🔮 feat: predict signal 🔮 (#124)

* almost whole model plan * save data * basically done * actually done w model creation * ready to predict * more done * new file * modify model * prediction trial * again * use metadata * make sure data is non-null * trade * typo * fix df * don't trigger every push * done * fix fb
suchak1 · Nov 8, 2021 · 762e8d8 · 762e8d8
1 parent 138a0eb
commit 762e8d8
Show file tree

Hide file tree

Showing 11 changed files with 111 additions and 30 deletions.
diff --git a/.github/workflows/model.yml b/.github/workflows/model.yml
@@ -12,14 +12,6 @@ on:
 jobs:
   build:
     runs-on: ubuntu-latest
-    strategy:
-      matrix:
-        include:
-          - pca: 2
-          - pca: 3
-          - pca: 5
-          - pca: 7
-      fail-fast: false
     steps:
       - name: Checkout repo
         uses: actions/checkout@v2
@@ -52,7 +44,6 @@ jobs:
 
       - name: Create model
         env:
-          PCA: ${{ matrix.pca }}
           AWS_ACCESS_KEY_ID: ${{ secrets.AWS_ACCESS_KEY_ID }}
           AWS_SECRET_ACCESS_KEY: ${{ secrets.AWS_SECRET_ACCESS_KEY }}
           AWS_DEFAULT_REGION: ${{ secrets.AWS_DEFAULT_REGION }}

diff --git a/.github/workflows/predict.yml b/.github/workflows/predict.yml
@@ -0,0 +1,59 @@
+# This workflow will install Python dependencies, run tests and lint with a single version of Python
+# For more information see: https://help.github.com/actions/language-and-framework-guides/using-python-with-github-actions
+
+name: Predict Signal
+
+on:
+  # push:
+  #   branches:
+  #     - feature/predict
+  schedule:
+    - cron: "0 2 * * *"
+    # 9pm EST every day
+  workflow_dispatch:
+
+jobs:
+  build:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout repo
+        uses: actions/checkout@v2
+
+      - name: Set up Python 3.8
+        uses: actions/setup-python@v2
+        with:
+          python-version: 3.8
+
+      - name: Cache pip dependencies
+        uses: actions/cache@v2
+        with:
+          path: ~/.cache/pip
+          key: ${{ runner.os }}-pip-${{ hashFiles('**/requirements.txt') }}
+          restore-keys: |
+            ${{ runner.os }}-pip-
+
+      - name: Install dependencies
+        run: |
+          python -m pip install --upgrade pip
+          pip install flake8 pytest coverage
+          if [ -f requirements.txt ]; then pip install -r requirements.txt; fi
+
+      - name: Decrypt script
+        env:
+          RH_PASSWORD: ${{ secrets.RH_PASSWORD }}
+          SALT: ${{ secrets.SALT }}
+        run: |
+          python util/decrypt.py encrypted/predict_signal.py
+
+      - name: Predict signal
+        env:
+          AWS_ACCESS_KEY_ID: ${{ secrets.AWS_ACCESS_KEY_ID }}
+          AWS_SECRET_ACCESS_KEY: ${{ secrets.AWS_SECRET_ACCESS_KEY }}
+          AWS_DEFAULT_REGION: ${{ secrets.AWS_DEFAULT_REGION }}
+          S3_BUCKET: ${{ secrets.S3_BUCKET }}
+          POLYGON: ${{ secrets.POLYGON }}
+          GLASSNODE: ${{ secrets.GLASSNODE }}
+          BINANCE_KEY: ${{ secrets.BINANCE_KEY }}
+          BINANCE_SECRET: ${{ secrets.BINANCE_SECRET }}
+        run: |
+          python encrypted/predict_signal.py
diff --git a/README.md b/README.md
@@ -111,4 +111,12 @@ Using Robinhood 2FA, we can simply provide our MFA one-time password in the `.en
 <!-- auto sklearn -->
 <!-- get data, make decision/trade, update model before 11pm est each night -->
 <!-- 8:30pm est polygon, 9pm est glassnode -->
-<!-- auto-encrypt and run on save extensions for encrypting model -->
+
+<!-- 2. connect model output to buy/sell ordering - update csvs with model result and order summary (signals.csv, orders.csv) -->
+<!-- 3. get new data and model input every day at 9pm? and order -->
+
+<!-- extra -->
+<!-- abstract away undersample fx from preprocess fx, and buy and sell from order fx, make oracle class -->
+<!-- 4. automate saving model and preprocessors (every 2 weeks ) -->
+<!-- 5. add live results on website / model vs buying and holding like alphahub - use dash or plotly? use pca visualization, tsne for higher dimensions, roc curve, etc-->
+<!-- 6. add authentication and business like report style like in dash example -->
diff --git a/encrypted/create_model.py.encrypted b/encrypted/create_model.py.encrypted
@@ -1 +1 @@
