4inVV0TOlKD1PUaKAc0V0GQUUUUAFFFUtX1W20bTpLy8ZhGmBtRC7OScBVUAkknsKNgJpLy3huI7eSeJJps+XGzgM+Bk4HU4Fcpc/EexS5cWGmalqVrHndeWscZjOOu3c4Z/qoIPbNcRZw2euwTDWrGRNTuUMtyt2AJsMTyrKT8oPC7T8uAODS6ZqUenXE2iahMgk0+BZY5B/y0t87VJA6MMBSPoR1wPLqY56qC2OuOHX2memWPjPw3qLxpZa7p80km0Ki3KbiT0GM9fbrW4DkV47p9vHDFe2EavHaSP58BjUgoJSScDGQQ4dh6ZHpXqPh+8lv/D9lc3JVp3iHmlRgFxw2PxBrqw+IVZtWtYyqUuSxo0UUV1GIUUUUAFFFFAFTVJbmDSbuWwhE91HC7QxE48xwpKrntk4FeRXU+meJLaIXniC7uJUw/GpNE6OO5jUgKwPYrwe1ezkZrI1Pwj4d1mYy6voWm3sp/5aXFojt+ZGa561F1UrSsaU5qD1VzyePw4NShu7W58SaveQhvKdBNHHwVBxlEBPDetZcWhLe6Xc6jqfiTWVhsLuSWF0mX9wIhg5BQ7ipD9eteqz/CnwPODnw1YRe9uhhP5oRVO5+Dng2WwntbeyurJJ1ZXNrqE6Z3dcjftOfcGuX6nJfa/A29uux5OfDviLU/E2nxatrOsQaRIn2uDzJYS9wUKkZ8pQE++CCcn0wRx1GleGdC8M7jGkInuJWfz7lg0r5ckDc3JxuC+/Hc1FpvhzVNBnvdIs9TtTDp032WGae1aWfytiuqsfMUDAcdBjvjk1DqPgSx1R5r7Vy2s6iIilv9tOIUPZRGmAAT1zk+9cVaS53C9kuxvBNxvbU321CxkuBZfaIppXyphU7yBjncBnA9zgVDq/h/Sdeh8vV7CC6x913T50/wB1uo/A1xGmeL9N8B6s/hXV4FtozcO1q1rFv2xscoHC8k4OM4JOOfU9db+MPDd1J5cOu2BkBIMbzqjgjsVbBB9sVjKlUpu8b27lqUZaMvWPiTVPA8cZvrt9U8OxkCZ7glrmxTpuDf8ALVB3z8wHOTXqdvcw3duk9tKk0MihkkjYMrA9CCOor57+IfjTTYNBuNE0m5h1DVtUU2cNvA4fb5nyksQcDhuPcis/wvrHi74LwQQ6i76zoPW4tV5NuOSzwN3AGSVOO5xj5q9bC1pOCVV69PM5K1NX9w+maKo6NrNjr+j22qaTcJc2dym+KVDww6fmDkfhV6u85gozRWZ4lurmx8K6rd2GPtVvZTSw5GRvVCV/UCgDyOx8b21/q7i6kXZf3U/2edGyqkSOojkP8DbQoXPXIx0qrqmr2eg6x/Z+yKV7uNp0gkVhDDPyQS5BCBgHJHX5CcfMa0NNkj1XwzbWsOm+ZaPaRr/pceyNhtHG0/MfXoB6GuJ1SyTQvFkX/CQX02t6TDGJGsrlGW2jbkKjuxYFgJNy+YegwT8wrwYqFSq3LQ9FtxgrE1x4uXwva6hpl3rWk2lxYSSTmG3tpbiO78394I42B+QZYqSc4J6ACtz4da3J4h1nwpDZ6nZXqJ++uLa3QiWzWOFlBkOTnczKOg5OBnk1d8Mw6UfD1xZW82lNaXEsrLb2EyvGkb9FPqe57ZPHAq9pdtbeB76HVNCit7W2keGLU18sYmgDY8zI6Mm4tkdRkHPGN6dSj7WzjbXczlGpyXuexUUA5or1ziCiiigArK17xDaeHrNZ7xZ5WckRwW0RkkkIGTgD2HetWvLfiHdazN44ttO0eW2j2aU8hFwGAO+UKSGXkEbFI4NZ1Z8kHIqEeaSQniPxBq2ua5pNpHHNpujSrMbqMygSzyBdyRsUJwuAxODzjB44OPaww23jy6jt4IoVbTISojQKBiWUHp65X8h6VWivru28LwHVZ43v9JvIYbuQHAc7lXeM4+9HJu/E1FqepyaX4omvbz7BYbdPCI13d4U/vhgsQvGcnAGc+3NeLUnUqt39DvjGMFoZuuJoms/ES40e6WOS8ksY4mTJU5+dt+V5yAyDHfcSeFr2n4f3P2v4d+H5icltOgyfcIAf5V5Rpuo2ejan4l1W+aFS93CokUgeaPs0TBFc4HOTgEjOa6T4e/EDRLf4d6ZbWMkuo3cKvH9ktky6gSMAXJwqcY+8R7V24SXKnfZJGFZXatueq5pM15nLq2ua7dP/AGl52mwL/qrWzuflkB/vSKAxbrkDA5H3uooeG7fVLvxLaXHhOC4j0yOf/TNRubyQwXKDIeOONmJc9fn4AI4J5FbxxMJz5IK/n0MnSlGPNI9cooHSiuoyCmyOsaM7sFVQSSTwBTq5b4lXs2n/AA41u4tW2S/ZjGGxnbvIUn8mNAGZq/j+4u9H1AeFtNuXukhb7Nc3aeVE7lcqVB+ZhyD0A7ZrhNZiRvBEWo3h+3XbrazTXVwN7yEyRs3J6DkjaMADgCrVjJr9j4mB124s5LS+TyYVti+I5kDMDhhxuXdxk8qKzdWubiXwPqVtFDFIlrNNbvKZgojCS/IAADuO3bgcV4k61SrJLpdbHoRpxgmXPiFd2Fn4QkOquscMlxAgkK7th8wMWx3IVScd+lbPwlitbPxVq8djCIIb7T7W8ESyM6g75lOCSfYccelc7qV5Drmo6ZGZtPvIbXWghS2fziu2GUjeMcNwDgZ6VpeHvHfh21+J08lzqcUS/wBlmEyFtwkZZVIVcZJb5mG3G7g8VeDvGSXqTWs4tntVGa891XxrquobYdE06eztHba97MVE4B43JEc49y3IGflNc9qkf+kxQW0Gqa1q7ruihj1CVXwDguzbwsag9+OeACeK7ZYqCkoR1fkc6oyau9D2Oisjwrb6ta+GLKDxFMk2pImJnRtwPJwCcDcQuATgZIJxWvXWYhRRRQA2RlSNmchVUZJPavAPhtEbvRF1d2DiYFYDj7qs7SOc9iXcqf8ArmK9a+JF/Jpvw08QXMBInFjIkRHXe42L+rCuM0XS4tE0Ky0y2AEdrCsQ9yByfxOT+NebmE7QUe51YZe82Y3jzxHY6D4auhfRySy3EDrbp9neRHkx8oJAwPmK9SK7n4VXGl2/gmw0KxR7W80uBI7uznAEqORlnIHVWYkhhkH6giuO8bgnwjOoXcDcW24Y7faI8/pUPibRdd1fUoX0e7g0owIwS+jZvtPP8ClcBVJ4OSfUAECubCV40kk9nua1qbnsbfj6TQ/EHjTTdKtj5Wp2DNcT39qu2RcIV+ziUc5Ik3FQchV7ZzUOlabp8XijQ/DtnZpHA0j38yJ02w4IZu5JlaLk9cHNZGnWVjN4JiSdpNPFoTI1y7KJYZlJ3SFzwXzkMTwSWByDXd+BfD97HfTa/q7MJprZLa1ikjCukQOWdgPutIwU7ewVe+RW0G8RX5ui/pEStTp26s7gDA4ooor1TjCiiigAooooAKKKKACiiigDn/HWsJofgfVb93CbINiuTjazkID+BYGvONP1m01+O60u4SMzxjbNERmOZCcb4z/GnuOh4PNdj8Tr42Wj6SHQyW8+qwxzqsZckBXZcAc58xU/KuQ1TTpfENqgEP2CZSTBdPzPCx43IFIx9C3sR2rysc05JP7zsw60bRzd1qsetQX2gadJYy3UYeH7dqivEvkgfKQ2MuwbcoIIBKFs1j6l8ULaS3024TWLIm6uIwbF7WRDZ4b78kucMMjkAdG46VVtIdMj/tKz8SyW+rXd0Wisr7W/3EcYVmZWj3ceW2/OIznIPQMMelS2Gl63plpBeva6gkOwgjayswGOnIwfSs5OjSS0uWueV9bGt8LLqbU9Z1q8F9balaRRW9tFfWibYp2HmO2Bk8rvUHk16XXnHgWb/hHdaTw3brFHo9xFJNZRKuDBKG3SRg91bcWAPI2sOmAPRxXp0JQlTThsclRNSfMFFFFbGYUUUUAFcT8WdZTR/h9eFpHRrllhxHKEdkzulCkkc+Wsneu1JrwrxBqtl4m/szxF4pt75rSJ1u7eFYi1vYpg4MuOp6bic84wAM5xrVFTjr1NIQcmUvFdtYa9DoVromnx6ndXDebbQRXAt1e125kBfIwCAAB/eA+hg02bwbd2sc2l6vcaReWsX2eJp7x99sqHGwq7MjICTxyvXFZmsxw6R46S8sZLuOy8t1hFguJI3kZVlEQIIYbmjYrjH7xiOlbMmm22r+GtL0vwTc2vl6flEvmOZrF1Ax8uAwYsTuGBkAg9cV5dkqcVd/p8zr1cmel/CGWe8+HsF/dzR3Fxd3VxLJNGMLJiVlBHttVce2K7muL+EySW/wAO7KyumU3lpLNFdBRwsvmMT+e4EexFdpXsRtyqxwvcKKKKoQUUUUAFcH8T9ebSbfSLaJ4hJd3TtslkCCURxO2zJ6Etsx74rvK8y8b6rJB8RrW0kglnRdJaS3WKLc29pcPz0HCx9SB+dYYh2pSZpTV5ow21TTtd8PnUo53tY7XdIZJEKzWki5+8g5BxnKn7ynHINc1d+JEurKz8QxRada2+mypNdJKXS5aRi0TqqYHA3HOck46cVreK9FuL6xutQt7waLcCEq81uhkkmXBAjYAhW5OBwxyflIzXOeHJNFkfS1uE06DWra5FzdahqUu27m3ZZhtbDF8nbk/LgZHUY8mjGnyuTO2bldJDYfH1vP4xhsj4r0og2j51GSyeGGIsyts2sw3H5OGz/Ee459u+GBuJfA0F3dqFkvLie4AUEAq0rbSM84K4YZ7GuK1Pw5oXiG483UrW3vWaExZbB+XOcgjkEZ4INdz4F1e9vbW80/VZo57vT5QqyqgUywMoMbsBwD95TjAyhwBXZhKlKUvdVmYVozS1d0dXRRRXonKFFFFABQelFBoA83+L/iCSLwbqel6VHK96Fhd5lVfLt18xWJfccEbVOVAPB5wDXm114O1HU9ae80bUNS0uK3gRC+oT+eL6QNHIXOHYKpZBkrx8oIyOB0c13ex/EO/vL+aN9O1i5ms4V/uSQsyID2+ZUk+uFFcn/wAJMND8QX2m6ffw3ehWdwkdxbM3mSQo6/cg28kIRJuBPyqvHSvNqVak5uMLafkdcYRUU5Evg/UPEs2pa7FbSaVd3KXIkuop5XVUfGwCNl3FlKxjkjgjHJzXqvwnurmPStS0jUbVbO5tL2SdYFl8xRFMxkBVsDI3+YOgxiuM1vwubeytrrwujWl7p6bY4oG2m4g3bngJOcZPIJ6H6mtH4UT3OoeNp9Q2G2gudHjla3a5adgrSfumdmPDcS/KoAGe/acLNSqXjsx1o2hZnsVFFFeocYUUUUAY3ijQI/EeivZmZra4VhLbXSKGa3lX7rgHr6EdwSO9eSxw6peakun3rNYa/pLiW4vY1x5bEMIzDxho3xkjpjKnk8e51w3xQ0dH8PHXIN8N1prI0s0LbWa0Ljz0J/u7NzexXIwa569LnXNHSS2Nac+V2exhaf471vxPG8dtcxaVLpz/AGe8+zIkwlnGclWfIEeMEcbskgkbeaeiLLdG8udW2XOpfbZkmuCgy6hzsA9Bs24XoKzbG4srDxhbW+nKsVhf2AjhEcbLFviJZVQ7QvKO54PO38763tvpevahBLKWM6pdpDGhkk6CNsIoLY+WM9P4q8itWqVLp7bnbCnGOwvhNt3huLaMKtxcoAOwE8g/pUXwn8VakPF+paRfWckenX15dmzuGfhpopG34HZShUfWNj1JrL8NXv8AZ+qahZ3v9qR4upbiOJrYmKCGVjIJHZV+XkuPnbjaeBXJaXqGqX1ukVnpV5YAXU1+NXM5DPAsrF1hGPlykzAc4OT3rehenUc2tGZ1LSikev8AiLxNF4o1e78P6bdlbCyIF80L7XuWOR5YI5EYIIZgeSNueGqtp3hP/hJFa1sJF0rSrKdUke2gUPM64YpGeiBflywBOcgYKmqGp2qWlhGNEggj1KSMWGnMoHBcjav+6Mbj14UmvUPDmjx6B4ftNNicyeQnzyt1lc8u592Yk/jW1G+IqOrLZbIzqfuo8i3NOiiivSOUKKKKAPJfjLI1z4i8KaZbzmKe4N2VKn5lyiRlh7hJHYf7tWo447a3WKCMJFEoVI414AA6ACqviKNdR+M9zcuNyaPpUMEQI4WSZ3dyPfaqj6Gr1eFj53q8vY9DDq0LnD+H/FHh68+MCa3NbTix0qz8h797Zoxb3ExwGlDAMF2LjcRgbsnA5r27XdV0nTfD897rDRNY7PmUr5nnZ6Iq/wAZbgADrmvJdOE//CQ+LDbwxvM01uEWVtquPs6Dk4PHXsaytI0PUtG17Tn1y6+1W7b1tbSAt9lsZPSNGJIJHIIxgCToMCumliowhy22Wn3XM50XKVzU8OaFpTWMOrCD5JJZLu2tpD+7sN7FtkadEK52nvkHp0r0P4bJHP4Rj1ZYjG+ryNend1KH5Ys/9s1jriDoVzf6teaZ4evZIJdQj33KgK0dmG63HqGbDYXozc8YY167YWUOnafbWVqgjgtolijQfwqowB+Qq8HFybqvrsRXaSUF0LFFFFeicwUUUhoA5fx14t/4RfS4ks0SfVb5jFZQOcLuAy0j+iIPmJ+g6muP0myWyseLj7XPO5muLs43XMjcs5xxz2A4AAA4AqBlPiTxBq+talG+2VpdOtImP+qtkYoxHoZGDNn02elcvPYw/C7TXvtOvFk0okNc6fcSKrMQAGeE8fN3KdD2xxXjYqp7aXs4vb8Tuow5FzM6bWbk6XcWepySlbZJPs9yP4QkhAV/+AuF57BmqDxZp9/Ppn2/Q8LrGn7prQEZEnGGiI7hh29Qp7UzxFrmmJ4d2XVre3sWp2xEcNtaSSs6Ov8AsjAOD3Iqr4X1y+/4QvRZb3T7+8neNoZnRFLIUcplwWBydvOM85rjjGSip22Zu2m7EVn8RtLmk/fLN9mMMcwu4I3lSMMPmWXap8plIOQe2Dmsv