Refactor (preparation for multiple models)

quickdudley

May 10, 2022, 3:59 AM

JZRK6Q4KJM2B5NHE257QXMMP4RY6GMNTSSPDPCW63P32SL4NBJWQC

Dependencies

• [2] 3QGP6RXL Automatically format python files
• [3] ROQCAPZJ Begin function for showing the map (for now just opens SDL window)
• [4] SJHJS463 Model trainer: main function to open sdl window and launch second thread for actual training
• [5] 2ABZP2KN Model trainer: save map.png after training
• [6] 7ML3OFE7 Model trainer: initial train and visualize thread pair (total crust mass; todo: altitude instead)
• [7] CWOSQTC4 Trust tensorflow's thread safety (it wasn't the cause of an earlier bug)
• [*] ZM2EMAZO Start doing python multi-module stuff properly

Change contents

• edit in trainmodel/src/model.py at line 53
  [2.4265]→[2.4265:4266](∅→∅)
  −
• edit in trainmodel/src/model.py at line 58
  [2.4411]
  +

  + class ModelSet:

  + def __init__(self):

  + self.heightmap = model()

  + self.heightmap.compile(optimizer=keras.optimizers.RMSprop(),

  + loss=shore_focused_loss)
• replacement in trainmodel/src/drawmap.py at line 28
  [2.5631]→[2.5631:5657](∅→∅)
  − outputs = m(inputs)
  [2.5631]

  [2.5657]
  + outputs = m.heightmap(inputs)
• replacement in trainmodel/src/__main__.py at line 14
  [2.8332]→[2.8332:8461](∅→∅)
  − m.fit(x=training_data[0], y=training_data[1], batch_size=100, epochs=5000)

  − tfjs.converters.save_keras_model(m, 'tfjs_model')
  [2.8332]

  [3.128]
  + m.heightmap.fit(x=training_data[0], y=training_data[1], batch_size=100, epochs=5000)

  + tfjs.converters.save_keras_model(m.heightmap, 'tfjs_model')
• replacement in trainmodel/src/__main__.py at line 19
  [2.8568]→[2.8568:8619](∅→∅)
  − outputs = tf.reshape(m(inputs), (1024, 2048, 1))
  [2.8568]

  [2.8619]
  + outputs = tf.reshape(m.heightmap(inputs), (1024, 2048, 1))
• replacement in trainmodel/src/__main__.py at line 28
  [3.151]→[2.8800:8913](∅→∅)
  − m = model.model()

  − m.compile(optimizer=keras.optimizers.RMSprop(),

  − loss=model.shore_focused_loss)
  [3.151]

  [2.8913]
  + m = model.ModelSet()