-gAAAAABhbkFAqCXq7xPvaZfuBxi8_2MO0VgPhAgZtDpPoz980prWyJ8QYgSpuItCQ5KVcJRghvJHIYmdQeBwQy_yJHpdzBGBW91GLo0XX3KRtqLUvi8Vw7DgX3goCoSeyZLRP0G3hp5vvYNuNyTJ63SOGiO-TmvVwvEYJFe15PfDft5vyZp-DL6Q4is289-IvpXEF7ria2P5c3d0bpWsrvx3QdXNzoCcttXg8U-ntxibFQCjycfO-M4-CB0jYzDcIx1266Ebp1IDxoyt3wDSriUeZaqwnXLrGrP_SESkRRSBBWzXm2SlRP4Ph8tuG6Q56vv8yJHr3QfIVc4G_EnJvMa1RN7BsirKEe_y0DRNPi_1E8Yok32lSp4SD733pM25zetWEHDGbBFzEeRUBo9oGLm3O0L2s4OjkdaXMYbDev_WxtDMoNNNELMwCzsHVxCQ3TdvrUpI7YZGOIFRt_GitxmP5FIPxSShWqJHLl-1Y7brk7bOOdpkioET12Dyty5I-7l0N5pCaJG3ki6VwgXhE7SevkQHyjOvFhHPtbnF1kKHzJVm6PT--18S1jB-ygo3P10wRj5RU3mLex8RqhGLiHCPL7dSP1_ZpaTWsyzvBBkkjSvrHK-pFrZERDPNze8iC9VSz2pszxyyKFemwexGNMGqh3MYXC3R5V_U0jl7SF_J5iMPLWTrey7Kx9_i-qov8585xDbyTYNryM4ZrlWRjoX9j25kQVAD31hCsWtV0FaqhIzLhGHAFNoiY2rAkeRIf3wRmd4ITTv3vp-UzV3JS0-3pJFxLM2UZOd5GAcRDoV-Xbe0LU5QK3rQ0MbUTCCX6G4o6YMhBXxh7V3quXbO-McxuDG0qOfgvpS4EnOeuSfV031OAyrSaJo34aKC6lOFZ-yhfTgGejHVtmhZn07QWbvrUYOslLORfyZLxiOKpIAJ_TN3cgDg882JpuwfYRViMcH0QUgsH0jwQXsBDkZfB7NGb0XevHY19MEXt21ePLCoDf8_kqTV2BgI1VlfvZfMjBTO16WQxUoF0m8CEeEgHb960IGD8EFWL-Qc-T8kcO1buUsmiB9zNxW6w9BNEJAZlr5rumRddZ0TIk0maK7_DtggUxFkhnK1SJ_xJS9Hc6OMDKsJppjzQiNIszpb4CowXzBp2-9isdcgG7B2Z1t3YJlIu3C7FVwlBOQ3yNAWC923ImAaH9Rsbgl3G4A61KWzgLUNSKxQ17fhyvnrXgXgteAzcxLrheuB5KsIfD2Can1YJOxjlBlV1ng3xk0SXIGeHc9GznN8IMgHEy3_kH46A35NypmD1OUBvKkxNp2g6bbog7ACt6hqEqdUgEVa1IRJM623IeAa4EoS24W3_VFUrwTGbheR_E4PsykVI4GyBiZZWFANYJePlr-DCM8bdD2eNG_3tfl_oQ5NajWG0LACCHdyAmyLl53kPY1lb925n3_9EFX3QEnUAR85-I75O_6175sv5xhmpVLoXlVMze87mTBvL-aFeCFn_yn5E5KbsdoGHMZ_1hQaTFr35JEhCSFP0HZBgcnOwd3dgYmPdtmAFCZzJvXB1-BtXibOLt4gckW4Zag5DUWHkjxMv6JHA06Lq2mYIlN_8ZE2D3_jdGZ3P3KANaEOzxPZ_0SSU34pjps7Z6CcMFSVTUBCZfo5spaRNLtEmg3fyKuphfJoXVMtnMwd-HGS7LBKiXD5E2Sw1w59BP67CndEAF5HoLHv2xq1rg_QpfExuJilSJO2NIORoVmsCI4SUz50DyHlvkHrGwYzcz0wdfwHvefPGvKTJZMHBtV1L1gwQgBDepZ5vVM2dzUlnkgfFsqXjjEGJ5suwTh3sEy16l30Y1LjiStkogdsmFi7bEcxrdHs3I4TUwqzmjEFmPjuoDuRAL6UYgq5f_HItt8Vn89q-FlIpweLEJORsA6iEFwGayxabmWFP_H7Dr-S79SeGxvcyZDajtfRJpHnToM1q2tM0z8PLUv9R4jT_7UtlzwdzEp_JqKI0Hc-4OKEAKdzPJenrp0QTcBl868yAOMbEh7YOQXtMUBLTctrwHwgiJ4xtwl5GM4qwrZmP0VWcR-_AAAdOSZFVop2q_pYUxEE93OF-BlnjGMEbK_qoEKI159OPO7vXjBxu0acQzTIV2VAeeNSU6y7Y7QeetJz2Wb7cy_sJetA6d-XVkQh2_LVuUn3r9wlQ6_4vdpTZ9rLjCds3H0UBK8uwPgBn5ktKig-BK19fDCwgbhVtnsKNK2Z7eDNMS8YttG1AewrmFi3AcRBG3OWYgUZ1IrtxyRqqFNY794v0sLYmai9gtJBEdJkTSo7mwR_VfAba-d9gDHt9YRKYJr-hBhhbVCY--SxzTF_aWlXMkD_bgd5PkDMx3MkVP3rIlUcwQpQzcXJ0JnIaxCC1XRald0gt0NA193EY2NCYW50yjS_vvVjW7_Ar6P2tlk3kiPNfUUyYqh7aCw4zrGwJFuKFE22V0xYSaNIY7k1rFXVZ6eZmedHZN9mOdPG05Wcb-1I9dtYaOEJB8axKpzg6xlgsjfvu63ps_poiLgk8hryIy6Nd03GmTSsdv42gQbtGKOkJMS7MaEtS6wpH4jDkvCcL8eBtXAnlcaThg8d6yJDMYqqXLUrFsJuC_vKUIKVfO8MFFJ725qXamWuHP66_rcgpwn1npoSI2JSKghOyj7UfqcgafS4ntn94s85s8HUwpAgYExF5F9UUKZBPu93uzVrELqxgVCPc6vAFQ4g_4oaEZnJAI6eYm6_7QQTmX15Glf5dAHMKLE8sjl8pmD0R1JI1QsBj5_ZkARr8Yg83n5VoLWI346D26bbcRjnB6m-yMwWvotszKoeikEDDsMc3Q7ijcV8q0yNrQs1LuEZUrytXebuNiFDCe4omAKa71lm2EAMKO14cy5FiEaLOUJbQC_jKB2zyg63-GDi6UiDmswZCeOFIq3svpXyK9CkuFbi4rldrHW283JVh4k2DoHKe6qIWEJSIms8q5tgYSFCBfJ_vmksex98IIPFsEeUzApjbmrrCAnnLBoxIUt4hA9E4Xhkc44p_gnCnI8L47GvRIEfZG--TLF0d2Kxc7TlvQZ398xSFr153ElJMXkH7vhsW2VKdL2czk6Rng6r1fKRsdmYppGfAw3AiPCISJqEUDUlACAFf9gjfYUW5kxZO4EV9bGo6j5UgcNzrVKTcAK5rKP79myBjKwB-54MS1QTh96Sg818Zs9wjd6lHgwZs31snGN1qTPUIlqM80WjXXwNCNdj1C8P7EBRIV_pwVzqYdpl86GcQs0cVcUz2kOwZCaByJ5LmsnIG0t8hS1GxbGC5IiHc6Ef6R3BaYGkuLQzUIMqADgzEgd43qQ057jaO5O8JIO3h1ByQ2eWBm9gkIfD9RRAE4IwYyLNPK6qOr_J8RnAlM5Wm--h-xS-4WgvcJDmLdc2y1am4RWNH9vhPlEwV1qUcDShqoX8G68alaASyotNvOUx568w4o1vDA7VPMo0OQT_K6DOtbwAzjpLnZQgAZpaN-FSK93YJrGhuYZp_mmf5vpa9f1j76F08iEPahF8ztBAIezUkrnxPiahgwAHPjuNllkSoJD08taPEKswKwjzeWnSFjnAjcpCok-g0E0iLNxKXaEcz1tEVd4_yN7jFijIFJLZJrRS88TGsjs6qn0QgtDfi2keaZ6l7xc-bP3znd1P8YX1M_n4di4uHHyBfB3HcMl98jLIQHf4H9MR8iIrwV3A4UhVnRiZ2YOMSuW9C8h7eyFPgrd_7I2NfiH8vi4j9CPJJpJ2K2UpfU7_OaslKy2CB55XOmdEauxUoSEF3_ZiIhYCD_iw0cZ0B2wqxK1qwR87qnAhA3S5prgCHJzvFlW6KrpaBz7ahEjddeyxIzmZoeZ7rP7CF0C1I4UIVbfuzSTz5xDGC9uto1rTETnNFtRj4srBQE1qLsvkp0VVHjEJvd2Y09VHsD8OtlJ7zg6U3Lc6gcbnQb9sQuOiU0kZgb4bz5TwB48ORyAfpgUzPZ_t-Y4KAtyyWXFIMc9Tq90n7nIeHhG7XF8O4hfPhr0s5bLfTdaptK9bEvtFtoxasVgwOUuQKwUwUbCdoR4kP47zMIeEPv2sl5DniTMTBcPA1uQjnec6z7pEzFnZA_XUAQVBhziOqLvSKxXC4wNgPJoYUY-MU2Us9rVSVwfK3gYcgaUq1T9kIUyrpRfQuHv3RbAPCgeLwHqxRq95iVLvj1FOfJyq-SFmqFC5VgbWgg7ARFpWf2bQsWG48igp0xyA5sqBeQFiuK_KaRCZLn7rcfJ1Bd8=
