Suppose I have a training data set shown X which is a 6000*51 matrix and the problem is a multi-output classification problem and the target matrix Y is 6000*10 and each column of the target matrix takes 0 or 1. I can define parameters based on the features as follows:
n = 10
p = X[:,:n]
a = X[:,2:2*n]
c = X[:,2*n]
Suppose the prediction of my model is Prediction. I want to define a loss function as follows:
-np.einsum('ij,ij ->i',p,y_test).mean() + 10 *
np.mean( np.maximum( np.einsum('ij,ij ->i',a,Prediction) - c, 0) )
def knapsack_loss(X, n, cvc=1):
input_a = X[:,n:2*n]
input_a = np.float64(deepcopy(input_a ))
input_p = X[:,:n]
input_p = np.float64(deepcopy(input_p))
input_c = X[:,2*n]
input_c = np.float64(deepcopy(input_c))
def loss(y_true, y_pred):
picks = y_pred
return (-1 * K.batch_dot(picks, input_p, 1)) + cvc * K.maximum(
K.batch_dot(picks, input_a, 1) - input_c, 0)
return loss
def get_model(n_inputs, n_outputs):
model = Sequential()
model.add(Dense(100, input_dim=n_inputs, kernel_initializer='he_uniform', activation='relu'))
model.add(Dense(n_outputs, activation='sigmoid'))
model.compile(loss= knapsack_loss(X_train, n, cvc=1),optimizer='adam')
return model
n_inputs, n_outputs = X.shape[1], Y.shape[1]
model = get_model(n_inputs, n_outputs)
model.fit(X_train, y_train, verbose=0, epochs=500)
When I run this code, I face the following error:
InvalidArgumentError: Incompatible shapes: [32] vs. [6000]
[[{{node training_14/Adam/gradients/loss_22/dense_55_loss/loss/MatMul_1_grad/BroadcastGradientArgs}}]]
I would be thankful if someone can correct it or provide and synthetic example.
Related
I have example data. This data is used to set the predictor (x) and response (y) variables:
x has an Age_Years feature and y has Age_Years and target features:
I have a custom cost function that accepts multiple parameters:
# this gives the y_pred values
def calc_prob(param1, param2, param3, age):
prob = (((100*param1*pow((100/param3),-(pow((age/param2),param1))))*pow(age/param2,param1)*math.log(100/param3))/age)/100
return prob
# this serves as the custom cost function
def brier_score(y_pred, y_true):
prob = calc_prob(y_pred[:, 0], y_pred[:, 1], y_pred[:, 2], y_true['Age_Years'])
brier_score = tf.reduce_mean((prob - y_true['target']) ** 2, axis=1)
return brier_score
Ultimately, I want an output that builds a 3-parameter model that minimizes the brier_score() function. If I try to build and run the model, I get an error:
from keras import models
from keras import layers
def build_model():
model = models.Sequential()
model.add(layers.Dense(1, activation='relu', input_shape=(x.shape[1],)))
model.add(layers.Dense(5, activation='relu'))
model.add(layers.Dense(3, activation='softmax'))
model.compile(loss=losses, optimizer='adam', metrics=['accuracy'])
return model
model = build_model()
losses=[brier_score]
model.compile(loss=losses, optimizer='adam', metrics=['accuracy'])
# this line causes the error
model.fit(x=x, y=y, epochs=100, verbose=0)
ValueError: slice index 2 of dimension 1 out of bounds. for '{{node brier_score/strided_slice_2}} = StridedSlice[Index=DT_INT32, T=DT_FLOAT, begin_mask=1, ellipsis_mask=0, end_mask=1, new_axis_mask=0, shrink_axis_mask=2](Cast, brier_score/strided_slice_2/stack, brier_score/strided_slice_2/stack_1, brier_score/strided_slice_2/stack_2)' with input shapes: [?,2], [2], [2], [2] and with computed input tensors: input[1] = <0 2>, input[2] = <0 3>, input[3] = <1 1>.
