ADD: 3 more festures, which increased the avg accuracy by 10%

bszek213 · Apr 3, 2023 · 5dcdb4a · 5dcdb4a
1 parent 4e8d0d0
commit 5dcdb4a
Show file tree

Hide file tree

Showing 43 changed files with 199 additions and 70 deletions.
diff --git a/ADA.csv b/ADA.csv
@@ -1961,4 +1961,13 @@ Open,High,Low,Close,Volume,Dividends,Stock Splits
 0.371192991733551,0.3869589865207672,0.35148099064826965,0.36061298847198486,792747065,0.0,0.0
 0.36057600378990173,0.3843640089035034,0.35505199432373047,0.37204501032829285,435026934,0.0,0.0
 0.37206000089645386,0.37272000312805176,0.357028990983963,0.3600830137729645,305578157,0.0,0.0
-0.35968679189682007,0.36287596821784973,0.34999141097068787,0.352222204208374,212685360,0.0,0.0
+0.3600989878177643,0.3628759980201721,0.34995201230049133,0.3519960045814514,211460876,0.0,0.0
+0.35200101137161255,0.3614110052585602,0.34979599714279175,0.3562450110912323,219918828,0.0,0.0
+0.3562529981136322,0.3569889962673187,0.33873799443244934,0.34551000595092773,302198799,0.0,0.0
+0.34556499123573303,0.36945900321006775,0.34380900859832764,0.3683539927005768,319572685,0.0,0.0
+0.36830300092697144,0.3895060122013092,0.36743101477622986,0.38158899545669556,472185946,0.0,0.0
+0.3816100060939789,0.3869290053844452,0.3708980083465576,0.376459002494812,391063045,0.0,0.0
+0.37649598717689514,0.40793800354003906,0.37508100271224976,0.39892300963401794,637648499,0.0,0.0
+0.3988800048828125,0.4012320041656494,0.38275301456451416,0.38538798689842224,355308379,0.0,0.0
+0.3931959867477417,0.3931959867477417,0.37731701135635376,0.38175299763679504,300032590,0.0,0.0
+0.3818819522857666,0.40330609679222107,0.3728748559951782,0.38308703899383545,578428544,0.0,0.0
diff --git a/ADA_model/fingerprint.pb b/ADA_model/fingerprint.pb
@@ -1 +1 @@
-ֈߐՐ��b����냰�����⏅�� ��������(䊘��ð��2
+�����Ĳ����܍�����ވ�đ��u ��ߺ����(�Į������2
diff --git a/ADA_model/keras_metadata.pb b/ADA_model/keras_metadata.pb
diff --git a/ADA_model/saved_model.pb b/ADA_model/saved_model.pb
diff --git a/ADA_model/variables/variables.data-00000-of-00001 b/ADA_model/variables/variables.data-00000-of-00001
diff --git a/ADA_model/variables/variables.index b/ADA_model/variables/variables.index
diff --git a/BTC.csv b/BTC.csv
@@ -3110,4 +3110,13 @@ Open,High,Low,Close,Volume,Dividends,Stock Splits
 28158.720703125,28803.3359375,26759.99609375,27307.4375,33382021890,0.0,0.0
 27301.95703125,28729.84375,27183.36328125,28333.97265625,24220433689,0.0,0.0
 28324.111328125,28388.4375,27039.265625,27493.28515625,25980310960,0.0,0.0
-27486.4140625,27780.93359375,27220.1875,27456.591796875,13416822784,0.0,0.0
+27487.337890625,27791.826171875,27196.234375,27494.70703125,13383005987,0.0,0.0
+27495.5234375,28178.14453125,27445.046875,27994.330078125,13878363192,0.0,0.0
+27994.068359375,28037.92578125,26606.689453125,27139.888671875,18188895178,0.0,0.0
+27132.888671875,27460.71875,26677.818359375,27268.130859375,17783600385,0.0,0.0
+27267.03125,28619.54296875,27259.662109375,28348.44140625,20684945906,0.0,0.0
+28350.140625,29159.90234375,27720.16015625,28033.5625,22435349951,0.0,0.0
+28032.26171875,28639.896484375,27583.71484375,28478.484375,19001327598,0.0,0.0
+28473.33203125,28802.45703125,28297.171875,28411.03515625,10876469901,0.0,0.0