4mS6FaQLcvfW9vqc8O3ymDFL6Hr5cmzoD/C5xgjg8cadjdWmiTeLIpk8nTIWS58l+zzR5dAP9psYHq+BW1oFo+k+DdOs9V2q1rYxpcB8EKVQbs+uOau8KclOKf+ZOslZm78GvHy+NfCSpcSM1/YhUlZxgyxnIRz6n5SCR/Ep9RXotfOnw5tn8M6v4f1+GS6QeINQkgOmIB5aw3BZ0IGMgKEjY9gAelfRYr34TU1dHnSi4uzCiiirJCuA+JXiC30u60y2lZGkKTXaxFgGcIUQhR3bbK5A7kYrvz0ryzxPqRi+J19A1vNcSpplt5AiTOFaSXfljgDkL1Izj2rnxLtRka0leaMa+1awvvDqa9b3Hki3XfDIYz5kchwPKZRk5JOwpjJJGOQDXNar4olhhtPEljHp1nY2xMbxSuVuHeaHOCgA+VZDGGxlsoTwBWh400W5fTLzVLK+/si42hmW0iMj3bj7inkDfuxtYKWU4IOBVDQ5PD0uq6W9qmm2WoadLL9pubyULeXO5SMsrYYOSSW3dCTj1ryqUaag5PU7JuXMkivZ/Ea3n8R3emjxNpQ8uBHGqT2rxQy7S2YwhbO75xyDyAe/Ne7+AzM/gXS5rqFoJbiI3DRN1j8xi+09Om7Fed3vhnRNXvXvLm2hmuWClZwFLRlcbWU9iCB+Veg+CdZvtV0eaPWNjX1jcNbzSIu0SjAZHx2yjKSPXNduEqUpSfKrM560ZpK7ujpKKKK9A5gooooAKKKKACiiigAooooA8/+IHhW6FnqGv8Ahy/ubG+2rLdRRRLOtwiAAlUYHEmwYBHXABB4xxDDQBpEWqar4iur60lXckk16UWX2EcW0E/7OCRXuxGa4nXvhF4R168kvXsG0++lDiS60+Uwu4cEMCB8pyDySM1yVsMqjunZm1Oq4qz1PLNA8K219rFnq8Fp9m0yCDYoljCvqPysqSPH0ACuwyfmbOSBXRReCvDUULxHRbKaNnLYngWQqT1ALAkD2zgdq6o/C1kiWK18Ya/CijCqRavgduTBn9aqy/C3VvJdIPH2rqWGAZLW2bB/CMGuSeExEn8X4s3VamuhwN9plrqXiiTQ9MtbeK0tLVY5IBFiD5tzMpUYGQDF8w+ZS4IrDu9Z1Hw7Z3OgazBcavbT/ubCNstcxyEfJHvXG5gSMMMEjlTkFRtaLDP8OZLjRda0rVNQ164meVZ7OBp/7QQsSHVuigbuVOME9813/hLwVe6hr9v4r8ZWMNre2sZTT9PVxKbUMcmR26GQ9MDhR7nNFOhN1OWS91df8glUio3T1O18NaUuh+F9M0tFVRZ2kcOFHGVUAn861KAMUV6xxBUdxAlzbyQyjdHIpRh6gjBqSigDyPQHRvDun+WwYR26Rtg5wygKw/Agj8Kp+KJZbPw88Ons0M99cRWgnXrEZpAhk+o3Z+uKvXOnv4Y8Z3+myjZp+qTPe6c56b25mhz2O7Lgdwxx0OKvix9N/wCEdu7XVtSh02O4iKpPLIFKMOVYZ6kEA8elfOVKbp1+V9z1Iy5oXRX1PwToepaTb6e1nHDHbvE8bxxgONmMc454yOc++afrlt9ss9O8MRu7y61MtkXCqrLCBumfCgAYjVhwAAWFc/dfE+38NlbPxjaNFebNyPYOJo5x03DkFckHhh+Nei+BvDmp3HiA+KvEVmdPlFsbaw093DvDGzBmkcjhXbCjA6AdTmujD0Kk5pz2WplUqRjF23PQYYxFEsa52qoUZOeBT6KK9s4AooooAK4fxnAieJ9Lu8APJaXEGf7x3xOB+Qau4rjfiMqQ2OkX8siRR2moqJHdgAFkR4uSf9p1/KscRHmpSXkaU3aaZwviLwlDr/m/6bLZidQJlSKORZGVWVHw6nayhzyMHp3AI5PSvDmvXHjC+uBNoazabmGS6OnNIL2SVdzNIpbllVh7DcR613+oazpmlW/n6lf29tH2LyDLewHUn6VieGdW086vrNgt0pupL+ScKTgOCFXCnuV2gMOoPsQT4lOrVhB2/I7pQg3qWtD8KWemab5F+sOpXDtulmmt0AOFVQFXGFUKigD2rN8NmDwnJrNheWxsYHvmuLONRu85HIUCNV5Jyv3RyNw4ren1/TobwWUMxvL8/dsrMedMf+Ar0+rYHvXVeGPCM/8Aay6/4hjQXkSGOytAQ62an7zbu8jAAEjgAADPJOlClUrN8+z/AK0JqTjBabmNbeBtS8VTwT+JC+m6VG29NOhkImnOCP3zqflXBP7tc+7dq9ItraK0t44LaNYoYlCRxouFVQOAB6VKOBRXs06caceWKOGUnJ3YUUUVoSFYvjCxGp+C9YsyMmaylVR77Dj9cVtUyRBJGysMhgQRQB5Vd2sWr6fH+8kjyUnhmiI3RsCGVhnI/AgggkdDXnXiTwZf2EFvp9jdWN9FqlyIEW8s1D20rIA88Xl7V3BIs5IzkE5JNd5oVzBHoMEElzEXsQbWYmQfK0RMZ3c8fdrE1nxJojazot6uoRTW1ndyrI8LbgrGIqTx1VQ/zFc4yCeM4+epSq05uMelz0pKMo3ZNp3hG5j1K2uNZvLK5htIljitrWwWFHKoyK8hySxAdvbmma9o1vp/iTQdbsNMiijtJZFu54EVNkbrtywHJGTnPbFdDdavpthaLdXuoWsFuwykskyhW+h7/hUljpF341g8iGK4stFm4uLuVDHJcp3SJTzhhkFyBwTtznIdL21apf5eQS5IRG2q6p4laSDwuI0iUlJdVnXdDGR1Ea/8tW+mFHc9q7Xwl4O0/wAI6cbeyM1xPIB595cuZJZiOmWPYdgOB2rbtLS3sLSK1s4Ugt4UCRxRrtVFHAAHYVNXs0aEKKtE4alSU3qFFFFbmYUUUUAcL8Vbrboek6djcNS1i2idfVEbzm/SL9azTV34hR/afE/hmI8rAbq7x6lUWMf+jqpgE9AcV4eYyvUS8jvwy925znj+b7P8PdZmzgpbEqfRsjH64raF0kOmi7vZVijSLzJXbgKMZJrzn4keK7G91Gx8IWly0hu5Qb42sRuHVRysaovVywBx2wM16h4U8PXfiWSHVPEOnSWGmQuHs9MuVxJKwPyyzL2wRlU9eT0GJp4WdSMVt/kVKtGLZD4H8Hz6lfS+IdbjkhsJLj7TpulyDbtJUDzpR/eJG4Ifukknnp6hikAxS17cIRguWJwSk5O7CiiirJCiiigAooooAKKKKACiiigDlfiIsSeFlup8BLW9tpSx6IPNVSfwDGuckRZI2STlWGGHqPSu/wBb0qDXNBvdLvBmC8geF8dQGGMj3HWvLtAlvBbPpmtHGr6cRBdjGPMI+7KuequMMD6kjqDXk5hTelRHZhpLWJR1hBqXizStHmQGyihk1CVGUFZWRlREx6Avu/Balm8I6W3iK21m3jFrdW8RiHkxoAw3Z5BBHrz154IxVXxLcp/aentpV7YnW7R3aKynnCmeNkO9DjkcAMDjGVFZtl8TrbVNYTQtL0e8ufEDuY1sFePYHAycy527QATu9B0rijCrJL2Z0OUU3zHY+H4DrfxOhjRmEGgW5uZSD96eYFI1P0TeSP8AaWvUxXL+AvDM3hrRZ/7RlWfU9QuGu72RB8odgAEUnkqigKPpnjNdRXu0afsqaiedUlzybCiiitiAooooAQ148ml2VxqWpQXsP2g6fqFxCkMrFo0Vn8xfk+7nbIoBxnHSvYq8y8S6eNA8dNeA7bLxBgNnnbdxrj/x+Nfzj9648ZFypNrob0JJTs+p5J4y0S5fUtH8ONm70sXi21tL9oZHg80hihA+/sjRwDn7sg7iuzsPBi6B4im1DwxLBZW1wirPYyQblcrxlZM7kyMZ6jPJBqXxlpF5eW9jquk+UbzR7n7YsMp2pMAuGUt2O08E+lVk+JXh+dLCPT5JbjUNQKLaWJiZXd3OFUkjAGTgnOPrXmc1WpCKhr3Ou0YybZ6H8N7fC+ILwMSt1qr4HYeXFHGf1Q121YvhHQf+Eb8L2mmvL50yBpJ5f+ekrsXdvoWY49q2q9yEeWCj2PPk7tsKKKKskKKKKACuF8bRww+KtIuHISSe1ubdSTjcQ0ThR+Ac/ga7quT+Img3Wt+G1m0pA+qaZOt7ZKTgO65DIT6MhZfxFZVoe0puK6lwlyyTOdaNXZN4zsOVB7H1rnrG2i1rxVq93qEKypYOthbwyoGCjYsjvz3Yuo+iD1rb0++g1LT4721YmJxkhhhkPdWHZgcgg9DXKahra6Z4hvtV0O4stUg8hBqdml0qyQlN2JFOdpO3IKkg/IK+dpxldx6npya0ZtaR4WsNF1a+vtNBhN9J5kkKooTOB04yOQTgEDnpxmui+GNuNRu9Z8Tgt5d3MLG1GflaGAsN/vmRpOfQCuF0bxZL8RjPpngW0uTJsC3V/cgRx2SvkBsAks3Bwo6kda9s0DR7bw/4fsdIsQRb2UCwR5xkhRjJx3PU+5r1cHRmpOpU32OOvUi1yxNGiiivSOUKKKKACiig0AeF6Zp8ut+BNPtbvMMzMrXEuMSRSpIS5X0fepGe3J56VhfD/RxceJfEOqXrfbI4XOnWsk8aBjEpO/IUAc5Hviu4hWWz8UeItNnG3yb9rmDP8UU4EmR7eYZR9VNU9Lht9J1bULMLFbrdzC6gXIHmFlAfA7ncuT/vD1r5+pKVN1IL+tT0opSUZGh5EdppLW8LyRxxQlVbzCWUYOPmOTkepqD9nuwD+Cjrjbs3iQ20e7+5Am0n8ZGl/Ss7xrrMelaBNaRvnU9SU2lhbqfnllk+RcD0BYZNereFdCh8M+E9M0W3ChLG2SElRwzAfM34nJ/GuzL4OzmzDEy1SNeiiivUOQKKKKACq2o2UepaZc2NwMxXMTwuPVWUg/oas0UAeOaVE+o+G4LfUpC13at5E8qcMk8LFGceh3Ln6HuDXL+I/DmsJDJfR6jZWC2I82G5s1aJy+GDFoh8rO4Kpktgc4XJru77TjonjjVrbBFvqRGo25PTcQEmUfRgrf8AbWsjxfb3U/he4exVXmtXiuxG3SQQyLKV/EIa+flejXcVtf8AM9JWnTTMix8ELdsbjWzNbxy2ywvpttqE7ocklvNkLZlOSePujnrnNdclvDGsSxRKiwpsjVRgIvAwB6cD8qy7Lxd4f1KziurfWLIJMgcLJcIrLkZwQTwfarFlcN4puzpvhi6SUdLvUISHjtFPow4MhH3V5x1PAwYtWrz5Wv8AgDvCEbmx4W0yPVPE8N4kKmy0cOsbA/L9pZduF/3ELg9svjqDj0QVU03TrbStOgsrCLyreBAqLkn8STySepJ5Jq3XvUaapQUEefOXPK4UUUVqQFFFFAHkGnXP9oa94l1BhzNq8sKt6pCiQj9Ub8606x/Cif8AFM20x5e8aS8Y+pmkaT/2erer6tY6Dpst/qtwlvBEpYliAW9l9Sewr5mu3OtK3c9SmuWCMbTrlm+Jmu2qn5FsbR2GP4syf0Iq7rNve65eWug6BxqbSR3JuSuUsY1bmVuxzgqF/iyewJri/hzqureJ9R1HVNI0y5utR1m4CmWSBks7GOPIXfL/ABkKRwvJwBx1r6B8N+HLbw5p5ihZp7mZvMuruQAPcSYxuOPYYA6AAAV6FLCN1Oaeyt89DnnWSjaPUd4d8PWnhzSxaWm+RmYyTXEp3STyHq7H19ugGAOBWtRRXrWtscYUUUUAFcD8VviNB4D8OlbZkl1u+Ux2NuWAw3TzGz0Vc/ieK7414T4u8IQD4/x6rr9zJdxXVkJdKinPyJJGfnRf9374H+0T2qKk+SDl2KjHmkkdDptvLa6TZ29wQZooESQg5ywUA89+e9Utb8MaX4iUR6pbq8f8e1QGfHQFuuB1xnGetZfih9d0vU49Y0SNri3EXl3FsAzhyGyCUGSOCfmTJHdWHTmJPjPPGhRPB+o/ahIsPltJtBdvuqMpkk44GM18/ClVk+emek5RWkjodH8WadofgXTX1i6EM1vG1m0YVmZpISY2+UAnHy8noMitXwVBPD4J0hbnPnvbLLJkYO5/nb9WNcVoOlaZ8Q7O71zSdTv9E1S4jeHULKwuSkZck/6xcZYMMZxgHnvms7xJca5eai+meK7q50rzZ0CJpepK0bQ4wUS3VTK7sc4LYAyM8Dnf2MZtwTs93/wO5nzuKu1obvhO2vvFXiLVtW1IKNFj1RpbKLb/AMfEkYESuT3VQgIH94k9q2fEOv6Xqum3fh+y1KMaleuLD7PyJY952u2084VNzZ6ce9cHLd+MPDPgaFb211zQ3heGHTmNvCtmY9wXM5K7kY/MTk85HFa+jWt3441S9ikAn8XRz+RBe6dLm10yIAESLIpIKkM3BJZjlcDttLCz5+Z7dLfr+pCqxtY9P8CabFq3iubVEj/4l+hobCxwflaYgCVh/ugLGPffXptZfhzQbTw14ftNJsAfJtk272+9Ix5Z292Ykn3NalelTgqcFFHJKTlK7CiiitCQrzvxVbfZ/iHHcYwb3TAi89TDKc/+jlr0SuL+JWjXV5pNnrGlxyTX2iz/AGkW8Qy1xCRtljHuVOQO7KorDEU3UpOKNKcuWaZimNWlR25KZ2+2e/1xx+J9a5rRILfXdX1rVNRto5ZY7trCJZUDeXFF6A9CzFmP1HpXRWt1b3ljHeW0qyW8i7lkB4x/SuOufECaJqep6rpL2eqaMzJJqKW9yvm2kgUKXUE7WBAX5cg5FfP04yd0tz0ZWVmzd0Pwzp3huS7fTS0MVzI0zxELsQk54OMgDpjOPbPNdV8LbeS60e98RTbh/bNyZIFJOBboPLiOPVgC2e+4elefaTrU3xVgudK8H29xFZOBFf6pcbYxbxt95UXks5XOOgHUmvdNPsoNN023sbOMRW9tEsUSD+FVGAPyFetg6M4t1Km7OSvUi7RiWKKKK9E5QooooAKKKKACiiigAooooAKKKKACiiigAxRiiigAooooAKKKKAPP/jBZ2OqeF9N0vUIxKl7q9tGq9+CXbBHI+RWGR615B4m+Hfhrw/b6bd29tI8v9qQK0lxKX+TLHYQeCM46jPvXv3izwhbeLLe0E93c2VzYzefa3FuVzG+0rkqwKtwSOR37VxeqfDHxNqbW1jda/ptxYxypO1w9gyzh0YMo2q+wg85PH0rjrU6sqicHp1N6coKLUtzyHx/INQ+FWgxjDSy37Rp/vBpEwPx/lX1fBH5VvGhbcVUKWPfA61wfhn4PaBoVvZLfy3OtyWLtJbm+IMcLsSxZYwAucnqcketegAYFbUabpx5fUzqS5ncKKKK2ICiiigArzv4rwrrkWkeF5l32eoTPcXygkFoIQDjI5GZHi/I16JWH4k8KWXiOOJ5ZJrS9tw32a9t2xJDnGRzwynAyrAg46VFRScWo7lRaTTZ4bc/DjwXobW+zTfMkMhmPnyNK3lxguwGTjkhV6dGNYXxH8N28XhzR7G2tY4tWupWkcw5BkkbqvHYNJkE9kwDzXouteBPHMmsR+RFpmoxrbtBHd/aWtlUMwJZ4trHdgY+VsYJ4FdXp/wAM7WfxBb694nlW/v7Uo1tBECkFsynIYDOXbJ6scegFcNGjWU05vbzOidSHK1E2vBng7S/Bnh2103S7WKNo4lWadUAedwOWY9SSc10OMUUV6JyhRRRQAUUUUAFUtY1GPSNFvdRn/wBVZ28k7/RVJP8AKrtQXlnb6hZzWl7Ck9vOhjljcZV1IwQRQB89wfDDw7dWqa14otJJdRuEN3qDee4RpGy7/KD2JI49Kdp3hrRbHSbu4k0u2WztLR5dgXB82T962G6/KvlqpznrjrXc+JPh3rtvpN3beFb5bu0nTy1sr2UpLCp4Iinwe3Zwf96odK+HWvatpc1p4mkttJtJ5g729jKZpigCgR+YwCqPkHIBOO4ry/YYhv3n17nZ7Smlojl/2efAFh9n1HxLqdql3MbjyLKSZNwVVALSKD03MfqMY9a9+AxWfoWh6f4b0W30rRrcW1nbgiOMEnGSSeTySSSefWtCvUOMKKKKACiiigAooooA828XXkc/xKs7V3WN7TS5GRWbBmM0i52jvt8gZ9N4rzr4uaNFfaNZy+fPHdT3kFlCIyqg72OcnG48Z4zivePEHhjSPFGn/Y9bso7qIHcjHKvG395GHKn3BFeX638GvE097pa6b4siu7GxvVuo11a3LywlQQuGTHmYz0bH1rhqYebre1izojViocjRy3wX8IaZY/FZ7q3QuILa8ltmf72wTiBCcADOFk5AGd1fRwFch4D+HOneB7TMU81/qDRCGW+uD8xQHIRQOFXJJx68kmuwrsimlZmDab0CiiiqEFFFFABRRRQAUUUUAFFFFABRRRQAV458TPDOk+LPiVDa6kjlrTR1ffDIUdd87beR/uNwcjk17HXG+Ivh9/auvSa3pOs3OlajNEkMx8tJopUQnaCjDI+8fusOtZVozlBqDsy6bSleR4PqfhrRvB/jmI6Tb+QBpnmEs5YlvNO5snocDHH5V0lna/av2o9BMPD21nI8pHXaIpBz9S4FdPqHwc1fxLqLy+I9dtII0j8iNtNtWWSSPcGyd7EI2QRgBhg13nh/wNovh6+l1G2ikuNUnUJNqF0++Z1HbPAUcDhQB7VhRozjJTm9bGlScWmonR0UUV2GAUUUUAFFFFABXE/E82l3oNlpFz5nm6jfRrC8L7XiMeZS6t2O2MjP+1XbVh+JfCtn4lt4RPLPa3VsWa2u7Z9skLMMEjscjqCCDUzTcWluOLSabPAfF/hKTTPDktzqPiPWdXXzIokt7mf9180igkqMbsLuPNReH9Cg/wCF8eEtAs7WKNdIsobm7ZV58yON2yT7sy/pXW+NPhx8RdS06PTLC70fU4EnEv2qQNbysNrLtZBlf4jyMdBwK6zwD8Mrnw94p1Dxb4i1CO813UovKkS2TbBAny/KufmP3VGTj6Vy0KVSNud9/wBP+CbVJxd+U9GHSiiiuwwCiiigAooooAKDRRQB8/ah4E8N+LPFPiHVLuCUltXmjPkylFcIqowIHB+dWORzz1rB8H2en6F46fTtNiWFH1K7i2g5IAiyq5PJACk8165P8M7zT7i4l8La+9rFPNJcNZX1us8XmOxZiGBVxkk9WPWsXTfgrcz6zFrPiDXTBeCd7hotIj8tQ7oqMN77iRtX0B+Y+1efKhWlKSb0ex0qpBJWWqMb9nyDd4x8e3aHETXiJtHQtvmY/kCK92rI8PeFtH8K2DWmg2SWkbsXkIJZ5WPVnY5LH3JrXrvSsrHO9WFFFFMQUUUUAFFFFAHm3xa026htdN1vQ5VtNUF1FYPcMm9PJlbGHT+LD7SO4yfUg+X+JvBlxqslo/i7VZdUlSC4dEhQW8UJGzG1V5J55JPOB6V9C+I9FHiDQ5dPM5ty7xyJKF3bHjkV1OO/KivM/FPgLxveX1ouly6PcwBJYpJXaSArvAAJX5+hGeDz7Vx1qVR1FKH/AATenOKi1I89vvBum+Hfih4POm+asc+qWoKyztIWZWck/Me3y/5NfUQ6Vw2i/DiOPxJb+IfEk8OoalapstI4oikNrnqygklmP94/gBXcjitqMZxppT3M6ji5Nx2CiiitiAooooAKKKKAOL+K1q58AajqdozQ3+lRG8tZ1GTGVHzcdwV3Ag8HNeX6j4c1W50f7R4j8QvemN4mC20P2ePb5i5yAcNkdyOO1e/3VrDe2k1rdxrLBPG0ckbjIdSMEH2INeea58JXvbGez0PxNqGnW00ZjNvcKt0iDGPlLYcY7fNxXLXoym04WNqc1FNSOBn8E6HL4OF1dacst1b6ftOGYAmNcZ2g4J+X0rrv2d7Yw/D/AFGRbdora41eeW2crgSx7UG4e2VIH0rQ0j4WX6KbbxL4j/tLTyzs9nb2f2dZdxJKu29m25J4BGenTivRLe3itbdILaNIoo1CpGigKoHQADoKMPSnTvzu4VZxnayJKKKK6jEKKKKACs/X9R/snw7qOo+W0v2S1km8tRkttUnH6VoUEZ60AeO6HZRf8IdYWMNylxEtkkHnxsrq+ECkg8g/rXi994Q0671vV2M9xPFbasbVBK4I8qKF5p+gGOVAG3GM19G698LbW7v31PwtqEvhzUZG3TG2QPBcH1khPyk+4weec1yPhv4E3y3TyeMNaint2uri4e105GjFz523d5jsc4IUDaB0yNxya8+lhqlKTaludMqsZpXR6B8MNMh0n4XeHbSBQqiwikb3Z1DsfxZia6uora2itLaK3to1ihiQJGijAVQMAD8Klr0DmCiiigAooooAK5X4g+EG8X+GjbWk62mp2sq3Wn3Z6QzL0z1+U8gj0NdVRQB4Hf8Aj5fD+lzxeKbKSw1+2QD+zWH/AB8ucANE3RkJPUHI5z05wLKaHQdOk8a+NmWS8iZlgiTbh524fy+vPGwHsEPVQtfQniHwtovivTzZeIdOhvoM5USLyh9VYcqfcGvP5v2dfBtzfRS3VxrNxawjEVhLfboYx6LxuA/4FXD9Tgr8rsnv6djo9u3ucF8MPhg3xIur3xz4pkutOgvpiLS20+TyPNQcFmIGSvGMjBJBJr3bw94J8O+FkA0LSLa1fGGnC7pX/wB6RssfxNbFnaQWFnDaWcSw28EaxxRoMBFAwAPwqau1JJWRg3crX2n2mp2MtlqNtFdW0y7JIZUDK49CDUWk6Lpug2K2WjWFvY2y8iK3jCDPrx1PvV6imIKKKKACiiigApDS0hGRigD50tPh94W8Uy3mtT2smy61K8fZHMyxyJ9ocLkD2A6Yqn8NLS00zxlr+nWEYjtwHCoOfuzEYPrjpXp//Cr9R0O3aLwjr4S2Xc0NhqdsJY1YknAkQq4GT1O4/Ws3Qvgg1trUmr6x4guBcz7zNBpi+TGTIxZl3tuYjLHBG08CvOlh603NSej2OpVIKzSMf9mW12aX4lukP7qW8iUY6bhHub/0MV7nWboPh7SvDGlpp2hWUVlaoc+XGPvHuxPUk+p5rSr0TlCiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAPSvOdY8Ra5D4yn0K3uZIlmuI0hmEanyxJ5bDt2CT9f8ACvRjzVd7C1kuhcyW0LTqQRKYwWBGQOevG5vzPrW9GpGm25K+n4mFanKokoux5bF461u7s5LyKZ0WxEYmjEa/vDJsjB6dnEh4rbtNc1rUfAY1qO6kj8y4M0vlKhljtl+VggYYJBUnkdM+1dkujaaqOi2FqEk271EK4bB3DPHOCSfqc1HL4f0ieMRz6XZSIowFe3QgDJOMEepJ/E10yxFF7Qtr+H/DnLHDV1vO+n49zi5vFd9/bF0ft9x9gtZBIogjj3SofJAB3D7o3knHPzVT0rxbr1xrAtby4ZVWWIA7F/eI92Ez0/ulk/D15r0N9E0uSRJJNOtGeN/MRmgUlWwBuBxwcAc+wpJNC0mZCk2mWciMuwq9uhBXdux06bucevNCxFHltyfkDw1fmvznAeJPGmp2viLVtP0y8XfGYUtoQFJVw8W4dP4vNI/4Can07xhfahcWdz9pufsbSb/LhSPdIsl3JEm7cM7QFRfl55rtxoGkKkSLpdmFhbfEot0xG3HK8cHgdPQU4aLpgkhcafah7di0LeQuYyTkleOCSSeO5pfWKHIoqHz+VhrD1+dyc9O3zueeX/jDV7fw2866hsne1i2ysi4R2uZELdP7i/pT5/GuoTWN1qEdzcRCWCRLaKJIzHE62qzbmyNxOScY44/P0CLRdNhlaSHT7WORpPNZ1gUEvz8xOOvJ59zTF8P6QpJXS7IExeScWycx4xs6fdxxjpT+sUP5PyJ+rV/5/wAziW1nXrU3Dtd3x+zaYk5EywEI7tIMvtHOMAjHHy81Dd674gt9eXTLTU57kQ3RiWZo4gHybfHmYUcZlZflwckV6Bb6NptpA8NpYWsEcibHSOBVVl54IA5HzHj3NNi0PS4I1jg060iRfuqkCgDkNwAPVVP1A9KSxNNO7gvuRTw1W1ud/ezz6fxfrMFjp919rZ1CQy3ChF+dfMn3jp3WMD8K07HXNXvfDehJcXrWt7qF/NBcTIiFowrTEKoIx/yzAyR0BrrhoWlL5m3TbMea++TFuvztgjceOTgnn3NEmhaVMjLNplnIr/eD26kN8xbnj+8zH6knvRLEUWrKFtfII4esnrPp5nnGteLde06OG5t9QeU3M0wNuY0xGIrhEAHy55UkHPr2rY0jxTdXujXOqXNy/lW095chcAboUQBEOP8AroD9QK67+wdJ855f7Ms/NkILv9nXcxByCTjnBANKmh6ZHG8aWFusci7XjEQ2MM5OV6HoPyFEsRRceXk1+QRw9ZSvz6fM4nRfEmp3NlpN5eam0loXltp5oEQh59+I93GQpXGCuOcZro9FvbvUNN8PyTXDGSW2+0zsOPN+QDBx7yA/8BrTbRNMe6Fy+nWjTht4lMC7g2AM5xnOAOfYUtlo2naa7Np1lb2m4YIgjCA/gOOw59h6VnUq05J8sbfd5/5mlOjVi1zSv9/l/kXqKKK5DsCiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKAP/2Q=="}}},{"cell_type":"markdown","source":"<center><div class=\"alert alert-block alert-info\" style=\"margin: 2em; line-height: 1.7em; font-family: Verdana;\">\n    <b style=\"font-size: 18px; color:green\"> &nbsp; DO YOU KNOW WHAT THE IMAGE ABOVE IS SAYING?</b><br><br><b style=\"font-size: 18px; color:green\">HELLO EVERYONE!</b><br>\n</div></center>","metadata":{}},{"cell_type":"markdown","source":"**I did my bachelor thesis on Continuous American Sign Language Translation from videos.** **Although this competition differs from that slightly,still I think this competition will be the perfect place to use the experience I gained working on my thesis.**\n\nIn this notebook I'll try to gather some important resources on Sign Language, discuss some of the conventions of Sign Language, explore the dataset and try to visualize them. **If you like my work, your upvote will be appreciated**.","metadata":{}},{"cell_type":"markdown","source":"<div class=\"alert alert-block alert-info\" style=\"padding:10px; line-height: 1.7em; font-family: Verdana;\">\n    <center style=\"font-family: consolas; font-size: 32px; font-weight: bold; color:purple\"> 🤟Table of Contents🤟</center></div>\n\n1.  [ASL Fingerspelling](#1)\n1.  [Dataset EDA : Phrases](#2)\n1.  [Dataset EDA : Sequence Landmarks](#3)\n1.  [Visualization ](#4)\n1. [Complete Function for visualization](#5)\n1. [Evaluation Metric](#6)\n1.  [Resources](#7)\n\n\n\n","metadata":{}},{"cell_type":"markdown","source":"<a id=\"1\"></a>\n<div class=\"alert alert-block alert-info\" style=\"padding:25px; line-height: 1.7em; font-family: Verdana;\">\n    <center style=\"font-family: consolas; font-size: 32px; font-weight: bold; color:purple\"> 🤟ASL Fingerspelling🤟</center>\n","metadata":{}},{"cell_type":"markdown","source":"*  Fingerspelling is the process of using **specific handshapes to represent the letters** of a spoken or written language. \n* In ASL, fingerspelling is done to spell out **words that don't have established signs, such as names, places, or new concepts**.\n* Fingerspelling expands the vocabulary of ASL users by enabling them to convey words that **do not have specific signs**","metadata":{}},{"cell_type":"markdown","source":"<center><div class=\"alert alert-block alert-info\" style=\"margin: 0.2em; line-height: 1.7em; font-family: Verdana;color:black\">\n    <b style=\"font-size: 15px;color:green\"> In short, some special words such as names, addresses, numbers, URLS etc. that doesn't have any specific signs are expressed by fingerspelling.</b>\n</div></center>\n\n","metadata":{}},{"cell_type":"markdown","source":"<span style=\"color:purple;font-size:20px;\">Some important things to note about ASL Fingerspelling : </span>\n* ASL fingerspelling is **one-handed**. Some other languages like **British Sign Language(BSL)** is two-handed. [[source](http://https://en.wikipedia.org/wiki/Fingerspelling)]\nThis will be crucial since we have landmarks for both of the hands in the dataset and how we approach the problem.\n\n* For word based signs, facial expression plays an important role. It will be interesting to see how much information facial expression contains in fingerspelling.\n\nHere are the ASL Fingerspelling alphabets. [[source](http://https://www.quora.com/What-are-the-differences-among-fingerspelling-and-sign-language-and-Braille)]\n   ","metadata":{}},{"cell_type":"markdown","source":"![main-qimg-3a270c09a578b9afab3dc79d9b546562-pjlq.jpeg](attachment:247f6d6a-0455-4823-a928-dd0b3a008707.jpeg)\n","metadata":{},"attachments":{"247f6d6a-0455-4823-a928-dd0b3a008707.jpeg":{"image/jpeg":"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"}}},{"cell_type":"markdown","source":"<a id=\"2\"></a>\n<div class=\"alert alert-block alert-info\" style=\"padding:25px; line-height: 1.7em; font-family: Verdana;\">\n    <center style=\"font-family: consolas; font-size: 32px; font-weight: bold; color:purple\"> 🤟EDA : Phrases🤟</center>","metadata":{}},{"cell_type":"code","source":"import pandas as pd\nimport numpy as np\nfrom tqdm import tqdm\nimport matplotlib.pyplot as plt\nfrom matplotlib.animation import FuncAnimation\nfrom IPython.display import HTML","metadata":{"execution":{"iopub.status.busy":"2023-05-18T20:45:25.606375Z","iopub.execute_input":"2023-05-18T20:45:25.60679Z","iopub.status.idle":"2023-05-18T20:45:25.613958Z","shell.execute_reply.started":"2023-05-18T20:45:25.606757Z","shell.execute_reply":"2023-05-18T20:45:25.612784Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Let's load the training and supplemental_metadata dataframe","metadata":{}},{"cell_type":"code","source":"df = pd.read_csv(\"/kaggle/input/asl-fingerspelling/train.csv\")\nmetadata = pd.read_csv(\"/kaggle/input/asl-fingerspelling/supplemental_metadata.csv\")\ndf.head()","metadata":{"execution":{"iopub.status.busy":"2023-05-18T20:45:25.706099Z","iopub.execute_input":"2023-05-18T20:45:25.706534Z","iopub.status.idle":"2023-05-18T20:45:26.021034Z","shell.execute_reply.started":"2023-05-18T20:45:25.706502Z","shell.execute_reply":"2023-05-18T20:45:26.019847Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"metadata.head()","metadata":{"execution":{"iopub.status.busy":"2023-05-18T20:45:26.023197Z","iopub.execute_input":"2023-05-18T20:45:26.023893Z","iopub.status.idle":"2023-05-18T20:45:26.037247Z","shell.execute_reply.started":"2023-05-18T20:45:26.023853Z","shell.execute_reply":"2023-05-18T20:45:26.035863Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"* **path** - The path to the landmark file.\n* **file_id** - A unique identifier for the data file.\n* **participant_id** - A unique identifier for the data contributor.\n* **sequence_id** - A unique identifier for the landmark sequence. Each data file may contain many sequences.\n* **phrase** - The labels for the landmark sequence. The train and test datasets contain **randomly generated addresses, phone numbers, and urls** derived from components of real addresses/phone numbers/urls. Any overlap with real addresses, phone numbers, or urls is purely accidental. The supplemental dataset consists of fingerspelled sentences. Note that some of the urls include adult content. The intent of this competition is to support the Deaf and Hard of Hearing community in engaging with technology on an equal footing with other adults.","metadata":{}},{"cell_type":"code","source":"print(f\"Total Training files : {df.shape[0]}\")\nprint(f\"Total Signers in training set : {df.participant_id.nunique()}\")\nprint(f\"Total unique phrases : {df.phrase.nunique()}\")","metadata":{"execution":{"iopub.status.busy":"2023-05-18T20:45:26.03888Z","iopub.execute_input":"2023-05-18T20:45:26.039305Z","iopub.status.idle":"2023-05-18T20:45:26.076733Z","shell.execute_reply.started":"2023-05-18T20:45:26.039263Z","shell.execute_reply":"2023-05-18T20:45:26.075529Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"print(f\"Total files in metadata: {metadata.shape[0]}\")\nprint(f\"Total Signers in metadata : {metadata.participant_id.nunique()}\")\nprint(f\"Total unique phrases in metadata : {metadata.phrase.nunique()}\")","metadata":{"execution":{"iopub.status.busy":"2023-05-18T20:45:26.079215Z","iopub.execute_input":"2023-05-18T20:45:26.079902Z","iopub.status.idle":"2023-05-18T20:45:26.096061Z","shell.execute_reply.started":"2023-05-18T20:45:26.079861Z","shell.execute_reply":"2023-05-18T20:45:26.094861Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"**That means some phrases are repeated!**\nMetadata contains much more repeatative phrases!