You are mixing up the order of y_true and y_pred in brier_score. Here is a working example:
import tensorflow as tf
from keras import models
from keras import layers
import pandas as pd
import numpy as np
x = np.random.random((500, 1))
y = {'Age_Years': np.squeeze(x, axis=1), 'target': np.random.randint(2, size=500)}
def calc_prob(param1, param2, param3, age):
prob = (((100*param1*tf.math.pow((100/param3),-(tf.math.pow((age/param2),param1))))*tf.math.pow(age/param2,param1)*tf.math.log(100/param3))/age)/100
return prob
# this serves as the custom cost function
def brier_score(y_true, y_pred):
prob = calc_prob(y_pred[:, 0], y_pred[:, 1], y_pred[:, 2], tf.cast(y_true['Age_Years'], dtype=tf.float32))
brier_score = tf.reduce_mean((prob - tf.cast(y_true['target'], dtype=tf.float32)) ** 2, axis=-1, keepdims=True)
return brier_score
def build_model():
model = models.Sequential()
model.add(layers.Dense(1, activation='relu', input_shape=(x.shape[1],)))
model.add(layers.Dense(5, activation='relu'))
model.add(layers.Dense(3, activation='softmax'))
return model
model = build_model()
optimizer = tf.keras.optimizers.Adam()
batch_size = 10
dataset = tf.data.Dataset.from_tensor_slices((x,y)).batch(batch_size)
epochs = 2
for epoch in range(epochs):
print("\nStart of epoch %d" % (epoch,))
for step, (x_batch_train, y_batch_train) in enumerate(dataset):
with tf.GradientTape() as tape:
logits = model(x_batch_train, training=True)
loss_value = brier_score(y_batch_train, logits)
grads = tape.gradient(loss_value, model.trainable_weights)
optimizer.apply_gradients(zip(grads, model.trainable_weights))
My model has two inputs and I want to calculate the loss of the two inputs separately because the loss of input 2 has to be multiplied by a weight. Then add up these two losses as the final loss for the model. The structure is somehow like this:
This is my model:
def final_loss(y_true, y_pred):
loss = x_loss_value.output + y_model.output*weight
return loss
def mymodel(input_shape): #pooling=max or avg
img_input1 = Input(shape=(input_shape[0], input_shape[1], input_shape[2], ))
image_input2 = Input(shape=(input_shape[0], input_shape[1], input_shape[2], ))
#for input1
x = Conv2D(32, (3, 3), strides=(2, 2))(img_input1)
x_dense = Dense(2, activation='softmax', name='predictions')(x)
x_loss_value = my_categorical_crossentropy_layer(x)[input1_y_true, input1_y_pred]
x_model = Model(inputs=img_input1, outputs=x_loss_value)
#for input2
y = Conv2D(32, (3, 3), strides=(2, 2))(image_input2)
y_dense = Dense(2, activation='softmax', name='predictions')(y)
y_loss_value = my_categorical_crossentropy_layer(y)[input2_y_true, input2_y_pred]
y_model = Model(inputs=img_input2, outputs=y_loss_value)
concat = concatenate([x_model.output, y_model.output])
final_dense = Dense(2, activation='softmax')(concat)
# Create model.
model = Model(inputs=[img_input1,image_input2], output = final_dense)
return model
model.compile(optimizer = optimizers.adam(lr=1e-7), loss = final_loss, metrics = ['accuracy'])
Most of the related solutions I found just customize the final loss and change the loss in Model.complie(loss=customize_loss).
However, I need to apply different losses for different inputs. I'm trying to use a customized layer like this, and get my loss value for final the loss calculation:
class my_categorical_crossentropy_layer1(Layer):
def __init__(self, **kwargs):
self.is_placeholder = True
super(my_categorical_crossentropy_layer1, self).__init__(**kwargs)
def my_categorical_crossentropy_loss(self, y_true, y_pred):
y_pred = K.constant(y_pred) if not K.is_tensor(y_pred) else y_pred
y_true = K.cast(y_true, y_pred.dtype)
return K.categorical_crossentropy(y_true, y_pred, from_logits=from_logits)
def call(self, y_true, y_pred):
loss = self.my_categorical_crossentropy_loss(y_true, y_pred)
self.add_loss(loss, inputs=(y_true, y_pred))
return loss
But, inside the keras model, I can't figure out how to get the y_true and y_pred of the current epoch/batch for my loss layer.