+28462.845703125,28518.958984375,27884.087890625,28199.30859375,12284641999,0.0,0.0
+28162.6875,28440.708984375,27644.890625,27764.0859375,17633808384,0.0,0.0
diff --git a/BTC_model/fingerprint.pb b/BTC_model/fingerprint.pb
@@ -1 +1 @@
-�۴���Ś����냰�����⏅�� ��������(�������2
+ɺ���յ����܍�����ވ�đ��u ��ߺ����(����ݑ��s2
diff --git a/BTC_model/keras_metadata.pb b/BTC_model/keras_metadata.pb
diff --git a/BTC_model/saved_model.pb b/BTC_model/saved_model.pb
diff --git a/BTC_model/variables/variables.data-00000-of-00001 b/BTC_model/variables/variables.data-00000-of-00001
diff --git a/BTC_model/variables/variables.index b/BTC_model/variables/variables.index
diff --git a/DOGE.csv b/DOGE.csv
@@ -1961,4 +1961,13 @@ Open,High,Low,Close,Volume,Dividends,Stock Splits
 0.07687000185251236,0.0776429995894432,0.07213400304317474,0.07399900257587433,693511781,0.0,0.0
 0.07399599999189377,0.07893399894237518,0.073294997215271,0.07741200178861618,637122733,0.0,0.0
 0.07740399986505508,0.07749000191688538,0.0733880028128624,0.07468000054359436,396748009,0.0,0.0
-0.07468792051076889,0.07588913291692734,0.07298284769058228,0.07375582307577133,287920000,0.0,0.0
+0.07468199729919434,0.07591400295495987,0.07296399772167206,0.07404399663209915,286831067,0.0,0.0
+0.07405500113964081,0.075375996530056,0.07372099906206131,0.07441700249910355,228916506,0.0,0.0
+0.07442600280046463,0.07583100348711014,0.07175599783658981,0.07289499789476395,435284210,0.0,0.0
+0.0728989988565445,0.07392500340938568,0.07193700224161148,0.0738229975104332,276514415,0.0,0.0
+0.07382400333881378,0.07673300057649612,0.07364200055599213,0.07562699913978577,423884022,0.0,0.0
+0.07564199715852737,0.07664100080728531,0.07368399947881699,0.07424499839544296,372289117,0.0,0.0
+0.07425899803638458,0.07772199809551239,0.07380799949169159,0.07703500241041183,378314248,0.0,0.0
+0.07702499628067017,0.08527900278568268,0.07614400237798691,0.08405099809169769,933070185,0.0,0.0
+0.08185900002717972,0.08527900278568268,0.07784800231456757,0.07908900082111359,929778146,0.0,0.0
+0.07901941239833832,0.10230608284473419,0.07663155347108841,0.09190445393323898,2925777664,0.0,0.0
diff --git a/DOGE_model/fingerprint.pb b/DOGE_model/fingerprint.pb
@@ -1 +1 @@
-ֈߐՐ��b����냰�����⏅�� ��������(��爏	2
+�����Ĳ����܍�����ވ�đ��u ��ߺ����(������±�2
diff --git a/DOGE_model/keras_metadata.pb b/DOGE_model/keras_metadata.pb
diff --git a/DOGE_model/saved_model.pb b/DOGE_model/saved_model.pb
diff --git a/DOGE_model/variables/variables.data-00000-of-00001 b/DOGE_model/variables/variables.data-00000-of-00001
diff --git a/DOGE_model/variables/variables.index b/DOGE_model/variables/variables.index
diff --git a/ETH.csv b/ETH.csv
@@ -1961,4 +1961,13 @@ Open,High,Low,Close,Volume,Dividends,Stock Splits
 1804.95703125,1821.458251953125,1717.76513671875,1737.717041015625,12484103560,0.0,0.0
 1737.140869140625,1853.8880615234375,1732.524658203125,1816.4051513671875,11012966854,0.0,0.0
 1815.81982421875,1819.935302734375,1730.3494873046875,1752.0447998046875,10913669383,0.0,0.0
-1752.5189208984375,1762.674560546875,1719.502685546875,1739.0853271484375,6727498752,0.0,0.0
+1751.469482421875,1762.674560546875,1716.6519775390625,1743.7647705078125,6614309912,0.0,0.0
+1743.8421630859375,1797.8802490234375,1741.182861328125,1775.6767578125,6990313307,0.0,0.0