+gAAAAABhiM3JJhBPTP4WNzaNOZvuIG5Oi_74X8GHY1_KxedZF4D6iYvXRQZaC6QSQFE31ozxKKFOQ6gFrHrke7PuBgqOa7We0D3958qvcQiY0veuIR_VEzjOfVRTMWlngXDO2HxQqmwat3ITZgrnrAPs7eB3HXa3S7O6MMGXZfOQH8wgAu1oMKaUxodtJtKH1O1odiQ0uBx9mT7tP-dnynRE6Ct81UrxdCvUpOd5Cv5gG5rVW0v163huo3ZWqsysdhLPQa_OCegXDciZQeDsIi6TdzQjwBMr01oIYNHYZofJiehB3URzWeAZX32xustj7Hi8RW2_ZSVZkjNXfSwaEF72rFCBahsVAAzFjjgbbM6yfgJL8HYWep4vNrVYHqw92uZPIvGKN-dq6MkVlZLxiIOk96o24kJw8YmUuacmGNbtCO5jN0J5v7pS2G5tJVQqbcOwNJIqyxCG7b2VsE7esnglds8CLxMrKBsV_BW1d08OIyQzQGIAldjbvOcQBadj0DMkpZjBuig79QonooFEykxjnr1n1CF8khNY1-B1gWOWc6NGssD4UneTSSHQ98lKjKS5cFjmozO05RKeGKjGEhIjANee2EdD4QKKMAPbijNpRtnCKzewtWEafcIUOduOLoDRgnCkXkvIc-tGO_Nh1vacviznTqgK3WRAYHHG7m9Q4RjkbIY3YjN0SadZmt4aaBvz0SbIfrjVJMfA969ZDEVYQynNaYHIT6WIACUM3qw9vQTMTD-Wl-2uoTO4WTl-uEPQIKPOYybvKNgBMT63ffqQjiYyzSslFKXpR0T3iIOS0tuI4Y6VCzDs9PHDzPtE2W1K0xSGVr96AUlXRCehzxAPbsiV2otjM8kEFkHk-HKDVidJ1RieuCfC0iFLOqg5krU4rg7IZKJ_0crSKDn9d9cVmOX10ZypDe1TqwdhS4spWdedTryAyVdBpvqr8OyAcPCR-pGUYKXyKaHICJl2GIWWGlYM4TfsG2dEDII9npf8guLxJIevpCGnaTYlM5ybFBvdGv9mpIKRBRjVyNl4ceAOQEbUyMiD5XjsABFT536B-tYJOjTGOs1-ZsXq-vnTYte8FY59uSzWsdsp0ivKzyYJDnFIsf9nhyfcef4HgoP-1Aay4-L-UL0SAyCLRTFSRQGrRlhNGXCTVq57Zzqa7WwRqgjs3nvukBPeYkD-czPT5uDo5AZtGgFU7q_4JBfB9hY4kRyZomlHK5coDgG13z4eXo3wlT8SMu7NP7F5tpGbF9J9tnA-vMOyz9BDUTSChEyBYUhID5J8OjIVV8BMRQsotQsc5FNPeJ5g_fLLhVF8kk8yPm-FlmwPz-p50lJMO0NvOxw4Pfm64VGOD4P_tn3M_pwmuue5t6qYcy4ZOzcE_Ps6cHLU22rwMC9PQWUlvOUAcGUILBID3fZlM-IAhBpN4jmaS7nkGjVdBB3hnqOCq3gLOeQ7muqZYuoABtBOxzp6klYpaC3L2UcEAKWPY78khE3gxhG4uIfcmQu0ZdoZXd7nQGxzcQoiXFZJKEofRhJxUZkBE9nTZjSzflgVM8OJufvYmBe3PBTuX7wNEPOjW6X8wqqrhvnga5PtOth4Yk7e8Zz98tPn6mI7J1GrCuJEFczTbD1nXcnlCYJsW6nCvll7rti-HD3yumRl9TcBp-GhlwHn05qhVMRgsWJloz2Z3BX8TLfOdYmFuwniJKFDaBo2XLlSdUxczaQu9bupt0udyUPuX_hFFrLknDlErKxcyj32tRRIzpfcQjZOnPcCBNtmzcDCjdikCiez6WnQYmxauNuU2bJRMflllJqAqu3XPi5D_cKYEbEqKUOiQ6cdKbkcqK8HJx-6ovHqaApyf4SexkdLRLAkuDAohzNdM-BSejqxD9AWIPh2eQ2JFNRsC3VXaASOZLWYHGIQxnX67bWuc7ojDQccTkfFjiJlqZah1KjwB1FfU98ppZ61JFSf8CVWXSfabaNDT60HdqewMAMB6V1m1vBmkcT8arP3okeXfj6az_Sb0su7Kw72zhBzO_FB_LzsJmCe_RJrSaKfVC9B1TSFA2_W2LCv-0-ERMpkbELtIfKkPSfjXV-SfaUeC8Zsu3qLslYN4nZ0lr12ImeMDK7z192SA1ZjPbKm8J7LyStN3EsmZxYpdQuAofK7TOPL_2X16Wr9n9ScHmBuCYUlJQ026whekBenTrUfd7G3eKq0VVg5qTECR5wIdsO1_zNjRqjAsLb70QKHvShuYmMShWUhrhrIDUiPz5zHDdoDijoMi_PPdMAU2MQL3tvLROG4LjWZin0u0abTxClVwpCB4c6Cxp6AYIkymnNN0-VmruCGVA8Pr6BG5-x5v9CDCqSpzJc1vuk3dtMbJZy3FCKfhLEZaSjLFXSWn_-2iQQQ4Xzsi-RkkVmOZaaraxltif_nQru0gAxn9tN-Q4-hT3iC3iuxiaWUUyh3Ujd0t7RSb7BfIhXnG4ilzfbzipcbAALyXNMd0OJd_4_KW8L5SpGmKsduo08_U7pPqSlr1vv-yW76zpyRMVp2epHjdHMI9pC0C7h0_rYVf7TjsN43u2kfoiJuKXmf1XnYU4EL76VvdWnhoUsAqfhtG85fHnvmLmms4zP1snvZGZ_xg8XS82q3TQkY3SO2EiL_yPyue0d7pexv_MenNCeG-PPBdEJGCSxdwL9e6pXwEiG4lGiZkKXMgqzN8XHPjc5KqVhuSFgAkoFl2-FtUfv7U1bFtz_9rvvVpgTxQmmdzls7eKlTDlihDq8o53zrzvK_73G9SBZuoQf2nIRmIsmTX-9OhfS1qOwJsJXDcXS2EiM5ETq6NnZLkqXebEgQzNnF9othL5Tu_wZGYQU1PXvigjzCK5akQ2692X7a62MLmCLAFiDfvNSolvPc2fSZKYWzmjVAXCczI0E8-_oejv9ZLTAk7-sVd26B_jx3Qlxipwll3TLw7JFYuyxyige0ectUuxWM044c-vQnu9bq4kJCdNDdrYlc2who5B_x4Jct1NqDjMrvM4ojtMoIFBRfDtUIoHy9Q-_uoMZWLZv2tT5luSJpnc3vXzCk6TD0Fszy4RiHUbHy2YXU7lSo4suLZ5mrGB0JNCbHzKylaLTGsw4deOOAAtHbbQ8EwZJS0aJ4sReE3_4yNZPsEBHssfjPXDma-Fm264YfhioP-iPDqVv2X0bhSzu_kVbG_F971SEK5TTIp_Gsl7Gnl1DyTbNDyrnPyYJvmisCdywcgDKl0RpyjAuWVy_EeMUY-pA7mqLpAVil9r2HSXMSM516yRMotXhXAxXQc7hn0n95xKsVBFpOVKYm56PFbhkdctCKWZwrzl2BkHby67qYGZaoghBXVg0Hcqy7yFhvjeR5l2Tp6-LyXw9KSQhLHtPPWQszbqU7eI246Gr3g1CMY4Zqcuk4HXQ1XOoOn4DK6ANvXlMBBCY3OB7DH2UBQD7pwld-w1RotwdN_Tcbcg62IFUVrTBp0krcGaCX5DdpyxReguNaPDZZyE1k93U8vVSTuHw1XNlJb_xRtSDXJ9VLCp62rerVNIEs-wU3yYcOItuT2__oMeapVDp2qtST5F_4UPOeaEAjs_hn5efudy59H1O69M3ilysQR7UjL6PLvKHxmdXUOIX_wplE9TXkF9h0wM7rfa4V77wjgIhlv1l-KJOAIob_QOxCJMJnTIFmYy28E-BuDNJYGSFLaplwWIazf_8WviVfPh2RHxS6BclzJKoT7in9KHIlkGegAQkcQ3V46sPPI5duo2Z9EJq1dcez_aS9xYsF6VcYwwrRF2ZYP8YRjQNj8jENvcNvmy9_CbFUbqrS2U829DWb-6U7Ig8Y1X0RiGFCANHdTNM5thFsIDlh67e08FFMGxkkcPDXQG_H_aV4b71qFMIe3dm1uwQWnswdkk2c1HNvbi4GjXQIts3IAN_bPurJs4I_KgL-XlkG_S35KM4atPFS30pbyzQk-Ae8oJpdqpWd2ep7RR9YT6m1kSLhpNIFo1Qqwcqn1Z4THFj9EOvcLvymuEDAJuZC1IDLwAktM1Aw_bg2u8MPbi6SmgyQVG73RccMK5f0pqxaFrU5RlUOKFo5OyEH5GWxVTMmr7B4-5z3-y4uYmYiIT6COLoXMdEcwVRSDDV_fzmhKBAJqpDef5AAFffuc70AjjkTF10M5tXZn2KuzTEnxJIoDU2y74lUf0vG49Hhrii2_bcrBfbr-dnDAEH0gBBowodTZltv7h6dPhkQRhM2ZQHX17i_KhMghUNgKagtIxZMPyPli546MyuukqA926QV9tN9MGP49CELjm3Xzcq6yGR36rnWqFxj1xvY2qlTQ9i8jOgrb-gNoh4FNycUYW8EYJ7EZ5mRkMYWSaR0NlXOcgI4KS53wGrJZMQOVHPLxdxeIOuocchGVQl3TBtXeRqL08S9jfPYM99Vqsklv6Xha-tQGuUm-sDEwUzbHjQV7UxFpj-AS2_FaHnqxKhAOw2mnfWOkUcd2dlM6vNguwVXpIkFXn3QC-vCL3rK-EGzdlWr6UVlbSL_TYu_tlHfggqdGJ3lVojRSeQMxEyuj7JJBvNrSUcfC-GsI7gVWvhnedZgTivL4J8nNo39J3LrLRv21fcgp_Hie-haP-jIhHwtLb9pX-jgxb_vLFL-t8gq0d5lVyyGqvzLhn9vVgaJiiNkVdHxFsO3hddeH83lcNASGJ9XvQ1WtBmMiL6BCYBeDrC-kjtrb63oAM-qnE-0qgKvws1dyWpPCKBhZTU6M2khYmy7agLRhtqz7vBDtLGIpVeqsZ4cvBhYsI65aQZXDzCONVMc-VkV8FmW2IjsUXxvhuNLCRZDZPIlN6FxYW2L0M1JoKfWI05xAgq5DTCRKxRdUZNcD_OL7bAT_6KSu5pi-uueIvUQ5kKO6UGmwdxULp4t9CiX8AXalCY6tup4HP9cutPR1h1LZaXh99CImHWd4ZSd8IyEUriC3MHS5q5VZxfJTWt1b8Hoxa5qEUeP142X0leST1Ewlqxzht_5Ig8U4vnMJNngtAhZby8mA2svbWpQQKj-YTRMnBgqjzHp-OB-Et4ld7dpMaFYlbIoI6B3LwtlRAAWOHLzujeqO_E_MshTuqnfahTX9vbPnNsXWLUQ6VUhPAXY2PGiZ9lIgHO6KC9wbHthHHgtevXfP4OQAtcmtgOLl-eGwKq70dOwamwvhqfBzU39A4kYT8SiFjgecQxbSxExzfhrWaZmjdaPef08Ple7yTF2lgdrImqJ7hMEVv-n8gFtUYy-B7O7DxJBRy8TLR2VPsuOCQ3I40oz3PmnIVVOVsHTU1qHKk2KLiTlX7qvK-4z74yEr4p9fdT4d2cG7g3n-_TLCYdCnzRmI6XFavLyB5URF1NAGfiPliK5i-P5Bh1zEEpp9yJgLqjj-tPxBpTbRWwL9kAe9UrcyCtt0bAGAE0NNeIRLQl0iqtH-04Z4SQzEZjjY2SSnJu5N93iQ_rNlxDnI2g8i37lpOSgnzt4hY8MJW8ZX1LoBlEA_xW7x9-Cd_hEYuaDo3h5anaQNf_hicDvm77aoXetGJfOCPWZK-JvnjQL76bDPj8oaEEVzr-nrzp-_eSefDDnPDpmeIlFhcHnKlMmjDQTOe1GDk0AxCBALk247HS3FH1HEoklGQEoB2oecG6AbZq_A3pYLdP7slQC19Vl1XMiGrtl_6NAISMLTnwSvDu2n4lv17Fc4N733ScXdPkCAsA-jYsC-ws3u5KtVAj0rZhY7w3ivk7sup9CEPE6mFWlzr3fz8tI0rO_TvmaDOhRV5_dODuGuX8X39BUgMk_Ug-dbp1-iRkGrlE0skOPYxfA_kxYVQESZtoZqAmOXYr7lHe8dpSGYp3JNZ0hNLVItbzMW5bLK8RRJxKl8UlI7SpoTVxeAQcKslpedYvymmO7xe2RhBsG4NfZg4JLmRSyio90_EBem4LiGxWAdLpXiNHQkO8KQ5Jwomj7_JtEiiHTSvMoyLXWECfFJOeVe6htGBQpIMnKuC80C0HeXCKu13GpwGTI6j-o45bBZb1j4_9ZyWzpQ3EbO3u1PeoZkrdyMaQh_oyg-oln-n7b6MO7OcZp-yfBj_xHqhXkwhPyEZj9qJ7bqj3XFfqPjVibfcoIkiclqD4JJsMQzMRJvbNBAAmVfcgp_QYu_a-DFwBt8iHwTW5xpIwZpYhyTLQFfmOMWuwlIhEovq55_qOt_7WvVgqun3gOJQhl3zlTsKsRIth9SrvwwD4wsRXBOnZmBPGErZxVCY2ieiOirh2u0UBJytTK-DuTydwIDF1a-ZVYkbivyGuo1JtFePmlzWuvHlhV-iXFIsTbSeVJZWCBTNT9NU6ViDDLKbbsEiQ-R3zpvF76L4-KsIIL72JN2AZChHsuYFlQPOo88Vdrl5OtXq5efE0Yq0fOk_c1_J-qi_Z9lmos1L95TVa4fWRBq4m3VySXMK-btSNG3RMa9rN6JRj5WMFeYvc_zlUkxdz_oASTLa1kwOk1yhKNHIn0G2HZ7Dq5gcLLU6WTUdfNly5b8ZTW3QuXQaJ0-kf4hNOtVn-cfhCRhpQZE-UQmHTM_JtovIWAf_1ODrhwItalLA==
diff --git a/encrypted/predict_signal.py.encrypted b/encrypted/predict_signal.py.encrypted
@@ -0,0 +1 @@
+gAAAAABhiNfzR8mYiYlCVUVAOMROC2txWlDSWnqprOIpu0Wgqkjk1TqoqtN5O8j9Wh6W4alt2dS4Ma2ijSCAyWaGWDC-kOeOU6fLaLZqzhqjImTCfCXbtEeMrVVHkPrZSP8sOoTv4RZx48br2xbg7BCvG9SSYMLkRhNh_JdpgncPSY9TAREEkW91G2XZgtvW7plKmDLHCsJDR1ZOK4j37mzVPGFifBcFvXtOtOOb6MirX6qO_imgj46TvNqBlOyVp5oKA8NyJw7uuYCn13n7zHZgvETN5gryTER9LQdqzhz687f8oeGDk3lwZENWrzIBBZcJOeWKjgFH6XsJFx0mDleS2D0RBnPhAfhNUQfcKBTvDIb9CPAcvieZIvZCpfoX7uUisnBj2kCA5AhBOGwc1FZKDcPYkKYQDRle9L1L169XoNzt0yp9UmizJAQbWIi_vl2M0K9EPnjW_2xdoEW_T49Emc-qL1i-8UmSGqPY2olJndh_dW3YiiszPRCTIsHNrFoiK-skEtDk1xpeWXhezXvGO735lXt_oQWSGuuYQi2nxgxXuEvycN9JFwVbs4VSjSTWpR-bIzkJzMdwJLXDuwcq8GEsp3wKfeIIhrp-gVHZxmBJDTib57dkpR_Nn72pWak8oJJj-CWT8_fhYwP8eSB_eYg3PVyKu9K83jP-cU-XdWb80qZXzM6mATP94OzUvNvJNPEur--vICcyYvEo67vgwWdl8e3bAQqa6JRVeDTPAeoPdgL0cR2zig1iHNDUTuaa6NfEGIvCcDPxGfEz09eg5DtDrMTyK4dM4kQ6mOKiIf4B0fkuPFdVgKul7v3V44CyY4vkCR9cUMxqC04Q9FKr2YkUSuLWlEVpdMpWBjJrTMbc-1k1WXndH8g2gC_T2pBAqM87QLCUDgbWtwTa3lwJfvg5dUZpKDb4zYDN5U-VeYy30AM7xJwGPaam28yOuR1hHVnkF7hVdQ1LGh0W1-t5GviKrcnooO9dsfDChc_0DaAKopIgacDXZfb_iBAJAF4dqvVy755PjkkUWdjeUvZd7Xqs89WsHgtA7elvAgdqHddA936fjIsEHLLdqlv-HbVqXghjA8ejk5OLN5Z3607cmmsSGWfOSosazpHXPjEpfpkDvDfler9GJCBAWzDdtZG0k45NFc_0HpuoZFm2fb7dJGg4bjuPyqFn3PlBdHgFQJWTMkbuz3amMrQ4P43KbjCP2_OKw_ciRxHPJAk2mzkzZFD_GB3ASqvws-JbOubrF5VQGzuQD9Y6AsSAT4QysIKD0CQOqX7BW1WBCniBifaWm7h0NiZtnKTYy_aKLmnxZkC7b9qa0fIF0TMySrRxlBqek3reGGXbK_Z67wQvIQV3U8gFFMDIPQ5D8KYoafz43UbLkt4KHtZBsPgnTo1411mWJ2A6UXDZncRJyLDkx8g5g94xjiPEmqwFfm8QhYlXThFVy1cHqoe55Q2yhLwMtFsA0kwcKkIL-EPmcoIqFhvXhtcoMDlaPw8zH2kEWYcuGtZHDy8fcxHSfY_sPDCEvmiENDMiWHOCLW2K3SgTJBdxKZ84QJQgGJZZNRWg1xR27pSaDF0PRLn6AA_cR7WrSC5e62gPLvGd7CLQy-zvVmHZdXeaGZxbmEaoqQUFT6RjSz75WnWHzWQYcMsZu7bIwSjEdxareC10oEyv_nWjHRydOqQMWt_tF0DSzZuOVERPDCZVNVSp1CJdApMGbumqNrLLzr0XMaVo7etObHgmQ9hNwlX5THFIH6jsqrZAspsRyuJ61HVlTeRwe-fT-Ct8LdtXQPmUXSp5yDoFZvgJEAe3BIFgVAufR_QYYvzSVzlseYR0VWKAWwOqBR8DteTtG_8FnGgL3-u1yvSRM5jLIPjqmEe0Lo8s9P1TyYSZRPCTy21h56garQCM500gekU52lkFbulBu_wXw3B5SMpMrzF4S51LPudUbDSAoM8RxsKAhro-SL0_J5eZFQ6a4TFmfb5k3OKqnSpHtbUcmtr16jIXQ8DjPPLE64Met1ZGGVbJdA213wHn_k4sziUg97deoqlLAxvDKRYGJiHG8f_1wVtSsT2cmUnrwiuEAyBtVw7YqHKl_U4Vs8wlDil4n0pOG8L4pOF2xjTlXe1vHubJB5YMOeEU_ZZDGBoPz87kXRGLIN1wX68bBbwUPq4rXLkW1b_okhkLzw5VgKNxLCiJHuddKHPb4VRj3GlSN4a2M3Oc78EF_SQYg_AeVwW7loYGdk0qZ9EG-DiSy4iV3FZ7Vy1LlM3NjqFZxOAYvwPU8zmeXFA4tHVPjPkLzH2Bi_8FQeEwBgXl9s__yIz9uZiQ796RW_8649FcmUU6R5GMZXA6_yX88KNnNAMVrQj0XAzQq8HDIZzCn_XSb36JdAAXkbasZuhv8Eu5rqUgvjftK1elcJ8k2Ja6gJ0RSlujwDaVEWkiyRh5r2dc7QowN5cyT8BNkhxf3ACQ9cnAfn8OAgOLOwqzx_o8lQT2q1bjnHRE9EwjskCqLWlVqBD1neoegCN24GcwZI73A2cx1BT-TUrTyg2ejMh2fZ5e1zPXzBnCTjSGpPjKLpig9chCASWd1dhfSeazeKZpZSKaNbdDKY9cYMTpAnsdXNiS4DlG1pDKjetxrq5hcmCnALb6E9DExGiwQGi__6j9t4vWHuonrlw3akm8RG1blWr6wRQaP4aRauyCBrKml5LStAw-8sb8qYN_TR_MuN-s7IPvDmRZGPat3xs762zdjVB0hDStcpdvZCmAQdJTaW6APwgqsI_M5sO6YIjqVWIhu-CdXURJelEm7-acTNYQ0gCZ_ikGft9kFhCs9lLN7JSjABEm8UMxCUnwFQRu3rdkZHtjZu7TcszU-wcrc9H3HmuROrBtipq8zf26QfkY2RRqV_Y64y3qFqdUtZ4r4YOJBEYRiRKroEm28WKAHx0iblAjB1kS7Egy3gp7KdTa-YuneNlqBd4Y-srMni_oFCaLSg9__hflWjWhtILOIUltKNLuIaxl2GR-2wzng1DeTmp3_TU_9StYpnEwZuHH8rDrG0_PL0CSVhHZTioRXMDYROS_80wdx3AvB6mvKAANXlen04Lqj0hcI5pQXWnOFlQyqafKmq7Oz0q8eTA7Qj_FRzXywcZqg3md30GY_FwHWlF6k3NhIE7EbYGZDNE8cZRllQzYJWwemQeNAfhQX0iY0qR9IgGcV7wUnfTX7w3E0ZvjyQcTiDyBHZ0BdHo9LS9HOw4Y6HpHcVlVj_vtyBbhih8ao8N7yU_Ya5ydM28xpPqGHkShLq1dIf8FJoWSFnctHLJdsjfyKDa9u7ZCt2XzAzhnalHF27NPCs4yN5fLFr9iAGhCO1rdQ6WYKw5xq2AyAY6_8V5f-_3CeVXvv6xTax8laN7YftDBaQP64DKT89v2_bZMI-MjjvxzEv0YKckGG4X1vjkaUOCLYr_qTHgKZ8s-3pSQ6KSX4Obnqwt6x_k4yMN_JpGwF1rdxlsCCKsoP8Lyz0CLt4GqAHsoX8V-aQ-s_YDNLOusSMgOF1ritI8f7uClvXL09nmaeglGL8Mlxvwok3UsbQJWEgRffBDPzqBRp14uG4Yn7M1xFzG6W53C3ywKZJR6GwFOdf1I16B1KmyOogGN5nVzPtyMBBTreglS7B1MAoSZbAL1J_PQ3zkd0NnT-4uqNAoT_83rm4_yrNclqha0hxbfZh33x9gy-OOMIYsKLAHoGnM-LSEdYQpYcueFZqgILEQR3O9Oc66Pube_YTNDjVE63nrHnM771iOpVkBckZrTdB9eFbtXsAlr2bo-LPjK9xrCC3JQt356_4yQU350fks7VvK_K8J_sqGqs0NTe0w4Vu48vw3br4TvHqQU-HADoTvwt4MDvLabKMGkfuokQsHsRg7u8l68xfUE3IK75yWn1dKo8RmVKyqdklw2tyE-YilFPpI8yNTYxLOAf6YYx8w9OfPBBcQeGf4zGLQwJeiLMtPwOOqKlQ-wxElfPP3_jLrzcdyFEfcc_uZzgPqhM4T4mEV6um0vChN1LfdGuhlH5efv0m66fKWMqfRVZKyZFX4ePEKjZRYO3dhIYRYP4k51fy2zH1sL6_vQM_CR188AHQQhzyTQ9OQfPxVrvGbRDM_c4RxuBDnFkcRMtZaOgh0jOt6uDm4-naat2u1CLMltL3SMD2tYc8Mvv7RLeGuOgjzCb1gfkKBPaLB6RTtzN0ZiQmrJ7un2S_uAQziNPuZVLEjrxUO3Hkd85so9xZNtadphE_XMYvzD47LJglItfQUpWRBKa4zKl1JhPRM4QyUVN4geRVbZvnJNf2wOs7YrrlthMjs3dX4dxqEvNq6IALomNd8FRNjpAHoGG3OoOg0gJURN1yH_u4b5QMuLjJ9t9UzpUjPsMmQATQyKqsQ4CIPvYlSvZzmyUEJN7OoNcUsKsGCKIgsIzZ61_oqJnsx4XT8Gc5MHN_8cZVQJ4eU15BAWY_hciFVPkn7B38Zp23c-mN3KNr7fapSEiIeAt52NRss-M5iRAxzb5GHM0s_AvsU33MbJh2NgxrY4pdSnyFkFHtoGjwYtXnAAXm5HDWC-rR5FtYyvk8lQ-idbv-7tvcDcejxTc0zHzy6pI9BNfCVZiB3F_0oY5NXtsNde5SVINQ0AUy863732__oweKWtP3AwWSLqOdDJkvnRFpM-7w_ehPjIsEUKBX5tMY9l76eBznGcYFUlnyEB-8QlsaEpkUhl4acP7Y3uytfyTeT9o7RO8vDA1OUlGA6jBQSIGQveCFUPBX0Eo8uaBZWGDlhyYHzzwiH2fxSJCGDc_zjEazUNF_dzITZZHjVXd45gAJG1JYuBB_K5pPdpRmvg1kBxqvdagJ8JEoA_jJjuTm73WwYf93jxjhC0lXs5a-GMwBKixi33WpfWuA7M78jPcQMR9bwoc3qetEF7shfE_fhrs_gvQapnmXOTaWjB
diff --git a/requirements.txt b/requirements.txt
@@ -6,6 +6,7 @@ polygon-api-client == 0.2.11
 pytz == 2021.3
 vectorbt == 0.21.0
 scipy == 1.7.1
+scikit-learn == 0.24.2
 auto-sklearn == 0.14.0
 cryptography == 35.0.0
 ta == 0.7.0

diff --git a/src/Constants.py b/src/Constants.py
@@ -84,6 +84,9 @@ def get_env_bool(var_name):
 # SOPR
 SOPR = 'SOPR'
 
+# Signals
+SIG = 'Sig'
+
 # Misc
 POLY_CRYPTO_SYMBOLS = [
     'X%3ABTCUSD', 'X%3AETHUSD',
Original file line number	Diff line number	Diff line change
		@@ -1 +1 @@
		gAAAAABhbkFAqCXq7xPvaZfuBxi8_2MO0VgPhAgZtDpPoz980prWyJ8QYgSpuItCQ5KVcJRghvJHIYmdQeBwQy_yJHpdzBGBW91GLo0XX3KRtqLUvi8Vw7DgX3goCoSeyZLRP0G3hp5vvYNuNyTJ63SOGiO-TmvVwvEYJFe15PfDft5vyZp-DL6Q4is289-IvpXEF7ria2P5c3d0bpWsrvx3QdXNzoCcttXg8U-ntxibFQCjycfO-M4-CB0jYzDcIx1266Ebp1IDxoyt3wDSriUeZaqwnXLrGrP_SESkRRSBBWzXm2SlRP4Ph8tuG6Q56vv8yJHr3QfIVc4G_EnJvMa1RN7BsirKEe_y0DRNPi_1E8Yok32lSp4SD733pM25zetWEHDGbBFzEeRUBo9oGLm3O0L2s4OjkdaXMYbDev_WxtDMoNNNELMwCzsHVxCQ3TdvrUpI7YZGOIFRt_GitxmP5FIPxSShWqJHLl-1Y7brk7bOOdpkioET12Dyty5I-7l0N5pCaJG3ki6VwgXhE7SevkQHyjOvFhHPtbnF1kKHzJVm6PT--18S1jB-ygo3P10wRj5RU3mLex8RqhGLiHCPL7dSP1_ZpaTWsyzvBBkkjSvrHK-pFrZERDPNze8iC9VSz2pszxyyKFemwexGNMGqh3MYXC3R5V_U0jl7SF_J5iMPLWTrey7Kx9_i-qov8585xDbyTYNryM4ZrlWRjoX9j25kQVAD31hCsWtV0FaqhIzLhGHAFNoiY2rAkeRIf3wRmd4ITTv3vp-UzV3JS0-3pJFxLM2UZOd5GAcRDoV-Xbe0LU5QK3rQ0MbUTCCX6G4o6YMhBXxh7V3quXbO-McxuDG0qOfgvpS4EnOeuSfV031OAyrSaJo34aKC6lOFZ-yhfTgGejHVtmhZn07QWbvrUYOslLORfyZLxiOKpIAJ_TN3cgDg882JpuwfYRViMcH0QUgsH0jwQXsBDkZfB7NGb0XevHY19MEXt21ePLCoDf8_kqTV2BgI1VlfvZfMjBTO16WQxUoF0m8CEeEgHb960IGD8EFWL-Qc-T8kcO1buUsmiB9zNxW6w9BNEJAZlr5rumRddZ0TIk0maK7_DtggUxFkhnK1SJ_xJS9Hc6OMDKsJppjzQiNIszpb4CowXzBp2-9isdcgG7B2Z1t3YJlIu3C7FVwlBOQ3yNAWC923ImAaH9Rsbgl3G4A61KWzgLUNSKxQ17fhyvnrXgXgteAzcxLrheuB5KsIfD2Can1YJOxjlBlV1ng3xk0SXIGeHc9GznN8IMgHEy3_kH46A35NypmD1OUBvKkxNp2g6bbog7ACt6hqEqdUgEVa1IRJM623IeAa4EoS24W3_VFUrwTGbheR_E4PsykVI4GyBiZZWFANYJePlr-DCM8bdD2eNG_3tfl_oQ5NajWG0LACCHdyAmyLl53kPY1lb925n3_9EFX3QEnUAR85-I75O_6175sv5xhmpVLoXlVMze87mTBvL-aFeCFn_yn5E5KbsdoGHMZ_1hQaTFr35JEhCSFP0HZBgcnOwd3dgYmPdtmAFCZzJvXB1-BtXibOLt4gckW4Zag5DUWHkjxMv6JHA06Lq2mYIlN_8ZE2D3_jdGZ3P3KANaEOzxPZ_0SSU34pjps7Z6CcMFSVTUBCZfo5spaRNLtEmg3fyKuphfJoXVMtnMwd-HGS7LBKiXD5E2Sw1w59BP67CndEAF5HoLHv2xq1rg_QpfExuJilSJO2NIORoVmsCI4SUz50DyHlvkHrGwYzcz0wdfwHvefPGvKTJZMHBtV1L1gwQgBDepZ5vVM2dzUlnkgfFsqXjjEGJ5suwTh3sEy16l30Y1LjiStkogdsmFi7bEcxrdHs3I4TUwqzmjEFmPjuoDuRAL6UYgq5f_HItt8Vn89q-FlIpweLEJORsA6iEFwGayxabmWFP_H7Dr-S79SeGxvcyZDajtfRJpHnToM1q2tM0z8PLUv9R4jT_7UtlzwdzEp_JqKI0Hc-4OKEAKdzPJenrp0QTcBl868yAOMbEh7YOQXtMUBLTctrwHwgiJ4xtwl5GM4qwrZmP0VWcR-_AAAdOSZFVop2q_pYUxEE93OF-BlnjGMEbK_qoEKI159OPO7vXjBxu0acQzTIV2VAeeNSU6y7Y7QeetJz2Wb7cy_sJetA6d-