diff --git a/video_prediction_savp/video_prediction/models/vanilla_convLSTM_model.py b/video_prediction_savp/video_prediction/models/vanilla_convLSTM_model.py
index 5a8a2e1f3fffe5c66d5b93e53137300bf792317e..744284fc6c5b52bcde249f1a58e04a41e80339fa 100644
--- a/video_prediction_savp/video_prediction/models/vanilla_convLSTM_model.py
+++ b/video_prediction_savp/video_prediction/models/vanilla_convLSTM_model.py
@@ -70,9 +70,14 @@ class VanillaConvLstmVideoPredictionModel(BaseVideoPredictionModel):
#self.context_frames_loss = tf.reduce_mean(
# tf.square(self.x[:, :self.context_frames, :, :, 0] - self.x_hat_context_frames[:, :, :, :, 0]))
# This is the loss function (RMSE):
+ #This is loss function only for 1 channel (temperature RMSE)
self.total_loss = tf.reduce_mean(
- tf.square(self.x[:, self.context_frames:, :, :, 0] - self.x_hat_context_frames[:, (self.context_frames-1):-1, :, :, 0]))
-
+ tf.square(self.x[:, self.context_frames:,:,:,0] - self.x_hat_predict_frames[:,:,:,:,0]))
+
+ #This is the loss for only all the channels(temperature, geo500, pressure)
+ #self.total_loss = tf.reduce_mean(
+ # tf.square(self.x[:, self.context_frames:,:,:,:] - self.x_hat_predict_frames[:,:,:,:,:]))
+
self.train_op = tf.train.AdamOptimizer(
learning_rate = self.learning_rate).minimize(self.total_loss, global_step = self.global_step)
self.outputs = {}
@@ -88,35 +93,19 @@ class VanillaConvLstmVideoPredictionModel(BaseVideoPredictionModel):
@staticmethod
def convLSTM_cell(inputs, hidden):
- conv1 = ld.conv_layer(inputs, 3, 2, 8, "encode_1", activate = "leaky_relu")
-
- conv2 = ld.conv_layer(conv1, 3, 1, 8, "encode_2", activate = "leaky_relu")
-
- conv3 = ld.conv_layer(conv2, 3, 2, 8, "encode_3", activate = "leaky_relu")
-
- y_0 = conv3
+ y_0 = inputs #we only usd patch 1, but the original paper use patch 4 for the moving mnist case, but use 2 for Radar Echo Dataset
# conv lstm cell
cell_shape = y_0.get_shape().as_list()
+ channels = cell_shape[-1]
with tf.variable_scope('conv_lstm', initializer = tf.random_uniform_initializer(-.01, 0.1)):
- cell = BasicConvLSTMCell(shape = [cell_shape[1], cell_shape[2]], filter_size = [3, 3], num_features = 8)
+ cell = BasicConvLSTMCell(shape = [cell_shape[1], cell_shape[2]], filter_size = [5, 5], num_features = 256)
if hidden is None:
hidden = cell.zero_state(y_0, tf.float32)
-
output, hidden = cell(y_0, hidden)
-
-
output_shape = output.get_shape().as_list()
-
-
z3 = tf.reshape(output, [-1, output_shape[1], output_shape[2], output_shape[3]])
-
- conv5 = ld.transpose_conv_layer(z3, 3, 2, 8, "decode_5", activate = "leaky_relu")
-
-
- conv6 = ld.transpose_conv_layer(conv5, 3, 1, 8, "decode_6", activate = "leaky_relu")
-
-
- x_hat = ld.transpose_conv_layer(conv6, 3, 2, 3, "decode_7", activate = "sigmoid") # set activation to linear
+ #we feed the learn representation into a 1 × 1 convolutional layer to generate the final prediction
+ x_hat = ld.conv_layer(z3, 1, 1, channels, "decode_1", activate="sigmoid")
return x_hat, hidden
@@ -124,20 +113,18 @@ class VanillaConvLstmVideoPredictionModel(BaseVideoPredictionModel):
network_template = tf.make_template('network',
VanillaConvLstmVideoPredictionModel.convLSTM_cell) # make the template to share the variables
# create network
- x_hat_context = []
x_hat = []
- hidden = None
- #This is for training
- for i in range(self.sequence_length):
- if i < self.context_frames:
- x_1, hidden = network_template(self.x[:, i, :, :, :], hidden)
- else:
- x_1, hidden = network_template(x_1, hidden)
- x_hat_context.append(x_1)
+
+ # for i in range(self.sequence_length-1):
+ # if i < self.context_frames:
+ # x_1, hidden = network_template(self.x[:, i, :, :, :], hidden)
+ # else:
+ # x_1, hidden = network_template(x_1, hidden)
+ # x_hat_context.append(x_1)
- #This is for generating video
+ #This is for training (optimization of convLSTM layer)
hidden_g = None
- for i in range(self.sequence_length):
+ for i in range(self.sequence_length-1):
if i < self.context_frames:
x_1_g, hidden_g = network_template(self.x[:, i, :, :, :], hidden_g)
else:
@@ -145,8 +132,6 @@ class VanillaConvLstmVideoPredictionModel(BaseVideoPredictionModel):
x_hat.append(x_1_g)
# pack them all together
- x_hat_context = tf.stack(x_hat_context)
x_hat = tf.stack(x_hat)
- self.x_hat_context_frames = tf.transpose(x_hat_context, [1, 0, 2, 3, 4]) # change first dim with sec dim
self.x_hat= tf.transpose(x_hat, [1, 0, 2, 3, 4]) # change first dim with sec dim
- self.x_hat_predict_frames = self.x_hat[:,self.context_frames:,:,:,:]
+ self.x_hat_predict_frames = self.x_hat[:,self.context_frames-1:,:,:,:]