So I can't add x = my_categorical_crossentropy_layer()[y_true, y_pred] to my model.
Is there any way to do the variable calculation like this in the keras model?
Further, can Keras get the previous epoch's training loss or val loss during training process?
I want to apply the previous epoch's training loss as my weight in the final loss.
this is my proposal...
your it's a double binary classification problem that you want to carry out using a single fit. the first thing to notice is that you need to take care of dimensionality: your input is 4d while your target is 2d one-hot encoded so your network needs something to reduce dimensionality, for example, flatten or global pooling. after this, you can start fitting creating a single model with two inputs and two outputs and use two losses. in your case, the losses are weighted categorical_crossentropy. keras enable by default to set the loss weights using loss_weights parameters. to reproduce the formula loss1*1+loss2*W set the weights to [1, W]. you can use the loss_weights parameter also specifying different losses for your output in this way losses=[loss1, loss2, ....] which are linearly combined with the weights specified in the loss_weights
below a working example
input_shape = (28,28,3)
n_sample = 10
# create dummy data
X1 = np.random.uniform(0,1, (n_sample,)+input_shape) # 4d
X2 = np.random.uniform(0,1, (n_sample,)+input_shape) # 4d
y1 = tf.keras.utils.to_categorical(np.random.randint(0,2, n_sample)) # 2d
y2 = tf.keras.utils.to_categorical(np.random.randint(0,2, n_sample)) # 2d
def mymodel(input_shape, weight):
img_input1 = Input(shape=(input_shape[0], input_shape[1], input_shape[2], ))
img_input2 = Input(shape=(input_shape[0], input_shape[1], input_shape[2], ))
# for input1
x = Conv2D(32, (3, 3), strides=(2, 2))(img_input1)
x = GlobalMaxPool2D()(x) # pass from 4d to 2d
x = Dense(2, activation='softmax', name='predictions1')(x)
# for input2
y = Conv2D(32, (3, 3), strides=(2, 2))(img_input2)
y = GlobalMaxPool2D()(y) # pass from 4d to 2d
y = Dense(2, activation='softmax', name='predictions2')(y)
# Create model
model = Model([img_input1,img_input2], [x,y])
model.compile(optimizer = 'adam', loss = 'categorical_crossentropy', metrics = ['accuracy'],
loss_weights=[1,weight])
return model
weight = 0.3
model = mymodel(input_shape, weight)
model.summary()
model.fit([X1,X2], [y1,y2], epochs=2)
I have a siamese network and I want to perform a grid seach on it using GridSearchCV.
So I create a model using the following function:
def createMod(learn_rate=0.01, optimizer='Adam'):
#K.clear_session()
# network definition
base_network = create_base_network(input_shape)
input_a = Input(shape=input_shape)
input_b = Input(shape=input_shape)
# because we re-use the same instance `base_network`,
# the weights of the network will be shared across the two branches
processed_a = base_network(input_a)
processed_b = base_network(input_b)
distance = Lambda(euclidean_distance,
output_shape=eucl_dist_output_shape)([processed_a, processed_b])
prediction = Dense(1,activation='sigmoid')(distance)
model = Model([input_a, input_b], prediction)
if(optimizer=='SGD'):
opt = SGD(lr=learn_rate)
elif (optimizer=='RMSprop'):
opt = RMSprop(lr=learn_rate)
else:
opt = Adam(lr=learn_rate)
model.compile(loss='binary_crossentropy', optimizer=opt, metrics=[accuracy])
return model
And then I create the KerasClassifier and GridSearch as follows:
model = KerasClassifier(build_fn=createMod, verbose=0)
param_grid = dict(epochs=epochs, batch_size=batch_size, learn_rate=learn_rate,optimizer=optimizer)
grid = GridSearchCV(estimator=model, param_grid=param_grid, n_jobs=-1, cv=5)