XVkQh2_LVuUn3r9wlQ6_4vdpTZ9rLjCds3H0UBK8uwPgBn5ktKig-BK19fDCwgbhVtnsKNK2Z7eDNMS8YttG1AewrmFi3AcRBG3OWYgUZ1IrtxyRqqFNY794v0sLYmai9gtJBEdJkTSo7mwR_VfAba-d9gDHt9YRKYJr-hBhhbVCY--SxzTF_aWlXMkD_bgd5PkDMx3MkVP3rIlUcwQpQzcXJ0JnIaxCC1XRald0gt0NA193EY2NCYW50yjS_vvVjW7_Ar6P2tlk3kiPNfUUyYqh7aCw4zrGwJFuKFE22V0xYSaNIY7k1rFXVZ6eZmedHZN9mOdPG05Wcb-1I9dtYaOEJB8axKpzg6xlgsjfvu63ps_poiLgk8hryIy6Nd03GmTSsdv42gQbtGKOkJMS7MaEtS6wpH4jDkvCcL8eBtXAnlcaThg8d6yJDMYqqXLUrFsJuC_vKUIKVfO8MFFJ725qXamWuHP66_rcgpwn1npoSI2JSKghOyj7UfqcgafS4ntn94s85s8HUwpAgYExF5F9UUKZBPu93uzVrELqxgVCPc6vAFQ4g_4oaEZnJAI6eYm6_7QQTmX15Glf5dAHMKLE8sjl8pmD0R1JI1QsBj5_ZkARr8Yg83n5VoLWI346D26bbcRjnB6m-yMwWvotszKoeikEDDsMc3Q7ijcV8q0yNrQs1LuEZUrytXebuNiFDCe4omAKa71lm2EAMKO14cy5FiEaLOUJbQC_jKB2zyg63-GDi6UiDmswZCeOFIq3svpXyK9CkuFbi4rldrHW283JVh4k2DoHKe6qIWEJSIms8q5tgYSFCBfJ_vmksex98IIPFsEeUzApjbmrrCAnnLBoxIUt4hA9E4Xhkc44p_gnCnI8L47GvRIEfZG--TLF0d2Kxc7TlvQZ398xSFr153ElJMXkH7vhsW2VKdL2czk6Rng6r1fKRsdmYppGfAw3AiPCISJqEUDUlACAFf9gjfYUW5kxZO4EV9bGo6j5UgcNzrVKTcAK5rKP79myBjKwB-54MS1QTh96Sg818Zs9wjd6lHgwZs31snGN1qTPUIlqM80WjXXwNCNdj1C8P7EBRIV_pwVzqYdpl86GcQs0cVcUz2kOwZCaByJ5LmsnIG0t8hS1GxbGC5IiHc6Ef6R3BaYGkuLQzUIMqADgzEgd43qQ057jaO5O8JIO3h1ByQ2eWBm9gkIfD9RRAE4IwYyLNPK6qOr_J8RnAlM5Wm--h-xS-4WgvcJDmLdc2y1am4RWNH9vhPlEwV1qUcDShqoX8G68alaASyotNvOUx568w4o1vDA7VPMo0OQT_K6DOtbwAzjpLnZQgAZpaN-FSK93YJrGhuYZp_mmf5vpa9f1j76F08iEPahF8ztBAIezUkrnxPiahgwAHPjuNllkSoJD08taPEKswKwjzeWnSFjnAjcpCok-g0E0iLNxKXaEcz1tEVd4_yN7jFijIFJLZJrRS88TGsjs6qn0QgtDfi2keaZ6l7xc-bP3znd1P8YX1M_n4di4uHHyBfB3HcMl98jLIQHf4H9MR8iIrwV3A4UhVnRiZ2YOMSuW9C8h7eyFPgrd_7I2NfiH8vi4j9CPJJpJ2K2UpfU7_OaslKy2CB55XOmdEauxUoSEF3_ZiIhYCD_iw0cZ0B2wqxK1qwR87qnAhA3S5prgCHJzvFlW6KrpaBz7ahEjddeyxIzmZoeZ7rP7CF0C1I4UIVbfuzSTz5xDGC9uto1rTETnNFtRj4srBQE1qLsvkp0VVHjEJvd2Y09VHsD8OtlJ7zg6U3Lc6gcbnQb9sQuOiU0kZgb4bz5TwB48ORyAfpgUzPZ_t-Y4KAtyyWXFIMc9Tq90n7nIeHhG7XF8O4hfPhr0s5bLfTdaptK9bEvtFtoxasVgwOUuQKwUwUbCdoR4kP47zMIeEPv2sl5DniTMTBcPA1uQjnec6z7pEzFnZA_XUAQVBhziOqLvSKxXC4wNgPJoYUY-MU2Us9rVSVwfK3gYcgaUq1T9kIUyrpRfQuHv3RbAPCgeLwHqxRq95iVLvj1FOfJyq-SFmqFC5VgbWgg7ARFpWf2bQsWG48igp0xyA5sqBeQFiuK_KaRCZLn7rcfJ1Bd8=
		gAAAAABhiM3JJhBPTP4WNzaNOZvuIG5Oi_74X8GHY1_KxedZF4D6iYvXRQZaC6QSQFE31ozxKKFOQ6gFrHrke7PuBgqOa7We0D3958qvcQiY0veuIR_VEzjOfVRTMWlngXDO2HxQqmwat3ITZgrnrAPs7eB3HXa3S7O6MMGXZfOQH8wgAu1oMKaUxodtJtKH1O1odiQ0uBx9mT7tP-dnynRE6Ct81UrxdCvUpOd5Cv5gG5rVW0v163huo3ZWqsysdhLPQa_OCegXDciZQeDsIi6TdzQjwBMr01oIYNHYZofJiehB3URzWeAZX32xustj7Hi8RW2_ZSVZkjNXfSwaEF72rFCBahsVAAzFjjgbbM6yfgJL8HYWep4vNrVYHqw92uZPIvGKN-dq6MkVlZLxiIOk96o24kJw8YmUuacmGNbtCO5jN0J5v7pS2G5tJVQqbcOwNJIqyxCG7b2VsE7esnglds8CLxMrKBsV_BW1d08OIyQzQGIAldjbvOcQBadj0DMkpZjBuig79QonooFEykxjnr1n1CF8khNY1-B1gWOWc6NGssD4UneTSSHQ98lKjKS5cFjmozO05RKeGKjGEhIjANee2EdD4QKKMAPbijNpRtnCKzewtWEafcIUOduOLoDRgnCkXkvIc-tGO_Nh1vacviznTqgK3WRAYHHG7m9Q4RjkbIY3YjN0SadZmt4aaBvz0SbIfrjVJMfA969ZDEVYQynNaYHIT6WIACUM3qw9vQTMTD-Wl-2uoTO4WTl-uEPQIKPOYybvKNgBMT63ffqQjiYyzSslFKXpR0T3iIOS0tuI4Y6VCzDs9PHDzPtE2W1K0xSGVr96AUlXRCehzxAPbsiV2otjM8kEFkHk-HKDVidJ1RieuCfC0iFLOqg5krU4rg7IZKJ_0crSKDn9d9cVmOX10ZypDe1TqwdhS4spWdedTryAyVdBpvqr8OyAcPCR-pGUYKXyKaHICJl2GIWWGlYM4TfsG2dEDII9npf8guLxJIevpCGnaTYlM5ybFBvdGv9mpIKRBRjVyNl4ceAOQEbUyMiD5XjsABFT536B-tYJOjTGOs1-ZsXq-vnTYte8FY59uSzWsdsp0ivKzyYJDnFIsf9nhyfcef4HgoP-1Aay4-L-UL0SAyCLRTFSRQGrRlhNGXCTVq57Zzqa7WwRqgjs3nvukBPeYkD-czPT5uDo5AZtGgFU7q_4JBfB9hY4kRyZomlHK5coDgG13z4eXo3wlT8SMu7NP7F5tpGbF9J9tnA-vMOyz9BDUTSChEyBYUhID5J8OjIVV8BMRQsotQsc5FNPeJ5g_fLLhVF8kk8yPm-FlmwPz-p50lJMO0NvOxw4Pfm64VGOD4P_tn3M_pwmuue5t6qYcy4ZOzcE_Ps6cHLU22rwMC9PQWUlvOUAcGUILBID3fZlM-IAhBpN4jmaS7nkGjVdBB3hnqOCq3gLOeQ7muqZYuoABtBOxzp6klYpaC3L2UcEAKWPY78khE3gxhG4uIfcmQu0ZdoZXd7nQGxzcQoiXFZJKEofRhJxUZkBE9nTZjSzflgVM8OJufvYmBe3PBTuX7wNEPOjW6X8wqqrhvnga5PtOth4Yk7e8Zz98tPn6mI7J1GrCuJEFczTbD1nXcnlCYJsW6nCvll7rti-HD3yumRl9TcBp-GhlwHn05qhVMRgsWJloz2Z3BX8TLfOdYmFuwniJKFDaBo2XLlSdUxczaQu9bupt0udyUPuX_hFFrLknDlErKxcyj32tRRIzpfcQjZOnPcCBNtmzcDCjdikCiez6WnQYmxauNuU2bJRMflllJqAqu3XPi5D_cKYEbEqKUOiQ6cdKbkcqK8HJx-6ovHqaApyf4SexkdLRLAkuDAohzNdM-BSejqxD9AWIPh2eQ2JFNRsC3VXaASOZLWYHGIQxnX67bWuc7ojDQccTkfFjiJlqZah1KjwB1FfU98ppZ61JFSf8CVWXSfabaNDT60HdqewMAMB6V1m1vBmkcT8arP3okeXfj6az_Sb0su7Kw72zhBzO_FB_LzsJmCe_RJrSaKfVC9B1TSFA2_W2LCv-0-ERMpkbELtIfKkPSfjXV-SfaUeC8Zsu3qLslYN4nZ0lr12ImeMDK7z192SA1ZjPbKm8J7LyStN3EsmZxYpdQuAofK7TOPL_2X16Wr9n9ScHmBuCYUlJQ026whekBenTrUfd7G3eKq0VVg5qTECR5wIdsO1_zNjRqjAsLb70QKHvShuYmMShWUhrhrIDUiPz5zHDdoDijoMi_PPdMAU2MQL3tvLROG4LjWZin0u0abTxClVwpCB4c6Cxp6AYIkymnNN0-VmruCGVA8Pr6BG5-x5v9CDCqSpzJc1vuk3dtMbJZy3FCKfhLEZaSjLFXSWn_-2iQQQ4Xzsi-RkkVmOZaaraxltif_nQru0gAxn9tN-Q4-hT3iC3iuxiaWUUyh3Ujd0t