keras regression ready for testing and deployment

data_preparation/cleaned/time_series_notnormalized_with_0Nan_rpe.csv

@@ -36,8 +36,8 @@
 34,34,34,2018-07-17,4,1,Skills,60.0,5.0,300.0,300.0,107.1,309.11,0.35,7.0,8.0,,0,0,1,0,0,0,0,0,0,0,0,1,350
 35,35,35,2018-07-17,5,1,Skills,60.0,5.0,300.0,300.0,42.9,100.71,0.43,9.0,9.0,,0,0,1,0,0,0,0,0,0,0,0,1,350
 36,36,36,2018-07-17,6,1,Speed,30.0,3.0,90.0,370.0,52.9,333.5,0.16,0.0,0.0,,0,0,0,0,0,0,1,0,0,0,0,1,350
-37,37,37,2018-07-17,6,1,Conditioning,35.0,8.0,280.0,0.0,0.0,0.0,0.0,0.0,0.0,,0,0,0,1,0,0,0,0,0,0,0,1,350
 38,38,38,2018-07-17,7,1,Skills,75.0,6.0,450.0,450.0,205.7,401.79,0.51,0.0,0.0,,0,0,1,0,0,0,0,0,0,0,0,1,350
+37,37,37,2018-07-17,6,1,Conditioning,35.0,8.0,280.0,0.0,0.0,0.0,0.0,0.0,0.0,,0,0,0,1,0,0,0,0,0,0,0,1,350
 39,39,39,2018-07-17,10,1,Skills,60.0,4.0,240.0,240.0,79.3,298.57,0.27,0.0,0.0,,0,0,1,0,0,0,0,0,0,0,0,1,350
 40,40,40,2018-07-17,11,1,Skills,90.0,5.0,450.0,450.0,210.0,391.36,0.54,9.0,9.0,Absolutely,0,0,1,0,0,0,0,0,0,1,0,0,350
 41,41,41,2018-07-17,13,1,Skills,90.0,6.0,540.0,540.0,268.6,309.25,0.87,0.0,0.0,Not at all,0,0,1,0,0,0,0,0,1,0,0,0,350

hypotheses_modeling/KerasRegressions.py

@@ -0,0 +1,96 @@
+import tensorflow as tf
+import pandas as pd
+import numpy as np
+
+
+def time_series_sigmoid_classification(X, Y, k, n0, x_columns, y_columns):
+  inp = X[x_columns]
+  out = Y[y_columns]
+  col = "day"
+  x = []
+  y = []
+  input_shape = 0
+  output_shape = 0
+  for player in Y["playerID"].unique():
+    XPlayer = inp[inp["playerID"] == player]
+    YPlayer = out[out["playerID"] == player]
+    for day in YPlayer[col][n0 - 1:]:
+      prev = day - k
+      xprev = XPlayer[XPlayer[col] == prev].drop(columns=[col]).to_numpy()
+      if xprev.shape[0] != 1:
+        continue
+      else:
+        xprev = xprev[0, :]
+      yt = YPlayer[YPlayer[col] == day].drop(columns=[col]).to_numpy()[0, :]
+      if input_shape == 0:
+        input_shape = xprev.shape[0]
+      else:
+        if input_shape != xprev.shape[0]:
+          print("INCONSISTENT INPUT DIMENSION")
+          exit(2)
+      if input_shape == 0:
+        output_shape = yt.shape[0]
+      else:
+        if output_shape != yt.shape[0]:
+          print("INCONSISTENT OUTPUT DIMENSION")
+          exit(2)
+      x.append(xprev)
+      y.append(yt)
+  x = np.array(x)
+  y = np.array(y)
+  model = tf.keras.Sequential([
+    tf.keras.layers.Flatten(input_shape=input_shape),
+    tf.keras.layers.Dense(output_shape, activation=tf.nn.softmax)
+  ])
+  model.compile(optimizer='adam', loss='sparse_categorical_crossentropy', metrics=['accuracy', 'categorical_accuracy'])
+  model.fit(x, y, epochs=100)
+  loss, accuracy = model.evaluate(x, y)
+  print(loss, accuracy)
+  return model.get_weights()
+
+
+def time_series_sigmoid_classification(X, Y, k, n0, x_columns, y_columns):
+  inp = X[x_columns]
+  out = Y[y_columns]
+  col = "day"
+  x = []
+  y = []
+  input_shape = 0
+  output_shape = 0
+  for player in Y["playerID"].unique():
+    XPlayer = inp[inp["playerID"] == player]
+    YPlayer = out[out["playerID"] == player]
+    for day in YPlayer[col][n0 - 1:]:
+      prev = day - k
+      xprev = XPlayer[XPlayer[col] == prev].drop(columns=[col]).to_numpy()
+      if xprev.shape[0] != 1:
+        continue
+      else:
+        xprev = xprev[0, :]
+      yt = YPlayer[YPlayer[col] == day].drop(columns=[col]).to_numpy()[0, :]
+      if input_shape == 0:
+        input_shape = xprev.shape[0]
+      else:
+        if input_shape != xprev.shape[0]:
+          print("INCONSISTENT INPUT DIMENSION")
+          exit(2)
+      if input_shape == 0:
+        output_shape = yt.shape[0]
+      else:
+        if output_shape != yt.shape[0]:
+          print("INCONSISTENT OUTPUT DIMENSION")
+          exit(2)
+      x.append(xprev)
+      y.append(yt)
+  x = np.array(x)
+  y = np.array(y)
+  model = tf.keras.Sequential([
+    tf.keras.layers.Flatten(input_shape=input_shape),
+    tf.keras.layers.Dense(32),
+    tf.keras.layers.Dense(output_shape, activation=tf.nn.softmax)
+  ])
+  model.compile(optimizer='adam', loss='sparse_categorical_crossentropy', metrics=['accuracy', 'categorical_accuracy'])
+  model.fit(x, y, epochs=100)
+  loss, accuracy = model.evaluate(x, y)
+  print(loss, accuracy)
+  return model.get_weights()