+1775.4813232421875,1781.3792724609375,1690.5166015625,1715.4608154296875,8585750760,0.0,0.0
+1715.327392578125,1791.1109619140625,1702.564697265625,1772.78564453125,8346890042,0.0,0.0
+1772.89990234375,1825.6009521484375,1772.89990234375,1793.0018310546875,9136176952,0.0,0.0
+1793.345947265625,1827.2802734375,1766.2489013671875,1792.7371826171875,9724496987,0.0,0.0
+1793.6201171875,1844.6773681640625,1783.1422119140625,1822.0220947265625,9149379763,0.0,0.0
+1821.704345703125,1840.173095703125,1811.1185302734375,1813.860595703125,5556443606,0.0,0.0
+1821.192138671875,1824.328857421875,1774.8724365234375,1795.7081298828125,6199721188,0.0,0.0
+1793.747802734375,1835.900390625,1765.5743408203125,1778.10546875,9191772160,0.0,0.0
diff --git a/ETH_model/fingerprint.pb b/ETH_model/fingerprint.pb
@@ -1 +1 @@
-ֈߐՐ��b����냰�����⏅�� ��������(�������2
+�����Ĳ����܍�����ވ�đ��u ��ߺ����(�΃��͖�2
diff --git a/ETH_model/keras_metadata.pb b/ETH_model/keras_metadata.pb
diff --git a/ETH_model/saved_model.pb b/ETH_model/saved_model.pb
diff --git a/ETH_model/variables/variables.data-00000-of-00001 b/ETH_model/variables/variables.data-00000-of-00001
diff --git a/ETH_model/variables/variables.index b/ETH_model/variables/variables.index
diff --git a/LINK.csv b/LINK.csv
@@ -1961,4 +1961,13 @@ Open,High,Low,Close,Volume,Dividends,Stock Splits
 7.406498908996582,7.697200775146484,7.0700459480285645,7.180304050445557,586895392,0.0,0.0
 7.17984676361084,7.629281044006348,7.119134902954102,7.621306896209717,329204523,0.0,0.0
 7.62053108215332,7.646345138549805,7.152146816253662,7.206892013549805,295497990,0.0,0.0
-7.2022294998168945,7.2510294914245605,6.994295120239258,7.052063941955566,191346096,0.0,0.0
+7.20620584487915,7.251605987548828,6.994295120239258,7.084826946258545,193197164,0.0,0.0
+7.085594177246094,7.284111976623535,7.06675910949707,7.262111186981201,165218457,0.0,0.0
+7.262615203857422,7.271117210388184,6.804008960723877,6.872753143310547,230593321,0.0,0.0
+6.8729681968688965,7.02065896987915,6.77236795425415,6.955334186553955,211056975,0.0,0.0
+6.954264163970947,7.458693027496338,6.94170618057251,7.382523059844971,297994491,0.0,0.0
+7.382503986358643,7.528229236602783,7.1701250076293945,7.232709884643555,279488316,0.0,0.0
+7.232532024383545,7.655243873596191,7.231266021728516,7.594569206237793,340384822,0.0,0.0
+7.595728874206543,7.6271820068359375,7.459664821624756,7.47448205947876,183579045,0.0,0.0
+7.55074405670166,7.55074405670166,7.224246025085449,7.259294033050537,188792643,0.0,0.0
+7.258115768432617,7.424676895141602,7.073147296905518,7.084716796875,284152512,0.0,0.0
diff --git a/LINK_model/fingerprint.pb b/LINK_model/fingerprint.pb
@@ -1 +1 @@
-ֈߐՐ��b����냰�����⏅�� ��������(������2
+�����Ĳ����܍�����ވ�đ��u ��ߺ����(�ݯ�����/2
diff --git a/LINK_model/keras_metadata.pb b/LINK_model/keras_metadata.pb
diff --git a/LINK_model/saved_model.pb b/LINK_model/saved_model.pb
diff --git a/LINK_model/variables/variables.data-00000-of-00001 b/LINK_model/variables/variables.data-00000-of-00001
diff --git a/LINK_model/variables/variables.index b/LINK_model/variables/variables.index
diff --git a/LTC.csv b/LTC.csv
@@ -3110,4 +3110,13 @@ Open,High,Low,Close,Volume,Dividends,Stock Splits