\n\n","metadata":{}},{"cell_type":"code","source":"train_unique = df.participant_id.unique()\nmeta_unique = metadata.participant_id.unique()","metadata":{"execution":{"iopub.status.busy":"2023-05-18T20:45:26.098393Z","iopub.execute_input":"2023-05-18T20:45:26.099261Z","iopub.status.idle":"2023-05-18T20:45:26.105918Z","shell.execute_reply.started":"2023-05-18T20:45:26.099225Z","shell.execute_reply":"2023-05-18T20:45:26.104905Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"both = []\nfor x in train_unique:\n    if x in meta_unique:\n        both.append(x)\n\nprint(f\"Participants present in both metadata and training set :{both}\\nTotal {len(both)} participants overlapped\")\n        ","metadata":{"execution":{"iopub.status.busy":"2023-05-18T20:45:26.187974Z","iopub.execute_input":"2023-05-18T20:45:26.188409Z","iopub.status.idle":"2023-05-18T20:45:26.196095Z","shell.execute_reply.started":"2023-05-18T20:45:26.188376Z","shell.execute_reply":"2023-05-18T20:45:26.19499Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Now let's look at some phrases","metadata":{}},{"cell_type":"code","source":"df.phrase.iloc[:20].tolist()","metadata":{"execution":{"iopub.status.busy":"2023-05-18T20:45:26.243252Z","iopub.execute_input":"2023-05-18T20:45:26.243659Z","iopub.status.idle":"2023-05-18T20:45:26.251956Z","shell.execute_reply.started":"2023-05-18T20:45:26.24363Z","shell.execute_reply":"2023-05-18T20:45:26.250807Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"The phrases in the training set contains **random websites/addresses/phone numbers**. ","metadata":{}},{"cell_type":"code","source":"metadata.phrase.iloc[:20].tolist()","metadata":{"execution":{"iopub.status.busy":"2023-05-18T20:45:26.320501Z","iopub.execute_input":"2023-05-18T20:45:26.320936Z","iopub.status.idle":"2023-05-18T20:45:26.328571Z","shell.execute_reply.started":"2023-05-18T20:45:26.320904Z","shell.execute_reply":"2023-05-18T20:45:26.327494Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Whereas the **metadata phrases are mostly normal sentences!**","metadata":{}},{"cell_type":"markdown","source":"**How long are the phrases?**","metadata":{}},{"cell_type":"code","source":"df['phrase_len'] = df.phrase.str.len()\nmetadata['phrase_len'] = metadata.phrase.str.len()","metadata":{"execution":{"iopub.status.busy":"2023-05-18T20:45:26.40768Z","iopub.execute_input":"2023-05-18T20:45:26.408054Z","iopub.status.idle":"2023-05-18T20:45:26.504769Z","shell.execute_reply.started":"2023-05-18T20:45:26.408026Z","shell.execute_reply":"2023-05-18T20:45:26.503532Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"\nplt.figure(figsize=(15,8))\nplt.title('Character occurences in each phrase in training set')\nplt.hist(df.phrase_len)\nplt.xlabel('Phrase length')\nplt.ylabel('Sample Count')\nplt.grid(axis='y')\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2023-05-18T20:45:26.506961Z","iopub.execute_input":"2023-05-18T20:45:26.507309Z","iopub.status.idle":"2023-05-18T20:45:26.88726Z","shell.execute_reply.started":"2023-05-18T20:45:26.50728Z","shell.execute_reply":"2023-05-18T20:45:26.886166Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"plt.figure(figsize=(15,8))\nplt.title('Character occurences in each phrase in Supplementary metadata')\nplt.hist(metadata.phrase_len)\nplt.xlabel('Unique characters')\nplt.ylabel('Sample Count')\nplt.grid(axis='y')\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2023-05-18T20:45:26.888447Z","iopub.execute_input":"2023-05-18T20:45:26.888796Z","iopub.status.idle":"2023-05-18T20:45:27.249697Z","shell.execute_reply.started":"2023-05-18T20:45:26.888763Z","shell.execute_reply":"2023-05-18T20:45:27.248865Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Let's look at the **vocabulary**","metadata":{}},{"cell_type":"code","source":"#Load the character_to_prediction json file\nimport json\nchars = json.load(open(\"/kaggle/input/asl-fingerspelling/character_to_prediction_index.json\"))\nchars","metadata":{"execution":{"iopub.status.busy":"2023-05-18T20:45:27.251597Z","iopub.execute_input":"2023-05-18T20:45:27.252791Z","iopub.status.idle":"2023-05-18T20:45:27.26685Z","shell.execute_reply.started":"2023-05-18T20:45:27.252747Z","shell.execute_reply":"2023-05-18T20:45:27.265758Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"<span style=\"color:green;font-size:18px;\">So there are 59 characters in total in the vocabulary. :</span> \n* **alphabets a-z** : total 26 characters\n* **digits : 0-9** , 10 in total\n* The rest are **special characters**. Let's look at them","metadata":{}},{"cell_type":"code","source":"alphabets = \"abcdefghijklmnopqrstuvwxyz\"\ndigits = \"0123456789\"","metadata":{"execution":{"iopub.status.busy":"2023-05-18T20:45:27.268168Z","iopub.execute_input":"2023-05-18T20:45:27.268484Z","iopub.status.idle":"2023-05-18T20:45:27.273087Z","shell.execute_reply.started":"2023-05-18T20:45:27.268458Z","shell.execute_reply":"2023-05-18T20:45:27.27227Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"special = []\nfor char in chars.keys():\n    if char not in alphabets and char not in digits:\n        special.append(char)\nprint(\"Special Characters :\")\nspecial","metadata":{"execution":{"iopub.status.busy":"2023-05-18T20:45:27.274206Z","iopub.execute_input":"2023-05-18T20:45:27.275095Z","iopub.status.idle":"2023-05-18T20:45:27.294657Z","shell.execute_reply.started":"2023-05-18T20:45:27.275062Z","shell.execute_reply":"2023-05-18T20:45:27.293464Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Let's look if the train dataset contain any other characters and let's check the frequency of the character tokens","metadata":{}},{"cell_type":"code","source":"vocab = {}\nfor sentences in tqdm(df.phrase.tolist()):\n    for char in sentences:\n        try:\n            vocab[char]+=1\n        except:\n            vocab[char] = 1\nprint(f\"Total number of characters present in the training dataset: {len(vocab)}\")\nsorted(vocab.items(),key = lambda x:x[1])\n\n    ","metadata":{"execution":{"iopub.status.busy":"2023-05-18T20:45:27.296182Z","iopub.execute_input":"2023-05-18T20:45:27.296579Z","iopub.status.idle":"2023-05-18T20:45:27.635979Z","shell.execute_reply.started":"2023-05-18T20:45:27.296548Z","shell.execute_reply":"2023-05-18T20:45:27.634855Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"<span style=\"color:purple;font-size:18px;\">okay this is interesting. Most of the special characters have low frequency as expected. Except these few : :</span>\n\n\n* **\" \" (blank token)**  :  Each address contains blanks in between words \n* **\"-\"  :**  This one is used mainly in phone numbers\n* **\".