X = [tr_pairs[:, 0], tr_pairs[:, 1]]
grid_result = grid.fit(X, tr_y)
However this throws the following value error:
ValueError: Found input variables with inconsistent numbers of samples: [2, 1054]
The shape of tr_pairs[:, 0] and tr_pairs[:, 1] is (1054, 6) and tr_y is (1054,)
The code for the base networks is:
def create_base_network(input_shape):
K.clear_session()
encoder = build_encoder(latent_dim, n_in)
decoder = build_decoder(latent_dim, n_in)
item = Input(shape=(n_in, ))
encoded_repr = encoder(item)
reconstructed_item = decoder(encoded_repr)
autoencoder = Model(item, reconstructed_item)
return autoencoder
The code for the encoder and decoder are:
def build_encoder(latent_dim, input_dim):
input_layer = Input(shape=(input_dim, ))
h = Dense(32, activation='relu', activity_regularizer=regularizers.l1(10e-5))(input_layer)
h = Dropout(0.1)(h)
h = Dense(64, activation='relu')(h)
h = Dropout(0.1)(h)
# h = Dense(128, activation='relu')(h)
# h = Dropout(0.1)(h)
latent_repr = Dense(latent_dim, activation='relu')(h)
return Model(input_layer, latent_repr)
def build_decoder(latent_dim, input_dim):
model = Sequential()
# model.add(Dense(128, input_dim=latent_dim, activation='relu'))
# model.add(Dropout(0.1))
model.add(Dense(64, activation='relu'))
model.add(Dropout(0.1))
model.add(Dense(32, activation='relu'))
model.add(Dropout(0.1))
model.add(Dense(input_dim, activation='relu'))
z = Input(shape=(latent_dim,))
item = model(z)
return Model(z, item)
The code works when I do the normal keras model's .fit function but doesn't work here.. is there a problem somewhere in my code or is it just not possible to feed multiple inputs in Grid Search and if that is the case is there a way I can still perform the grid search?
this is workaround to pass multiple input. I create a dummy model that receives a SINGLE input in the format (n_sample, 2, 6) and then split it into two parts using Lambda layer. you can modify this according to your siamese structure.
def createMod(optimizer='Adam'):
combi_input = Input((2,6)) # (n_sample, 2, 6)
input_a = Lambda(lambda x: x[:,0])(combi_input) # (n_sample, 6)
input_b = Lambda(lambda x: x[:,1])(combi_input) # (n_sample, 6)
c = Concatenate()([input_a,input_b])
x = Dense(32)(c)
prediction = Dense(1,activation='sigmoid')(x)
model = Model(combi_input, prediction)
model.compile(loss='binary_crossentropy', optimizer=optimizer, metrics='accuracy')
return model
tr_pairs = np.random.uniform(0,1, (1054, 2, 6))
tr_y = np.random.randint(0,2, 1054)
model = tf.keras.wrappers.scikit_learn.KerasClassifier(build_fn=createMod, verbose=0)
batch_size = [10, 20]
epochs = [10, 5]
optimizer = ['adam','SGD']
param_grid = dict(batch_size=batch_size, epochs=epochs)
grid = GridSearchCV(estimator=model, param_grid=param_grid, n_jobs=-1, cv=3)
grid_result = grid.fit(tr_pairs, tr_y)
I'm converting a basic LSTM many-to-one architecture to predict the next single element in a sequence, written in Keras to Pytorch. NN architecture is the following (whole code can be found here):
model = Sequential()
model.add(LSTM(
512,
input_shape=(network_input.shape[1], network_input.shape[2]),
return_sequences=True
))
model.add(Dropout(0.3))
model.add(LSTM(512, return_sequences=True))
model.add(Dropout(0.3))
model.add(LSTM(512))
model.add(Dense(256))
model.add(Dropout(0.3))
model.add(Dense(n_vocab))
model.add(Activation('softmax'))
model.compile(loss='categorical_crossentropy', optimizer='rmsprop')
Running both models with the same data (yes, I've explicitly checked that), both start with a loss value ~ 4, but after 100 epochs or so, Keras already reached a loss ~ 0.02, which gives the desired results.