7RSb7BfIhXnG4ilzfbzipcbAALyXNMd0OJd_4_KW8L5SpGmKsduo08_U7pPqSlr1vv-yW76zpyRMVp2epHjdHMI9pC0C7h0_rYVf7TjsN43u2kfoiJuKXmf1XnYU4EL76VvdWnhoUsAqfhtG85fHnvmLmms4zP1snvZGZ_xg8XS82q3TQkY3SO2EiL_yPyue0d7pexv_MenNCeG-PPBdEJGCSxdwL9e6pXwEiG4lGiZkKXMgqzN8XHPjc5KqVhuSFgAkoFl2-FtUfv7U1bFtz_9rvvVpgTxQmmdzls7eKlTDlihDq8o53zrzvK_73G9SBZuoQf2nIRmIsmTX-9OhfS1qOwJsJXDcXS2EiM5ETq6NnZLkqXebEgQzNnF9othL5Tu_wZGYQU1PXvigjzCK5akQ2692X7a62MLmCLAFiDfvNSolvPc2fSZKYWzmjVAXCczI0E8-_oejv9ZLTAk7-sVd26B_jx3Qlxipwll3TLw7JFYuyxyige0ectUuxWM044c-vQnu9bq4kJCdNDdrYlc2who5B_x4Jct1NqDjMrvM4ojtMoIFBRfDtUIoHy9Q-_uoMZWLZv2tT5luSJpnc3vXzCk6TD0Fszy4RiHUbHy2YXU7lSo4suLZ5mrGB0JNCbHzKylaLTGsw4deOOAAtHbbQ8EwZJS0aJ4sReE3_4yNZPsEBHssfjPXDma-Fm264YfhioP-iPDqVv2X0bhSzu_kVbG_F971SEK5TTIp_Gsl7Gnl1DyTbNDyrnPyYJvmisCdywcgDKl0RpyjAuWVy_EeMUY-pA7mqLpAVil9r2HSXMSM516yRMotXhXAxXQc7hn0n95xKsVBFpOVKYm56PFbhkdctCKWZwrzl2BkHby67qYGZaoghBXVg0Hcqy7yFhvjeR5l2Tp6-LyXw9KSQhLHtPPWQszbqU7eI246Gr3g1CMY4Zqcuk4HXQ1XOoOn4DK6ANvXlMBBCY3OB7DH2UBQD7pwld-w1RotwdN_Tcbcg62IFUVrTBp0krcGaCX5DdpyxReguNaPDZZyE1k93U8vVSTuHw1XNlJb_xRtSDXJ9VLCp62rerVNIEs-wU3yYcOItuT2__oMeapVDp2qtST5F_4UPOeaEAjs_hn5efudy59H1O69M3ilysQR7UjL6PLvKHxmdXUOIX_wplE9TXkF9h0wM7rfa4V77wjgIhlv1l-KJOAIob_QOxCJMJnTIFmYy28E-BuDNJYGSFLaplwWIazf_8WviVfPh2RHxS6BclzJKoT7in9KHIlkGegAQkcQ3V46sPPI5duo2Z9EJq1dcez_aS9xYsF6VcYwwrRF2ZYP8YRjQNj8jENvcNvmy9_CbFUbqrS2U829DWb-6U7Ig8Y1X0RiGFCANHdTNM5thFsIDlh67e08FFMGxkkcPDXQG_H_aV4b71qFMIe3dm1uwQWnswdkk2c1HNvbi4GjXQIts3IAN_bPurJs4I_KgL-XlkG_S35KM4atPFS30pbyzQk-Ae8oJpdqpWd2ep7RR9YT6m1kSLhpNIFo1Qqwcqn1Z4THFj9EOvcLvymuEDAJuZC1IDLwAktM1Aw_bg2u8MPbi6SmgyQVG73RccMK5f0pqxaFrU5RlUOKFo5OyEH5GWxVTMmr7B4-5z3-y4uYmYiIT6COLoXMdEcwVRSDDV_fzmhKBAJqpDef5AAFffuc70AjjkTF10M5tXZn2KuzTEnxJIoDU2y74lUf0vG49Hhrii2_bcrBfbr-dnDAEH0gBBowodTZltv7h6dPhkQRhM2ZQHX17i_KhMghUNgKagtIxZMPyPli546MyuukqA926QV9tN9MGP49CELjm3Xzcq6yGR36rnWqFxj1xvY2qlTQ9i8jOgrb-gNoh4FNycUYW8EYJ7EZ5mRkMYWSaR0NlXOcgI4KS53wGrJZMQOVHPLxdxeIOuocchGVQl3TBtXeRqL08S9jfPYM99Vqsklv6Xha-tQGuUm-sDEwUzbHjQV7UxFpj-AS2_FaHnqxKhAOw2mnfWOkUcd2dlM6vNguwVXpIkFXn3QC-vCL3rK-EGzdlWr6UVlbSL_TYu_tlHfggqdGJ3lVojRSeQMxEyuj7JJBvNrSUcfC-GsI7gVWvhnedZgTivL4J8nNo39J3LrLRv21fcgp_Hie-haP-jIhHwtLb9pX-jgxb_vLFL-t8gq0d5lVyyGqvzLhn9vVgaJiiNkVdHxFsO3hddeH83lcNASGJ9XvQ1WtBmMiL6BCYBeDrC-kjtrb63oAM-qnE-0qgKvws1dyWpPCKBhZTU6M2khYmy7agLRhtqz7vBDtLGIpVeqsZ4cvBhYsI65aQZXDzCONVMc-VkV8FmW2IjsUXxvhuNLCRZDZPIlN6FxYW2L0M1JoKfWI05xAgq5DTCRKxRdUZNcD_OL7bAT_6KSu5pi-uueIvUQ5kKO6UGmwdxULp4t9CiX8AXalCY6tup4HP9cutPR1h1LZaXh99CImHWd4ZSd8IyEUriC3MHS5q5VZxfJTWt1b8Hoxa5qEUeP142X0leST1Ewlqxzht_5Ig8U4vnMJNngtAhZby8mA2svbWpQQKj-YTRMnBgqjzHp-OB-Et4ld7dpMaFYlbIoI6B3LwtlRAAWOHLzujeqO_E_MshTuqnfahTX9vbPnNsXWLUQ6VUhPAXY2PGiZ9lIgHO6KC9wbHthHHgtevXfP4OQAtcmtgOLl-eGwKq70dOwamwvhqfBzU39A4kYT8SiFjgecQxbSxExzfhrWaZmjdaPef08Ple7yTF2lgdrImqJ7hMEVv-n8gFtUYy-B7O7DxJBRy8TLR2VPsuOCQ3I40oz3PmnIVVOVsHTU1qHKk2KLiTlX7qvK-4z74yEr4p9fdT4d2cG7g3n-_TLCYdCnzRmI6XFavLyB5URF1NAGfiPliK5i-P5Bh1zEEpp9yJgLqjj-tPxBpTbRWwL9kAe9UrcyCtt0bAGAE0NNeIRLQl0iqtH-04Z4SQzEZjjY2SSnJu5N93iQ_rNlxDnI2g8i37lpOSgnzt4hY8MJW8ZX1LoBlEA_xW7x9-Cd_hEYuaDo3h5anaQNf_hicDvm77aoXetGJfOCPWZK-JvnjQL76bDPj8oaEEVzr-nrzp-_eSefDDnPDpmeIlFhcHnKlMmjDQTOe1GDk0AxCBALk247HS3FH1HEoklGQEoB2oecG6AbZq_A3pYLdP7slQC19Vl1XMiGrtl_6NAISMLTnwSvDu2n4lv17Fc4N733ScXdPkCAsA-jYsC-ws3u5KtVAj0rZhY7w3ivk7sup9CEPE6mFWlzr3fz8tI0rO_TvmaDOhRV5_dODuGuX8X39BUgMk_Ug-dbp1-iRkGrlE0skOPYxfA_kxYVQESZtoZqAmOXYr7lHe8dpSGYp3JNZ0hNLVItbzMW5bLK8RRJxKl8UlI7SpoTVxeAQcKslpedYvymmO7xe2RhBsG4NfZg4JLmRSyio90_EBem4LiGxWAdLpXiNHQkO8KQ5Jwomj7_JtEiiHTSvMoyLXWECfFJOeVe6htGBQpIMnKuC80C0HeXCKu13GpwGTI6j-o45bBZb1j4_9ZyWzpQ3EbO3u1PeoZkrdyMaQh_oyg-oln-n7b6MO7OcZp-yfBj_xHqhXkwhPyEZj9qJ7bqj3XFfqPjVibfcoIkiclqD4JJsMQzMRJvbNBAAmVfcgp_QYu_a-DFwBt8iHwTW5xpIwZpYhyTLQFfmOMWuwlIhEovq55_qOt_7WvVgqun3gOJQhl3zlTsKsRIth9SrvwwD4wsRXBOnZmBPGErZxVCY2ieiOirh2u0UBJytTK-DuTydwIDF1a-ZVYkbivyGuo1JtFePmlzWuvHlhV-iXFIsTbSeVJZWCBTNT9NU6ViDDLKbbsEiQ-R3zpvF76L4-KsIIL72JN2AZChHsuYFlQPOo88Vdrl5OtXq5efE0Yq0fOk_c1_J-qi_Z9lmos1L95TVa4fWRBq4m3VySXMK-btSNG3RMa9rN6JRj5WMFeYvc_zlUkxdz_oASTLa1kwOk1yhKNHIn0G2HZ7Dq5gcLLU6WTUdfNly5b8ZTW3QuXQaJ0-kf4hNOtVn-cfhCRhpQZE-UQmHTM_JtovIWAf_1ODrhwItalLA==
Original file line number	Diff line number	Diff line change
		@@ -0,0 +1 @@