\" (full stop)** : In URLs and end of sentence\n* **\"/\"**  :  In URLs and addresses\n* **\"+\"**  :  At the start oh phone numbers\n* **\":\"**  :  in URLs\n* \"_\" :  in URLS/addresses ","metadata":{}},{"cell_type":"markdown","source":"Let's plot the frequencies to visualize the differences","metadata":{}},{"cell_type":"code","source":"sorted_vocab = sorted(vocab.items(),key = lambda x:x[1],reverse = True)\nsorted_vocab = dict(sorted_vocab)","metadata":{"execution":{"iopub.status.busy":"2023-05-18T20:45:27.638144Z","iopub.execute_input":"2023-05-18T20:45:27.638625Z","iopub.status.idle":"2023-05-18T20:45:27.644937Z","shell.execute_reply.started":"2023-05-18T20:45:27.638564Z","shell.execute_reply":"2023-05-18T20:45:27.643642Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"plt.figure(figsize=(15,8))\nplt.title('Character Occurance')\nplt.bar(sorted_vocab.keys(),sorted_vocab.values())\nplt.xlabel('Unique characters')\nplt.ylabel('Sample Count')\nplt.grid(axis='y')\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2023-05-18T20:45:27.647647Z","iopub.execute_input":"2023-05-18T20:45:27.648014Z","iopub.status.idle":"2023-05-18T20:45:28.441663Z","shell.execute_reply.started":"2023-05-18T20:45:27.64798Z","shell.execute_reply":"2023-05-18T20:45:28.440655Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"<a id=\"3\"></a>\n<div class=\"alert alert-block alert-info\" style=\"padding:25px; line-height: 1.7em; font-family: Verdana;\">\n    <center style=\"font-family: consolas; font-size: 32px; font-weight: bold; color:purple\"> 🤟Sequence Landmarks🤟</center>\n","metadata":{}},{"cell_type":"markdown","source":"The sequence data for each phrase is load into the parquet files. Let's load a file and try to visualize the landmarks","metadata":{}},{"cell_type":"code","source":"df.head()","metadata":{"execution":{"iopub.status.busy":"2023-05-18T20:45:50.033846Z","iopub.execute_input":"2023-05-18T20:45:50.034246Z","iopub.status.idle":"2023-05-18T20:45:50.047759Z","shell.execute_reply.started":"2023-05-18T20:45:50.034216Z","shell.execute_reply":"2023-05-18T20:45:50.046524Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"base_path = \"/kaggle/input/asl-fingerspelling\"\nsequence_no = 0\nfile_name = df.path.iloc[sequence_no]\nsequence_id = df.sequence_id.iloc[sequence_no]\nrandom_df = pd.read_parquet(f\"{base_path}/{file_name}\")","metadata":{"execution":{"iopub.status.busy":"2023-05-18T20:45:50.123366Z","iopub.execute_input":"2023-05-18T20:45:50.12398Z","iopub.status.idle":"2023-05-18T20:46:05.769356Z","shell.execute_reply.started":"2023-05-18T20:45:50.123944Z","shell.execute_reply":"2023-05-18T20:46:05.768014Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Let's see what's in this parquet file","metadata":{}},{"cell_type":"code","source":"random_df","metadata":{"execution":{"iopub.status.busy":"2023-05-18T20:46:05.773813Z","iopub.execute_input":"2023-05-18T20:46:05.774176Z","iopub.status.idle":"2023-05-18T20:46:05.838204Z","shell.execute_reply.started":"2023-05-18T20:46:05.774145Z","shell.execute_reply":"2023-05-18T20:46:05.836569Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"* **sequence_id** - A unique identifier for the landmark sequence. Most landmark files contain 1,000 sequences. The sequence ID is used as the dataframe index.\n* **frame** - The frame number within a landmark sequence.\n* **[x/y/z]_[type]_[landmark_index]** - There are now 1,629 spatial coordinate columns for the x, y and z coordinates for each of the 543 landmarks. The type of landmark is one of ['face', 'left_hand', 'pose', 'right_hand']. Details of the hand landmark locations can be found here. The spatial coordinates have already been normalized by MediaPipe. Note that the MediaPipe model is not fully trained to predict depth so you may wish to ignore the z values. The landmarks have been converted to float32.","metadata":{}},{"cell_type":"markdown","source":"So we have a frame column here. \n\nThere are **543** landmarks in total for each frame, each having 3 co-ordinates. So **543*3 = 1629** columns for co-ordinates","metadata":{}},{"cell_type":"markdown","source":"**But wait! How much of them contains NaN values?**","metadata":{}},{"cell_type":"code","source":"random_df[~random_df.isnull().any(axis=1)]","metadata":{"execution":{"iopub.status.busy":"2023-05-18T20:46:05.839826Z","iopub.execute_input":"2023-05-18T20:46:05.840467Z","iopub.status.idle":"2023-05-18T20:46:06.369574Z","shell.execute_reply.started":"2023-05-18T20:46:05.840428Z","shell.execute_reply":"2023-05-18T20:46:06.368303Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"<span style=\"color:red;font-size:18px;\">Only 459 rows among 162699 rows! That means a lot of frame contain NaN values. This will be interesting to handle.</span>\n\n","metadata":{}},{"cell_type":"markdown","source":"Let's see how much frame for a random phrase.","metadata":{}},{"cell_type":"code","source":"for sequence,sequence_data in random_df.groupby(\"sequence_id\"):\n    if sequence==sequence_id:\n        print(f\"Total Frames for this file {sequence_data.frame.nunique()}\")\n        print(f\"Total Frames without any NaN values for this file {sequence_data[~sequence_data.isnull().any(axis=1)].frame.nunique()}\")\n        ","metadata":{"execution":{"iopub.status.busy":"2023-05-18T20:46:06.371906Z","iopub.execute_input":"2023-05-18T20:46:06.372244Z","iopub.status.idle":"2023-05-18T20:46:06.976237Z","shell.execute_reply.started":"2023-05-18T20:46:06.372215Z","shell.execute_reply":"2023-05-18T20:46:06.975035Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Okay !","metadata":{}},{"cell_type":"markdown","source":"**Now let's have a look at the frame distribuition for the training files.**","metadata":{}},{"cell_type":"code","source":"import os\nnumber_of_frames = {}\n\nfiles_path = \"/kaggle/input/asl-fingerspelling/train_landmarks\"\nall_files = os.listdir(\"/kaggle/input/asl-fingerspelling/train_landmarks\")\nfor file in tqdm(all_files):\n    parq = pd.read_parquet(f\"{files_path}/{file}\")\n    for seq,seq_data in parq.groupby(\"sequence_id\"):\n        total_frames = seq_data.frame.nunique()\n        \n        try:\n            number_of_frames[total_frames]+=1\n        except:\n            number_of_frames[total_frames]=1\n","metadata":{"execution":{"iopub.status.busy":"2023-05-18T20:46:06.978022Z","iopub.execute_input":"2023-05-18T20:46:06.978497Z","iopub.status.idle":"2023-05-18T21:05:29.803424Z","shell.execute_reply.started":"2023-05-18T20:46:06.978459Z","shell.execute_reply":"2023-05-18T21:05:29.801276Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"\n","metadata":{}},{"cell_type":"code","source":"plt.xlabel(\"Number of frames\")\nplt.ylabel(\"occurances\")\nplt.hist(number_of_frames,bins=[i for i in range(0,1000,100)])","metadata":{"execution":{"iopub.status.busy":"2023-05-18T21:09:06.136352Z","iopub.execute_input":"2023-05-18T21:09:06.136982Z","iopub.status.idle":"2023-05-18T21:09:06.400822Z","shell.execute_reply.started":"2023-05-18T21:09:06.136937Z","shell.execute_reply":"2023-05-18T21:09:06.399327Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"<a id=\"4\"></a>\n<div class=\"alert alert-block alert-info\" style=\"padding:25px; line-height: 1.7em; font-family: Verdana;\">\n    <center style=\"font-family: consolas; font-size: 32px; font-weight: bold; color:purple\"> 🤟Visualization🤟</center>","metadata":{}},{"cell_type":"markdown","source":"Now let's visualize a video file.