inference training

Author: JordiCorbilla

GOOG_20200704_b5f47746c83698528343678663ac3c96/inference.csv

@@ -0,0 +1,306 @@
+,0
+0,56.317787
+1,61.1736
+2,59.11235
+3,51.75302
+4,44.940697
+5,40.462807
+6,41.294895
+7,43.31001
+8,44.81763
+9,47.855385
+10,52.92124
+11,55.755062
+12,57.223644
+13,56.635033
+14,57.511547
+15,59.58694
+16,63.648922
+17,64.33477
+18,61.26138
+19,54.586243
+20,45.356445
+21,38.392834
+22,38.82387
+23,44.561275
+24,53.99945
+25,64.60877
+26,68.661064
+27,69.210075
+28,67.4986
+29,64.83073
+30,62.80525
+31,64.013275
+32,65.53788
+33,64.15403
+34,62.78204
+35,59.893867
+36,55.375698
+37,52.26238
+38,49.460854
+39,47.497906
+40,49.75747
+41,57.295845
+42,65.48818
+43,71.12546
+44,74.50774
+45,75.54332
+46,73.96777
+47,68.08885
+48,60.82775
+49,55.07597
+50,50.1526
+51,46.071827
+52,46.693165
+53,50.04703
+54,53.85379
+55,59.57086
+56,61.534496
+57,58.8781
+58,52.60207
+59,47.339054
+60,44.448997
+61,44.268307
+62,46.91525
+63,49.86658
+64,52.91868
+65,52.80223
+66,49.065258
+67,45.557472
+68,45.603893
+69,45.840015
+70,45.054554
+71,43.806393
+72,43.132607
+73,41.45051
+74,38.122086
+75,35.239216
+76,35.33435
+77,38.355473
+78,46.436802
+79,58.515705
+80,70.7686
+81,78.05592
+82,76.954414
+83,70.74484
+84,64.735886
+85,63.01453
+86,64.14086
+87,63.317654
+88,59.54807
+89,55.843624
+90,55.81856
+91,61.91248
+92,68.16476
+93,70.137184
+94,68.0748
+95,62.092987
+96,53.696465
+97,44.343925
+98,39.22013
+99,39.434055
+100,43.877895
+101,49.032516
+102,51.076084
+103,51.850807
+104,53.12258
+105,55.388523
+106,59.828415
+107,64.840935
+108,66.76945
+109,63.774567
+110,57.750874
+111,52.568798
+112,50.183372
+113,53.494473
+114,56.89796
+115,58.940105
+116,60.908463
+117,61.19233
+118,57.541798
+119,51.310024
+120,43.53133
+121,36.001343
+122,30.339222
+123,30.615528
+124,37.21069
+125,45.64332
+126,53.047337
+127,56.65195
+128,54.84111
+129,51.537575
+130,46.755543
+131,40.368595
+132,38.07337
+133,41.267372
+134,44.704098
+135,47.45567
+136,49.058475
+137,47.58637
+138,43.73454
+139,39.442696
+140,38.29924
+141,40.792885
+142,46.879128
+143,52.35109
+144,57.020733
+145,61.6833
+146,61.660603
+147,56.735455
+148,49.486317
+149,42.166763
+150,36.485874
+151,37.0594
+152,43.95215
+153,55.357227
+154,65.641914
+155,72.614265
+156,74.36655
+157,71.297844
+158,66.821594
+159,63.27379
+160,59.72942
+161,56.386448
+162,56.91229
+163,62.779285
+164,69.28576
+165,69.75186
+166,67.148056
+167,65.34721
+168,64.33941
+169,63.53279
+170,62.19789
+171,63.085403
+172,66.47886
+173,66.45321
+174,63.676586
+175,59.308422
+176,53.356888
+177,49.497444
+178,52.106655
+179,60.3396
+180,69.805885
+181,74.687325
+182,75.10766
+183,69.59556
+184,60.116352
+185,51.873676
+186,50.33226
+187,53.81782
+188,60.847607
+189,65.59432
+190,67.287094
+191,64.59502
+192,58.593567
+193,53.955673
+194,53.37181
+195,53.184578
+196,52.266994
+197,50.162815
+198,46.178715
+199,44.045643
+200,42.675144
+201,42.893635
+202,43.600872
+203,46.006886
+204,50.690353
+205,52.809723
+206,52.129833
+207,53.13782
+208,52.83069
+209,49.7728
+210,47.725334
+211,49.06386
+212,53.814777
+213,57.18265
+214,57.02118
+215,53.349995
+216,49.718597
+217,47.61544
+218,50.68281
+219,58.765865
+220,64.69569
+221,67.62983
+222,68.16974
+223,65.18015
+224,62.783768
+225,59.59185
+226,54.691048
+227,50.74944
+228,49.848854
+229,51.690643
+230,56.996826
+231,59.457962
+232,58.94028
+233,59.245613
+234,62.481335
+235,63.806664
+236,63.254936
+237,62.024567
+238,60.391136
+239,57.0773
+240,52.222057
+241,50.081013
+242,49.750244
+243,52.658707
+244,53.908863
+245,53.55174
+246,51.184677
+247,48.30691
+248,45.421215
+249,42.475918
+250,40.117344
+251,40.10504
+252,40.437687
+253,43.964
+254,53.066326
+255,62.52203
+256,66.6212
+257,67.47826
+258,63.850708
+259,59.55885
+260,55.92516
+261,56.11536
+262,55.969868
+263,53.46881
+264,48.46161
+265,42.624256
+266,36.411953
+267,31.52565
+268,30.58871
+269,34.71736
+270,40.704815
+271,47.795994
+272,52.016693
+273,55.41635
+274,55.413677
+275,55.19111
+276,53.69374
+277,49.16145
+278,46.50018
+279,49.031788
+280,52.096092
+281,55.295612
+282,60.036728
+283,62.336414
+284,61.83132
+285,57.88668
+286,51.97578
+287,49.119843
+288,52.664986
+289,58.810623
+290,64.130844
+291,65.87548
+292,66.13987
+293,63.86755
+294,60.205463
+295,55.652504
+296,52.544983
+297,53.78646
+298,57.270405
+299,62.693134
+300,67.834274
+301,68.108246
+302,64.94127
+303,60.485107
+304,57.789303