However, Pytorch model is stuck around ~ 3.4 after 20 epochs. I've tried many things:
Play with LR: It explodes when LR is too high, so this means that at least parameters are being updated.
Different optimizers, SGD, Adam, RMSprop, but same results with all.
Swap between .view[], .squeeze_ and indexing when accessing last sequence element.
Add, remove and modify non-linear activation functions and dropout.
Remove manual initialization for x_0 and h_0.
Here is the code for my model:
class NNP_RNN(nn.Module):
def __init__(self):
super(NNP_RNN, self).__init__()
self.lstm_1 = nn.LSTM(input_size=1, hidden_size=512, batch_first=True)
self.lstm_2 = nn.LSTM(input_size=512, hidden_size=512, batch_first=True)
self.lstm_3 = nn.LSTM(input_size=512, hidden_size=512, batch_first=True)
self.dense_1 = nn.Linear(in_features=512, out_features=256)
self.dense_2 = nn.Linear(in_features=256, out_features=58)
def forward(self, x):
batch_size = x.size(0)
h_0 = NNP_RNN.init_hidden((1, batch_size, 512))
c_0 = NNP_RNN.init_hidden((1, batch_size, 512))
x, _ = self.lstm_1(x, (h_0, c_0))
x = F.dropout(x, 0.3)
x, _ = self.lstm_2(x, (h_0, c_0))
x = F.dropout(x, 0.2)
_, (x, _) = self.lstm_3(x, (h_0, c_0))
x = x.squeeze_(0)
x = self.dense_1(x)
x = F.dropout(x, 0.1)
x = self.dense_2(x)
return x
#staticmethod
def init_hidden(dims):
return torch.zeros(dims, device=device)
And the training process:
optimizer = torch.optim.Adam(model.parameters(), lr=0.05)
scheduler = torch.optim.lr_scheduler.ReduceLROnPlateau(optimizer, factor=0.2, verbose=True, patience=5)
criterion = nn.CrossEntropyLoss()
for epoch in range(1, epochs + 1):
epoch_loss = 0
epoch_corrects = 0
for features, labels in tqdm(data, ncols=800):
features = features.to(device)
labels = labels.to(device)
optimizer.zero_grad()
batch_size = features.size(0)
output = model(features)
loss = criterion(output, labels)
loss.backward()
optimizer.step()
corrects = torch.argmax(output, dim=1)
corrects = torch.eq(corrects, labels).sum().item()
epoch_corrects += corrects
epoch_loss += loss.clone() * batch_size
epoch_loss /= len(data.dataset)
epoch_corrects /= len(data.dataset)
print(f'Loss epoch #{epoch} = {epoch_loss:.10f}, Accuracy = {epoch_corrects}')
scheduler.step(epoch_loss)
I was trying to write masked MSE loss:
def mae_loss_masked(mask):
def loss_fn(y_true, y_pred):
abs_vec = tf.multiply(tf.abs(y_pred-y_true), mask)
loss = tf.reduce_mean(abs_vec)
return loss
return loss_fn
My model:
def MobileNet_v1():
# MobileNet with dense layer on top
# Keras 2.1.6
mobilenet = MobileNet(input_shape=(config.IMAGE_H, config.IMAGE_W, config.N_CHANNELS),
alpha=1.0,
depth_multiplier=1,
include_top=False,
weights='imagenet'
)
x = Flatten()(mobilenet.output)
x = Dropout(0.5)(x)
x = Dense(config.N_LANDMARKS * 2, activation='linear')(x)
# -------------------------------------------------------
model = Model(inputs=mobilenet.input, outputs=x)
optimizer = Adadelta()
model.compile(optimizer=optimizer, loss=mae_loss_masked)
model.summary()
import sys
sys.exit()
return model
But it give an error:
TypeError: mae_loss_masked() takes 1 positional argument but 2 were given
Also a question how batch generator output should look like in this case.