 82.71697998046875,88.15037536621094,79.94864654541016,87.35688781738281,969686696,0.0,0.0
 87.34841918945312,94.44811248779297,86.03134155273438,93.70100402832031,1054746379,0.0,0.0
 93.69886016845703,95.75912475585938,89.8736801147461,93.70990753173828,866952422,0.0,0.0
-93.96833038330078,93.96833038330078,90.43731689453125,91.44096374511719,515442624,0.0,0.0
+93.7067642211914,93.96833038330078,90.42384338378906,91.58114624023438,493921885,0.0,0.0
+91.5821762084961,94.64312744140625,90.98301696777344,93.52519226074219,489851737,0.0,0.0
+93.51691436767578,93.6174545288086,88.17765808105469,89.24687194824219,552145190,0.0,0.0
+89.24317932128906,89.42103576660156,86.03421020507812,89.13937377929688,556233998,0.0,0.0
+89.13932800292969,92.87533569335938,89.09891510009766,91.34173583984375,541611636,0.0,0.0
+91.34050750732422,92.9992904663086,87.33723449707031,89.05989074707031,593012954,0.0,0.0
+89.07251739501953,90.20854187011719,87.93004608154297,89.68412780761719,452243261,0.0,0.0
+89.68643951416016,93.85362243652344,89.5528335571289,92.87535095214844,548762293,0.0,0.0
+92.662841796875,94.4293441772461,90.8553237915039,93.01597595214844,549452266,0.0,0.0
+92.9996566772461,95.37003326416016,90.96139526367188,91.40279388427734,637808896,0.0,0.0
diff --git a/LTC_model/fingerprint.pb b/LTC_model/fingerprint.pb
@@ -1 +1 @@
-�۴���Ś����냰�����⏅�� ��������(��������2
+ɺ���յ����܍�����ވ�đ��u ��ߺ����(�ǅِ���2
diff --git a/LTC_model/keras_metadata.pb b/LTC_model/keras_metadata.pb
diff --git a/LTC_model/saved_model.pb b/LTC_model/saved_model.pb
diff --git a/LTC_model/variables/variables.data-00000-of-00001 b/LTC_model/variables/variables.data-00000-of-00001
diff --git a/LTC_model/variables/variables.index b/LTC_model/variables/variables.index
diff --git a/classifier_ADA.joblib b/classifier_ADA.joblib
diff --git a/classifier_BTC.joblib b/classifier_BTC.joblib
diff --git a/classifier_DOGE.joblib b/classifier_DOGE.joblib
diff --git a/classifier_ETH.joblib b/classifier_ETH.joblib
diff --git a/classifier_LINK.joblib b/classifier_LINK.joblib
diff --git a/classifier_LTC.joblib b/classifier_LTC.joblib
diff --git a/classify_pos_neg.py b/classify_pos_neg.py
@@ -19,6 +19,7 @@
 from sklearn.metrics import accuracy_score
 import joblib
 from sys import argv
+import ta 
 
 def get_ohlc(crypt):
     crypt_name = crypt + '-USD'
@@ -38,50 +39,53 @@ def OBV(data):
         If the closing price of the asset is lower than the previous day?s closing price:
         OBV = Previous OBV - Current Day's Volume
         """
-        #TODO: add a linear regression lineover like 10-20 data points and use that 
-        # as a buy signal if it is positive
-        data['OBV'] = np.zeros(len(data))
-        # crypto_df_final['OBV'].iloc[0] = crypto_df_final['volume'].iloc[0]
-        OBV_iter = data['Volume'].iloc[0]
-        data['OBV'].iloc[0] = OBV_iter
-        for i in range(1,len(data['OBV'])):
-            if (data['Close'].iloc[i-1] < data['Close'].iloc[i]):
-                OBV_iter += data['Volume'].iloc[i]
-            if (data['Close'].iloc[i-1] > data['Close'].iloc[i]):
-                OBV_iter -= data['Volume'].iloc[i]
-            if (data['Close'].iloc[i-1] == data['Close'].iloc[i]):
-                OBV_iter += 0