		gAAAAABhiNfzR8mYiYlCVUVAOMROC2txWlDSWnqprOIpu0Wgqkjk1TqoqtN5O8j9Wh6W4alt2dS4Ma2ijSCAyWaGWDC-kOeOU6fLaLZqzhqjImTCfCXbtEeMrVVHkPrZSP8sOoTv4RZx48br2xbg7BCvG9SSYMLkRhNh_JdpgncPSY9TAREEkW91G2XZgtvW7plKmDLHCsJDR1ZOK4j37mzVPGFifBcFvXtOtOOb6MirX6qO_imgj46TvNqBlOyVp5oKA8NyJw7uuYCn13n7zHZgvETN5gryTER9LQdqzhz687f8oeGDk3lwZENWrzIBBZcJOeWKjgFH6XsJFx0mDleS2D0RBnPhAfhNUQfcKBTvDIb9CPAcvieZIvZCpfoX7uUisnBj2kCA5AhBOGwc1FZKDcPYkKYQDRle9L1L169XoNzt0yp9UmizJAQbWIi_vl2M0K9EPnjW_2xdoEW_T49Emc-qL1i-8UmSGqPY2olJndh_dW3YiiszPRCTIsHNrFoiK-skEtDk1xpeWXhezXvGO735lXt_oQWSGuuYQi2nxgxXuEvycN9JFwVbs4VSjSTWpR-bIzkJzMdwJLXDuwcq8GEsp3wKfeIIhrp-gVHZxmBJDTib57dkpR_Nn72pWak8oJJj-CWT8_fhYwP8eSB_eYg3PVyKu9K83jP-cU-XdWb80qZXzM6mATP94OzUvNvJNPEur--vICcyYvEo67vgwWdl8e3bAQqa6JRVeDTPAeoPdgL0cR2zig1iHNDUTuaa6NfEGIvCcDPxGfEz09eg5DtDrMTyK4dM4kQ6mOKiIf4B0fkuPFdVgKul7v3V44CyY4vkCR9cUMxqC04Q9FKr2YkUSuLWlEVpdMpWBjJrTMbc-1k1WXndH8g2gC_T2pBAqM87QLCUDgbWtwTa3lwJfvg5dUZpKDb4zYDN5U-VeYy30AM7xJwGPaam28yOuR1hHVnkF7hVdQ1LGh0W1-t5GviKrcnooO9dsfDChc_0DaAKopIgacDXZfb_iBAJAF4dqvVy755PjkkUWdjeUvZd7Xqs89WsHgtA7elvAgdqHddA936fjIsEHLLdqlv-HbVqXghjA8ejk5OLN5Z3607cmmsSGWfOSosazpHXPjEpfpkDvDfler9GJCBAWzDdtZG0k45NFc_0HpuoZFm2fb7dJGg4bjuPyqFn3PlBdHgFQJWTMkbuz3amMrQ4P43KbjCP2_OKw_ciRxHPJAk2mzkzZFD_GB3ASqvws-JbOubrF5VQGzuQD9Y6AsSAT4QysIKD0CQOqX7BW1WBCniBifaWm7h0NiZtnKTYy_aKLmnxZkC7b9qa0fIF0TMySrRxlBqek3reGGXbK_Z67wQvIQV3U8gFFMDIPQ5D8KYoafz43UbLkt4KHtZBsPgnTo1411mWJ2A6UXDZncRJyLDkx8g5g94xjiPEmqwFfm8QhYlXThFVy1cHqoe55Q2yhLwMtFsA0kwcKkIL-EPmcoIqFhvXhtcoMDlaPw8zH2kEWYcuGtZHDy8fcxHSfY_sPDCEvmiENDMiWHOCLW2K3SgTJBdxKZ84QJQgGJZZNRWg1xR27pSaDF0PRLn6AA_cR7WrSC5e62gPLvGd7CLQy-zvVmHZdXeaGZxbmEaoqQUFT6RjSz75WnWHzWQYcMsZu7bIwSjEdxareC10oEyv_nWjHRydOqQMWt_tF0DSzZuOVERPDCZVNVSp1CJdApMGbumqNrLLzr0XMaVo7etObHgmQ9hNwlX5THFIH6jsqrZAspsRyuJ61HVlTeRwe-fT-Ct8LdtXQPmUXSp5yDoFZvgJEAe3BIFgVAufR_QYYvzSVzlseYR0VWKAWwOqBR8DteTtG_8FnGgL3-u1yvSRM5jLIPjqmEe0Lo8s9P1TyYSZRPCTy21h56garQCM500gekU52lkFbulBu_wXw3B5SMpMrzF4S51LPudUbDSAoM8RxsKAhro-SL0_J5eZFQ6a4TFmfb5k3OKqnSpHtbUcmtr16jIXQ8DjPPLE64Met1ZGGVbJdA213wHn_k4sziUg97deoqlLAxvDKRYGJiHG8f_1wVtSsT2cmUnrwiuEAyBtVw7YqHKl_U4Vs8wlDil4n0pOG8L4pOF2xjTlXe1vHubJB5YMOeEU_ZZDGBoPz87kXRGLIN1wX68bBbwUPq4rXLkW1b_okhkLzw5VgKNxLCiJHuddKHPb4VRj3GlSN4a2M3Oc78EF_SQYg_AeVwW7loYGdk0qZ9EG-DiSy4iV3FZ7Vy1LlM3NjqFZxOAYvwPU8zmeXFA4tHVPjPkLzH2Bi_8FQeEwBgXl9s__yIz9uZiQ796RW_8649FcmUU6R5GMZXA6_yX88KNnNAMVrQj0XAzQq8HDIZzCn_XSb36JdAAXkbasZuhv8Eu5rqUgvjftK1elcJ8k2Ja6gJ0RSlujwDaVEWkiyRh5r2dc7QowN5cyT8BNkhxf3ACQ9cnAfn8OAgOLOwqzx_o8lQT2q1bjnHRE9EwjskCqLWlVqBD1neoegCN24GcwZI73A2cx1BT-TUrTyg2ejMh2fZ5e1zPXzBnCTjSGpPjKLpig9chCASWd1dhfSeazeKZpZSKaNbdDKY9cYMTpAnsdXNiS4DlG1pDKjetxrq5hcmCnALb6E9DExGiwQGi__6j9t4vWHuonrlw3akm8RG1blWr6wRQaP4aRauyCBrKml5LStAw-8sb8qYN_TR_MuN-s7IPvDmRZGPat3xs762zdjVB0hDStcpdvZCmAQdJTaW6APwgqsI_M5sO6YIjqVWIhu-CdXURJelEm7-acTNYQ0gCZ_ikGft9kFhCs9lLN7JSjABEm8UMxCUnwFQRu3rdkZHtjZu7TcszU-wcrc9H3HmuROrBtipq8zf26QfkY2RRqV_Y64y3qFqdUtZ4r4YOJBEYRiRKroEm28WKAHx0iblAjB1kS7Egy3gp7KdTa-YuneNlqBd4Y-srMni_oFCaLSg9__hflWjWhtILOIUltKNLuIaxl2GR-2wzng1DeTmp3_TU_9StYpnEwZuHH8rDrG0_PL0CSVhHZTioRXMDYROS_80wdx3AvB6mvKAANXlen04Lqj0hcI5pQXWnOFlQyqafKmq7Oz0q8eTA7Qj_FRzXywcZqg3md30GY_FwHWlF6k3NhIE7EbYGZDNE8cZRllQzYJWwemQeNAfhQX0iY0qR9IgGcV7wUnfTX7w3E0ZvjyQcTiDyBHZ0BdHo9LS9HOw4Y6HpHcVlVj_vtyBbhih8ao8N7yU_Ya5ydM28xpPqGHkShLq1dIf8FJoWSFnctHLJdsjfyKDa9u7ZCt2XzAzhnalHF27NPCs4yN5fLFr9iAGhCO1rdQ6WYKw5xq2AyAY6_8V5f-_3CeVXvv6xTax8laN7YftDBaQP64DKT89v2_bZMI-MjjvxzEv0YKckGG4X1vjkaUOCLYr_qTHgKZ8s-3pSQ6KSX4Obnqwt6x_k4yMN_JpGwF1rdxlsCCKsoP8Lyz0CLt4GqAHsoX8V-aQ-s_YDNLOusSMgOF1ritI8f7uClvXL09nmaeglGL8Mlxvwok3UsbQJWEgRffBDPzqBRp14uG4Yn7M1xFzG6W53C3ywKZJR6GwFOdf1I16B1KmyOogGN5nVzPtyMBBTreglS7B1MAoSZbAL1J_PQ3zkd0NnT-4uqNAoT_83rm4_yrNclqha0hxbfZh33x9gy-OOMIYsKLAHoGnM-LSEdYQpYcueFZqgILEQR3O9Oc66Pube_YTNDjVE63nrHnM771iOpVkBckZrTdB9eFbtXsAlr2bo-LPjK9xrCC3JQt356_4yQU350fks7VvK_K8J_sqGqs0NTe0w4Vu48vw3br4TvHqQU-HADoTvwt4MDvLabKMGkfuokQsHsRg7u8l68xfUE3IK75yWn1dKo8RmVKyqdklw2tyE-YilFPpI8yNTYxLOAf6YYx8w9OfPBBcQeGf4zGLQwJeiLMtPwOOqKlQ-wxElfPP3_jLrzcdyFEfcc_uZzgPqhM4T4mEV6um0vChN1LfdGuhlH5efv0m66fKWMqfRVZKyZFX4ePEKjZRYO3dhIYRYP4k51fy2zH1sL6_vQM_CR188AHQQhzyTQ9OQfPxVrvGbRDM_c4RxuBDnFkcRMtZaOgh0jOt6uDm4-naat2u1CLMltL3SMD2tYc8Mvv7RLeGuOgjzCb1gfkKBPaLB6RTtzN0ZiQmrJ7un2S_uAQziNPuZVLEjrxUO3Hkd85so9xZNtadphE_XMYvzD47LJglItfQUpWRBKa4zKl1JhPRM4QyUVN4geRVbZvnJNf2wOs7YrrlthMjs3dX4dxqEvNq6IALomNd8FRNjpAHoGG3OoOg0gJURN1yH_u4b5QMuLjJ9t9UzpUjPsMmQATQyKqsQ4CIPvYlSvZzmyUEJN7OoNcUsKsGCKIgsIzZ61_oqJnsx4XT8Gc5MHN_8cZVQJ4eU15BAWY_hciFVPkn7B38Zp23c-mN3KNr7fapSEiIeAt52NRss-M5iRAxzb5GHM0s_AvsU33MbJh2NgxrY4pdSnyFkFHtoGjwYtXnAAXm5HDWC-rR5FtYyvk8lQ-idbv-7tvcDcejxTc0zHzy6pI9BNfCVZiB3F_0oY5NXtsNde5SVINQ0AUy863732__oweKWtP3AwWSLqOdDJkvnRFpM-7w_ehPjIsEUKBX5tMY9l76eBznGcYFUlnyEB-8QlsaEpkUhl4acP7Y3uytfyTeT9o7RO8vDA1OUlGA6jBQSIGQveCFUPBX0Eo8uaBZWGDlhyYHzzwiH2fxSJCGDc_zjEazUNF_dzITZZHjVXd45gAJG1JYuBB_K5pPdpRmvg1kBxqvdagJ8JEoA_jJjuTm73WwYf93jxjhC0lXs5a-GMwBKixi33WpfWuA7M78jPcQMR9bwoc3qetEF7shfE_fhrs_gvQapnmXOTaWjB