stock_prediction_deep_learning_inference.py

@@ -16,8 +16,7 @@
 from absl import app
 import tensorflow as tf
 import pandas as pd
-from sklearn.preprocessing import MinMaxScaler
-
+from stock_prediction_class import StockPrediction
 from stock_prediction_numpy import StockData
 from datetime import date
 os.environ["PATH"] += os.pathsep + 'C:/Program Files (x86)/Graphviz2.38/bin/'
@@ -26,10 +25,13 @@
 def main(argv):
     print(tf.version.VERSION)
     inference_folder = os.path.join(os.getcwd(), 'GOOG_20200704_b5f47746c83698528343678663ac3c96')
+    stock = StockPrediction(STOCK_TICKER, STOCK_START_DATE, STOCK_VALIDATION_DATE, inference_folder)
+
+    data = StockData(stock)
+
+    (x_train, y_train), (x_test, y_test), (min_max, test_data) = data.download_transform_to_numpy(TIME_STEPS)
 
     # load future data
-    data = StockData()
-    min_max = MinMaxScaler(feature_range=(0, 1))
     x_test, y_test = data.generate_future_data(TIME_STEPS, min_max, date(2020, 7, 5), date(2021, 7, 5))
 
     # load the weights from our best model
@@ -52,4 +54,7 @@ def main(argv):
 
 if __name__ == '__main__':
     TIME_STEPS = 60
+    STOCK_TICKER = 'GOOG'
+    STOCK_START_DATE = pd.to_datetime('2004-08-01')
+    STOCK_VALIDATION_DATE = pd.to_datetime('2017-01-01')
     app.run(main)