-            data['OBV'].iloc[i] = OBV_iter
+        # #TODO: add a linear regression lineover like 10-20 data points and use that 
+        # # as a buy signal if it is positive
+        # data['OBV'] = np.zeros(len(data))
+        # # crypto_df_final['OBV'].iloc[0] = crypto_df_final['volume'].iloc[0]
+        # OBV_iter = data['Volume'].iloc[0]
+        # data['OBV'].iloc[0] = OBV_iter
+        # for i in range(1,len(data['OBV'])):
+        #     if (data['Close'].iloc[i-1] < data['Close'].iloc[i]):
+        #         OBV_iter += data['Volume'].iloc[i]
+        #     if (data['Close'].iloc[i-1] > data['Close'].iloc[i]):
+        #         OBV_iter -= data['Volume'].iloc[i]
+        #     if (data['Close'].iloc[i-1] == data['Close'].iloc[i]):
+        #         OBV_iter += 0
+        #     data['OBV'].iloc[i] = OBV_iter
+        data['OBV'] = ta.volume.OnBalanceVolumeIndicator(data['Close'], data['Volume']).on_balance_volume()
         return data
 
 def RSI(data):
-    update = pd.DataFrame()
-    update['change'] = data.Close.diff()
-    # crypto_df['U'] = [x if x > 0 else 0 for x in crypto_df.change]
-    # crypto_df['D'] = [abs(x) if x < 0 else 0 for x in crypto_df.change]
-    update['U']  = update['change'].clip(lower=0)
-    update['D'] = -1*update['change'].clip(upper=0)
-    update['U'] = update['U'].ewm(span=14,
-                min_periods=1).mean()
-    update['D'] = update['D'].ewm(span=14,
-                min_periods=1).mean()
-    update['RS'] = update['U'] / update['D']
-    data['RSI'] = 100 - (100/(1+update['RS']))
+    # update = pd.DataFrame()
+    # update['change'] = data.Close.diff()
+    # # crypto_df['U'] = [x if x > 0 else 0 for x in crypto_df.change]
+    # # crypto_df['D'] = [abs(x) if x < 0 else 0 for x in crypto_df.change]
+    # update['U']  = update['change'].clip(lower=0)
+    # update['D'] = -1*update['change'].clip(upper=0)
+    # update['U'] = update['U'].ewm(span=14,
+    #             min_periods=1).mean()
+    # update['D'] = update['D'].ewm(span=14,
+    #             min_periods=1).mean()
+    # update['RS'] = update['U'] / update['D']
+    # data['RSI'] = 100 - (100/(1+update['RS']))
+    data['RSI'] = ta.momentum.RSIIndicator(data['Close'], window=14).rsi()
     return data
 
 def vol_RSI(data):
-    update = pd.DataFrame()
-    update['change_vol'] = data.Volume.diff()
-    # crypto_df['U'] = [x if x > 0 else 0 for x in crypto_df.change]
-    # crypto_df['D'] = [abs(x) if x < 0 else 0 for x in crypto_df.change]
-    update['U_vol']  = update['change_vol'].clip(lower=0)
-    update['D_vol'] = -1*update['change_vol'].clip(upper=0)
-    update['U_vol'] = update.U_vol.ewm(span=14,
-                min_periods=1).mean()
-    update['D'] = update.D_vol.ewm(span=14,
-                min_periods=1).mean()
-    update['RS_vol'] = update.U_vol / update.D_vol
-    data['RSI_vol'] = 100 - (100/(1+update.RS_vol))
+    # update = pd.DataFrame()
+    # update['change_vol'] = data.Volume.diff()
+    # # crypto_df['U'] = [x if x > 0 else 0 for x in crypto_df.change]
+    # # crypto_df['D'] = [abs(x) if x < 0 else 0 for x in crypto_df.change]
+    # update['U_vol']  = update['change_vol'].clip(lower=0)
+    # update['D_vol'] = -1*update['change_vol'].clip(upper=0)