\nThe functions for this visualization was modified from [this notebook](https://www.kaggle.com/code/tatamikenn/islr-eda-let-s-get-landmarks-animated?scriptVersionId=124884880)","metadata":{}},{"cell_type":"code","source":"base_path = \"/kaggle/input/asl-fingerspelling\"\n#If you want to visualize a different video changehere\n\nsequence_no = 1\nfile_name = df.path.iloc[sequence_no]\nsequence_id = df.sequence_id.iloc[sequence_no]\nrandom_df = pd.read_parquet(f\"{base_path}/{file_name}\")\nrandom_df.head()","metadata":{"execution":{"iopub.status.busy":"2023-05-18T21:05:30.122546Z","iopub.execute_input":"2023-05-18T21:05:30.123356Z","iopub.status.idle":"2023-05-18T21:05:45.58488Z","shell.execute_reply.started":"2023-05-18T21:05:30.123313Z","shell.execute_reply":"2023-05-18T21:05:45.583811Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"**The Video file for sequence Id**:","metadata":{}},{"cell_type":"code","source":"#Let's get the data for a single sequence\nsign = random_df.groupby(\"sequence_id\").get_group(sequence_id)\nsign","metadata":{"execution":{"iopub.status.busy":"2023-05-18T21:05:45.586354Z","iopub.execute_input":"2023-05-18T21:05:45.58703Z","iopub.status.idle":"2023-05-18T21:05:45.643096Z","shell.execute_reply.started":"2023-05-18T21:05:45.586999Z","shell.execute_reply":"2023-05-18T21:05:45.641825Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"**Let's find out the indexes for face,left_hand,right_hand and pose**","metadata":{}},{"cell_type":"markdown","source":"Here\n* **face - 468 keypoints**\n* **left hand - 21 keypoints**\n* **right hand - 21 keypoints**\n* **pose - 33 keypoints**","metadata":{}},{"cell_type":"code","source":"def get_hand_points(hand):\n    x = [[hand.iloc[0].x, hand.iloc[1].x, hand.iloc[2].x, hand.iloc[3].x, hand.iloc[4].x], # Thumb\n         [hand.iloc[5].x, hand.iloc[6].x, hand.iloc[7].x, hand.iloc[8].x], # Index\n         [hand.iloc[9].x, hand.iloc[10].x, hand.iloc[11].x, hand.iloc[12].x], \n         [hand.iloc[13].x, hand.iloc[14].x, hand.iloc[15].x, hand.iloc[16].x], \n         [hand.iloc[17].x, hand.iloc[18].x, hand.iloc[19].x, hand.iloc[20].x], \n         [hand.iloc[0].x, hand.iloc[5].x, hand.iloc[9].x, hand.iloc[13].x, hand.iloc[17].x, hand.iloc[0].x]]\n\n    y = [[hand.iloc[0].y, hand.iloc[1].y, hand.iloc[2].y, hand.iloc[3].y, hand.iloc[4].y],  #Thumb\n         [hand.iloc[5].y, hand.iloc[6].y, hand.iloc[7].y, hand.iloc[8].y], # Index\n         [hand.iloc[9].y, hand.iloc[10].y, hand.iloc[11].y, hand.iloc[12].y], \n         [hand.iloc[13].y, hand.iloc[14].y, hand.iloc[15].y, hand.iloc[16].y], \n         [hand.iloc[17].y, hand.iloc[18].y, hand.iloc[19].y, hand.iloc[20].y], \n         [hand.iloc[0].y, hand.iloc[5].y, hand.iloc[9].y, hand.iloc[13].y, hand.iloc[17].y, hand.iloc[0].y]] \n    return x, y\n\ndef get_pose_points(pose):\n    x = [[pose.iloc[8].x, pose.iloc[6].x, pose.iloc[5].x, pose.iloc[4].x, pose.iloc[0].x, pose.iloc[1].x, pose.iloc[2].x, pose.iloc[3].x, pose.iloc[7].x], \n         [pose.iloc[10].x, pose.iloc[9].x], \n         [pose.iloc[22].x, pose.iloc[16].x, pose.iloc[20].x, pose.iloc[18].x, pose.iloc[16].x, pose.iloc[14].x, pose.iloc[12].x, \n          pose.iloc[11].x, pose.iloc[13].x, pose.iloc[15].x, pose.iloc[17].x, pose.iloc[19].x, pose.iloc[15].x, pose.iloc[21].x], \n         [pose.iloc[12].x, pose.iloc[24].x, pose.iloc[26].x, pose.iloc[28].x, pose.iloc[30].x, pose.iloc[32].x, pose.iloc[28].x], \n         [pose.iloc[11].x, pose.iloc[23].x, pose.iloc[25].x, pose.iloc[27].x, pose.iloc[29].x, pose.iloc[31].x, pose.iloc[27].x], \n         [pose.iloc[24].x, pose.iloc[23].x]\n        ]\n\n    y = [[pose.iloc[8].y, pose.iloc[6].y, pose.iloc[5].y, pose.iloc[4].y, pose.iloc[0].y, pose.iloc[1].y, pose.iloc[2].y, pose.iloc[3].y, pose.iloc[7].y], \n         [pose.iloc[10].y, pose.iloc[9].y], \n         [pose.iloc[22].y, pose.iloc[16].y, pose.iloc[20].y, pose.iloc[18].y, pose.iloc[16].y, pose.iloc[14].y, pose.iloc[12].y, \n          pose.iloc[11].y, pose.iloc[13].y, pose.iloc[15].y, pose.iloc[17].y, pose.iloc[19].y, pose.iloc[15].y, pose.iloc[21].y], \n         [pose.iloc[12].y, pose.iloc[24].y, pose.iloc[26].y, pose.iloc[28].y, pose.iloc[30].y, pose.iloc[32].y, pose.iloc[28].y], \n         [pose.iloc[11].y, pose.iloc[23].y, pose.iloc[25].y, pose.iloc[27].y, pose.iloc[29].y, pose.iloc[31].y, pose.iloc[27].y], \n         [pose.iloc[24].y, pose.iloc[23].y]\n        ]\n    return x, y","metadata":{"execution":{"iopub.status.busy":"2023-05-18T21:05:45.644736Z","iopub.execute_input":"2023-05-18T21:05:45.645143Z","iopub.status.idle":"2023-05-18T21:05:45.674957Z","shell.execute_reply.started":"2023-05-18T21:05:45.645111Z","shell.execute_reply":"2023-05-18T21:05:45.673763Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"#helper function to get the landmark point from the datafra,e\ndef get_points(df,landmark_name):\n    X = []\n    Y = []\n    for index,name in enumerate(df.columns.tolist()):\n        if \"x_\"+landmark_name in name:\n            X.append(df[f\"{name}\"].iloc[0])\n        if \"y_\"+landmark_name in name:\n            Y.append(df[f\"{name}\"].iloc[0])\n    return X,Y\n    \n            ","metadata":{"execution":{"iopub.status.busy":"2023-05-18T21:05:45.678961Z","iopub.execute_input":"2023-05-18T21:05:45.679528Z","iopub.status.idle":"2023-05-18T21:05:45.696287Z","shell.execute_reply.started":"2023-05-18T21:05:45.679489Z","shell.execute_reply":"2023-05-18T21:05:45.69507Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"#function to generate animation\n\ndef animation_frame(f):\n    frame = sign[sign.frame==f]\n    lhx, lhy = get_points(frame,'left_hand')\n    rhx, rhy = get_points(frame,'right_hand')\n    phx, phy = get_points(frame,'pose')\n    fx,fy = get_points(frame,'face')\n    \n    lh = pd.DataFrame({'x':lhx,'y':lhy})\n    rh = pd.DataFrame({'x':rhx,'y':rhy})\n    \n    lx,ly = get_hand_points(lh)\n    rx,ry = get_hand_points(rh)\n    px,py = get_pose_points(pd.DataFrame({'x':phx,'y':phy}))\n\n    \n    \n    ax.clear()\n    ax.plot(fx, fy, '.')\n    for i in range(len(lx)):\n        ax.plot(lx[i], ly[i])\n    for i in range(len(rx)):\n        ax.plot(rx[i], ry[i])\n    for i in range(len(px)):\n        ax.plot(px[i], py[i])\n        \n    plt.xlim(xmin, xmax)\n    #Inverting the video since the original video seem to be vertically flipped\n    \n    plt.ylim(ymax, ymin)","metadata":{"execution":{"iopub.status.busy":"2023-05-18T21:05:45.698243Z","iopub.execute_input":"2023-05-18T21:05:45.699414Z","iopub.status.idle":"2023-05-18T21:05:45.712257Z","shell.execute_reply.started":"2023-05-18T21:05:45.699375Z","shell.execute_reply":"2023-05-18T21:05:45.71141Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"print(f\"The sign being shown here is: {df[df.sequence_id==sequence_id].phrase.values}\")\n\n## These values set the limits on the graph to stabilize the video\nxmin = sign[sign.columns.tolist()[1:543]].min().min() - 0.2\nxmax = sign[sign.columns.tolist()[1:543]].max().max() + 0.2\nymin = sign[sign.columns.tolist()[543:543*2]].min().min() - 0.2\nymax = sign[sign.columns.tolist()[543:543*2]].max().max() + 0.2\n\nfig, ax = plt.subplots()\nl, = ax.plot([], [])\nanimation = FuncAnimation(fig, func=animation_frame, frames=sign.frame.unique())\n\nHTML(animation.to_html5_video())","metadata":{"execution":{"iopub.status.busy":"2023-05-18T21:05:45.71357Z","iopub.execute_input":"2023-05-18T21:05:45.714692Z","iopub.status.idle":"2023-05-18T21:06:15.977028Z","shell.execute_reply.started":"2023-05-18T21:05:45.714651Z","shell.execute_reply":"2023-05-18T21:06:15.975684Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"<a id=\"5\"></a>\n<div class=\"alert alert-block alert-info\" style=\"padding:25px; line-height: 1.7em; font-family: Verdana;\">\n    <center style=\"font-family: consolas; font-size: 32px; font-weight: bold; color:purple\"> 🤟Complete function for Visualization🤟</center>\n    ","metadata":{}},{"cell_type":"code","source":"df.head()","metadata":{"execution":{"iopub.status.busy":"2023-05-18T21:06:15.978316Z","iopub.execute_input":"2023-05-18T21:06:15.978654Z","iopub.status.idle":"2023-05-18T21:06:15.991825Z","shell.execute_reply.started":"2023-05-18T21:06:15.978625Z","shell.execute_reply":"2023-05-18T21:06:15.990891Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Suppose you want to visualize the video for the phrase \"988 franklin lane\". Then first find out the sequence_id.\nSo we'll visualize by defining the sequence_id","metadata":{}},{"cell_type":"code","source":"sequence_id = 1816796431\n\ndef visualize(seq_id):\n    \n    base_path = \"/kaggle/input/asl-fingerspelling\"\n    train_df = pd.read_csv(\"/kaggle/input/asl-fingerspelling/train.csv\")\n    \n    path = df[df.sequence_id==seq_id].path\n    phrase = df[df.sequence_id==seq_id].phrase.values\n    \n    print(f\"Showing sign for the phrase :{phrase}\")\n    \n    \n    def animation_framee(f):\n        frame = sign[sign.frame==f]\n        lhx, lhy = get_points(frame,'left_hand')\n        rhx, rhy = get_points(frame,'right_hand')\n        phx, phy = get_points(frame,'pose')\n        fx,fy = get_points(frame,'face')\n\n        lh = pd.DataFrame({'x':lhx,'y':lhy})\n        rh = pd.DataFrame({'x':rhx,'y':rhy})\n\n        lx,ly = get_hand_points(lh)\n        rx,ry = get_hand_points(rh)\n        px,py = get_pose_points(pd.DataFrame({'x':phx,'y':phy}))\n\n\n\n        ax.clear()\n        ax.plot(fx, fy, '.')\n        for i in range(len(lx)):\n            ax.plot(lx[i], ly[i])\n        for i in range(len(rx)):\n            ax.plot(rx[i], ry[i])\n        for i in range(len(px)):\n            ax.plot(px[i], py[i])\n\n        plt.xlim(xmin, xmax)\n        #Inverting the video since the original video seem to be vertically flipped\n\n        plt.ylim(ymax, ymin)\n    \n    landmark_df = pd.read_parquet(f\"{base_path}/{file_name}\")\n    sign = landmark_df.groupby(\"sequence_id\").get_group(seq_id)\n    \n    xmin = sign[sign.columns.tolist()[1:543]].min().min() - 0.2\n    xmax = sign[sign.columns.tolist()[1:543]].max().max() + 0.2\n    ymin = sign[sign.columns.tolist()[543:543*2]].min().min() - 0.2\n    ymax = sign[sign.columns.tolist()[543:543*2]].max().max() + 0.2\n\n    fig, ax = plt.subplots()\n    l, = ax.plot([], [])\n    animation = FuncAnimation(fig, func=animation_framee, frames=sign.frame.unique())\n\n    return HTML(animation.to_html5_video())\n    ","metadata":{"execution":{"iopub.status.busy":"2023-05-18T21:06:15.992944Z","iopub.execute_input":"2023-05-18T21:06:15.993277Z","iopub.status.idle":"2023-05-18T21:06:16.014667Z","shell.execute_reply.started":"2023-05-18T21:06:15.993242Z","shell.execute_reply":"2023-05-18T21:06:16.013257Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"visualize(1816909464)","metadata":{"execution":{"iopub.status.busy":"2023-05-18T21:06:16.01624Z","iopub.execute_input":"2023-05-18T21:06:16.016595Z","iopub.status.idle":"2023-05-18T21:07:14.36115Z","shell.execute_reply.started":"2023-05-18T21:06:16.016567Z","shell.execute_reply":"2023-05-18T21:07:14.359837Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"<a id=\"6\"></a>\n<div class=\"alert alert-block alert-info\" style=\"padding:25px; line-height: 1.7em; font-family: Verdana;\">\n    <center style=\"font-family: consolas; font-size: 32px; font-weight: bold; color:purple\"> 🤟Evaluation Metric🤟</center>\n    \n\n    ","metadata":{}},{"cell_type":"markdown","source":"* The evaluation metric for this contest is the normalized total [levenshtein distance](http://https://en.wikipedia.org/wiki/Levenshtein_distance). Let's first see what's Lev distance.","metadata":{}},{"cell_type":"markdown","source":"<center><div class=\"alert alert-block alert-info\" style=\"margin: 0.2em; line-height: 1.7em; font-family: Verdana;color:rgb(30, 30, 30)\">\n    <b style=\"font-size: 15px;\"> Levenshtein distance, also known as edit distance, is a metric used to quantify the difference between two strings. It measures the minimum number of single-character edits (insertions, deletions, or substitutions) required to transform one string into another.</b>\n</div></center>","metadata":{}},{"cell_type":"markdown","source":"**Impelementation**","metadata":{}},{"cell_type":"code","source":"from nltk.metrics.distance import edit_distance\n\nstring1 = \"scales/kuhaylah\"\nstring2 = \"scales+kuhaylah\"\n  \n# the Levenshtein distance between string1 and string2\nprint(f\"Levenshtein Distance btween the strings: {edit_distance(string1, string2)}\")","metadata":{"execution":{"iopub.status.busy":"2023-05-18T21:07:14.36256Z","iopub.execute_input":"2023-05-18T21:07:14.362961Z","iopub.status.idle":"2023-05-18T21:07:16.304031Z","shell.execute_reply.started":"2023-05-18T21:07:14.362907Z","shell.execute_reply":"2023-05-18T21:07:16.302669Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"However, in this competetion, Normalized lev distance will be used.\n\n```\nThe evaluation metric for this contest is the normalized total levenshtein distance. Let the total number of characters in all of the labels be N and the total levenshtein distance be D. The metric equals (N - D) / N.\n```","metadata":{}},{"cell_type":"markdown","source":"So for the previous case, \n> N = 15 ( length of strings 1)\n\n> D = 1 (Lev distance between string1 and string2 )\n \n > So normalized lev distance will be ``` (15-1)/15 = 0.933 ```","metadata":{}},{"cell_type":"code","source":"N = len(string1)\nD = edit_distance(string1, string2)\n\nprint(\"Normalized Lev distance : \",(N-D)/N)","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"<a id=\"7\"></a>\n<div class=\"alert alert-block alert-info\" style=\"padding:25px; line-height: 1.7em; font-family: Verdana;\">\n    <center style=\"font-family: consolas; font-size: 32px; font-weight: bold; color:purple\"> 🤟Resources🤟</center>","metadata":{}},{"cell_type":"markdown","source":"**Some awesome Resources for ASL**\n1. [Goldmine](https://research.sign.mt/)\n2. [Generate your own signs from text](https://www.signlanguageforum.com/asl/fingerspelling/message/)(The one I used at the beginning)","metadata":{}},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]}]}