+    # update['U_vol'] = update.U_vol.ewm(span=14,
+    #             min_periods=1).mean()
+    # update['D'] = update.D_vol.ewm(span=14,
+    #             min_periods=1).mean()
+    # update['RS_vol'] = update.U_vol / update.D_vol
+    # data['RSI_vol'] = 100 - (100/(1+update.RS_vol))
+    data['RSI_vol'] = ta.momentum.RSIIndicator(data['Volume'], window=14).rsi()
     return data
 
 def moving_averages(data):
@@ -139,19 +143,86 @@ def aroon_ind(data,lb=25):
         if up[i] >= 70 and down[i] <= 30: buy
         if up[i] <= 30 and down[i] >= 70: sell
         """
-        data['aroon_up'] = 100 * ((data['High'].rolling(lb).apply(lambda x: x.argmax())) / lb)
-        data['aroon_down'] = 100 * ((data['Low'].rolling(lb).apply(lambda x: x.argmin())) / lb)
+        # data['aroon_up'] = 100 * ((data['High'].rolling(lb).apply(lambda x: x.argmax())) / lb)
+        # data['aroon_down'] = 100 * ((data['Low'].rolling(lb).apply(lambda x: x.argmin())) / lb)
+        aroon = ta.trend.AroonIndicator(data['Close'],window=25)
+        data['aroon'] = aroon.aroon_indicator()
         return data
 
 def stoch_RSI(data):
-        min_val  = data['RSI'].rolling(window=14, center=False).min()
-        max_val = data['RSI'].rolling(window=14, center=False).max()
-        data['Stoch_RSI'] = ((data['RSI'] - min_val) / (max_val - min_val)) * 100
+        # min_val  = data['RSI'].rolling(window=14, center=False).min()
+        # max_val = data['RSI'].rolling(window=14, center=False).max()
+        # data['Stoch_RSI'] = ((data['RSI'] - min_val) / (max_val - min_val)) * 100
+        data['Stoch_RSI'] = ta.momentum.StochRSIIndicator(data['RSI']).stochrsi()
         return data
 
+def volume_osc(data):
+    """
+    Volume Oscillator = [(Shorter Period SMA of Volume – Longer Period SMA of Volume)
+                            / Longer Period SMA of Volume ] * 100
+    """
+    short = data['Volume'].ewm(span=14,min_periods=13).mean() 
+    long = data['Volume'].ewm(span=28,min_periods=27).mean() 
+    data['volume_os'] = ((short - long) / long) * 100
+
+    return data
+
+def cmf_line(df):
+    """
+    If the CMF is above zero, it suggests that buying pressure is increasing and the price is likely to rise. 
+    If the CMF is below zero, it suggests that selling pressure is increasing and the price is likely to fall.
+    """
+    # Calculate the money flow multiplier
+    mf_mult = ((df['Close'] - df['Low']) - (df['High'] - df['Close'])) / (df['High'] - df['Low'])
+    mf_mult = mf_mult.fillna(0) # Replace NaN values with 0
+
+    # Calculate the money flow volume
+    mf_vol = mf_mult * df['Volume']
+
+    # Calculate the cumulative money flow volume and volume
+    cumulative_mfv = mf_vol.cumsum()
+    cumulative_vol = df['Volume'].cumsum()
+
+    # Calculate the CMF
+    cmf = cumulative_mfv / cumulative_vol
+    cmf *= (2 * (df['Close'] - df['Low']) - (df['High'] - df['Close'])) / (df['High'] - df['Low'])
+    cmf[:2] = [np.nan, np.nan]
+
+    df['cmf'] = cmf
+
+    return df
+
+def vwma_macd(df):
+    # Calculate the volume-weighted moving average (VWMA) of the volume
+    df['VWMA'] = (df['Volume'] * (df['High'] + df['Low']) / 2).cumsum() / df['Volume'].cumsum()
+
+    # Calculate the 12-day and 26-day exponential moving averages (EMA) of the VWMA
+    ema12 = df['VWMA'].ewm(span=12, adjust=False).mean()
+    ema26 = df['VWMA'].ewm(span=26, adjust=False).mean()
+
+    # Calculate the MACD line as the difference between the 12-day and 26-day EMAs
+    macd_line = ema12 - ema26
+
+    # Calculate the 9-day EMA of the MACD line as the signal line
+    signal_line = macd_line.ewm(span=9, adjust=False).mean()
+
+    # Calculate the MACD histogram as the difference between the MACD line and the signal line
+    macd_hist = macd_line - signal_line
+
+    df['vol_macd'] = macd_line
+    df['vol_macd_signal'] = signal_line
+
+    return df
+
+def force_index_indicator(df):
+    force = ta.volume.ForceIndexIndicator(df['close'],df['volume'])
+    df['force_index'] = force.force_index()
+    return df
+
 def split_x_y(df):
     X = df[['OBV', 'macd_diff', 'signal_line','MFI','ewmshort','ewmmedium','ewmlong',
-            'RSI_vol','RSI','aroon_up','Stoch_RSI']].fillna(method='bfill')#.values[:-1]
+            'RSI_vol','RSI','aroon','Stoch_RSI','volume_os','vol_macd','vol_macd_signal',
+            'cmf']].fillna(method='bfill')#.values[:-1]
     #Make sure this is right
     Y = np.where(df['Close'].shift(-1).values > df['Close'].values, 1, 0)
     Y = np.pad(Y, (1, 0), mode='constant')
@@ -232,7 +303,8 @@ def deep_model(X_train,y_train,X_test,y_test):
     # print(classification_report(y_test, y_pred_class))
 def random_forest(df):
     X = df[['OBV', 'macd_diff', 'signal_line','MFI','ewmshort','ewmmedium','ewmlong',
-            'RSI_vol','RSI','aroon_up','Stoch_RSI']].fillna(method='bfill')#.values[:-1]
+            'RSI_vol','RSI','aroon','Stoch_RSI','volume_os','vol_macd','vol_macd_signal',
+            'cmf']].fillna(method='bfill')#.values[:-1]
     #Make sure this is right
     Y = np.where(df['Close'].shift(-1).values > df['Close'].values, 1, 0)
     Y = np.pad(Y, (1, 0), mode='constant')
@@ -260,7 +332,7 @@ def random_forest(df):
     print("Accuracy:", acc)
 
 def main():
-    #Possible Cryptos: BTC, ETH, DOGE, LTC, SOL
+    #Possible Cryptos: BTC, ETH, DOGE, LTC, ADA, LINK
     print(f'creating model for {argv[1]}')
     crypt = argv[1]
     df = get_ohlc(crypt)
@@ -272,6 +344,9 @@ def main():
     df = macd(df)
     df = aroon_ind(df)
     df = stoch_RSI(df)
+    df = volume_osc(df)
+    df = cmf_line(df)
+    df = vwma_macd(df)
     X_train, X_test, y_train, y_test = split_x_y(df)
     deep_model(X_train,y_train,X_test,y_test)
     random_forest(df)
Original file line number	Diff line number	Diff line change
		@@ -1 +1 @@
		ֈߐՐ��b��냰��⏅�� (䊘��ð��2
		��Ĳ��܍��ވ�đ��u ��ߺ��(�Į��2
Original file line number	Diff line number	Diff line change
		@@ -1 +1 @@
		�۴��Ś��냰��⏅�� (��2
		ɺ��յ��܍��ވ�đ��u ��ߺ��(��ݑ��s2
Original file line number	Diff line number	Diff line change
		@@ -1 +1 @@
		ֈߐՐ��b��냰��⏅�� (��爏 2
		��Ĳ��܍��ވ�đ��u ��ߺ��(��±�2
Original file line number	Diff line number	Diff line change
		@@ -1 +1 @@
		ֈߐՐ��b��냰��⏅�� (��2
		��Ĳ��܍��ވ�đ��u ��ߺ��(�΃��͖�2
Original file line number	Diff line number	Diff line change
		@@ -1 +1 @@
		ֈߐՐ��b��냰��⏅�� (��2
		��Ĳ��܍��ވ�đ��u ��ߺ��(�ݯ��/2
Original file line number	Diff line number	Diff line change
		@@ -1 +1 @@
		�۴��Ś��냰��⏅�� (��2
		ɺ��յ��܍��ވ�đ��u ��ߺ��(�ǅِ��2