Develop Reference

A Classifier Network Seems to be "Forgetting" older samples

This is a strange problem: Imagine a neural network classifier. It is a simple linear layer followed by a sigmoid activation that has an input size of 64, and an output size of 112. There also are 112 training samples, where I expect the output to be a one-hot vector. So the basic structure of a training loop is as follows, where samples is a list of integer indices:
model = nn.Sequential(nn.Linear(64,112),nn.Sequential())
loss_fn = nn.BCELoss()
optimizer = optim.AdamW(model.parameters(),lr=3e-4)
for epoch in range(500):
for input_state, index in samples:
one_hot = torch.zeros(112).float()
one_hot[index] = 1.0
optimizer.zero_grad()
prediction = model(input_state)
loss = loss_fn(prediction,one_hot)
loss.backward()
optimizer.step()
This model does not perform well, but I don't think it's a problem with the model itself, but rather how it's trained. I think that this is happening because for the most part, all of the one_hot tensor is zeros, that the model just tends to gravitate toward all of the outputs being zeros, which is what's happening. The question becomes: "How does this get solved?" I tried using the average loss with all the samples, to no avail. So what do I do?

So this is very embarrassing, but the answer actually lies in how I process my data. This is a text-input project, so I used basic python lists to create blocks of messages, but when I did this, I ended up making it so that all of the inputs the net got were the same, but the output was different every time. I solved tho s problem with the copy method.

CNN-LSTM model for sentiment analysis has low validation accuracy

I am working on a project to implement CNN-LSTM sentiment analysis. Below is the code
from keras.models import Sequential
from keras import regularizers
from keras import backend as K
from keras.callbacks import ModelCheckpoint
from keras.layers import Dense, Conv1D , MaxPool1D , Flatten , Dropout
from keras.layers import BatchNormalization
from keras import regularizers
model7 = Sequential()
model7.add(Embedding(max_words, 40,input_length=max_len)) #The embedding layer
model7.add(Conv1D(20, 5, activation='relu', kernel_regularizer = regularizers.l2(l = 0.0001), bias_regularizer=regularizers.l2(0.01)))
model7.add(Dropout(0.5))
model7.add(Bidirectional(LSTM(20,dropout=0.5, kernel_regularizer=regularizers.l2(0.01), recurrent_regularizer=regularizers.l2(0.01), bias_regularizer=regularizers.l2(0.01))))
model7.add(Dense(1,activation='sigmoid'))
model7.compile(optimizer='adam',loss='binary_crossentropy', metrics=['accuracy'])
checkpoint7 = ModelCheckpoint("best_model7.hdf5", monitor='val_accuracy', verbose=1,save_best_only=True, mode='auto', period=1,save_weights_only=False)
history = model7.fit(X_train_padded, y_train, epochs=10,validation_data=(X_test_padded, y_test),callbacks=[checkpoint7])
Even after adding regularizers and dropout, my model has very high validation loss and low accuracy.
Epoch 3: val_accuracy improved from 0.54517 to 0.57010, saving model to best_model7.hdf5
2188/2188 [==============================] - 290s 132ms/step - loss: 0.4241 - accuracy: 0.8301 - val_loss: 0.9713 - val_accuracy: 0.5701
My train and test data:
train: (70000, 7)
test: (30000, 7)
train['sentiment'].value_counts()
1 41044
0 28956
test['sentiment'].value_counts()
1 17591
0 12409
Can anyone please let me know how to reduce overfitting.

Since your code works, I believe that your network is failing silently by 'not learning' a lot from the data. Here's a list of some of the things you can generally check:
Is your textual data well transformed into numerical data? Is it well reprented using TF-IDF or bag of words or any other method that returns a numerical representation?
I see that you imported batch normalization but you do not apply it. Batch norm actually helps and most importantly, does the job of regularizers since each input to each layer is normalized using the mini-batch the network has seen. So maybe remove your L2 regularizations in all layers and apply a simple batch norm instead which should reduce overfitting (also, use it without the drop out since some empirical studies show that they should not be combined together)
Your embedding output is currently set to 40, that is 40 numerical elements of a text vector that may contain more than 10,000 elements. It seems a bit low. Try something more 'standard' such as 128 or 256 instead of 40.
Lastly, you set the adam optimizer with all the default parameters. However, the learning rate can have a big impact on the way your loss function is computed. As I am sure you know, the gradient step uses this learning rate to progress in its calculation of the derivatives for each neuron. the default is learning_rate=0.001. So try the following code and increase a bit the learning rate (for example 0.01 or even 0.1).
A simple example :
# define model
model = Sequential()
model.add(LSTM(32)) # or CNN
model.add(BatchNormalization())
model.add(Dense(1))
# define optimizer
optimizer = keras.optimizers.Adam(0.01)
# define loss function
loss = keras.losses.binary_crossentropy
# define metric to optimize
metric = [keras.metrics.Accuracy(name='accuracy')] # you can add more
# compile model
model.compile(optimizer=optimizer, loss=loss, metrics=metric)
Final thought: I see that you went for a combination of CNN and LSTM which has great merite. However, it is always recommended to try a simple MLP network to establish a baseline score that you may later try to beat. Does a simple MLP with 1 or 2 layers and not a lot of units produce a low accuracy score as well? If it performs better than maybe the problem is in the implementation or in the hyper parameters that you chose for the layers (or even theoretical).
I hope this answer helps and cheers!

Could not increase accuracy from a fixed threshold using Keras Dense layer ANN

I'm learning the simplest neural networks using Dense layers using Keras. I'm trying to implement face recognition on a relatively small dataset (In total ~250 images with 50 images per class).
I've downloaded the images from google images and resized them to 100 * 100 png files. Then I've read those files into a numpy array and also created a one hot label array for training my model.
Here is my code for processing the training data:
X, Y = [], []
feature_map = {
'Alia Bhatt': 0,
'Dipika Padukon': 1,
'Shahrukh khan': 2,
'amitabh bachchan': 3,
'ayushmann khurrana': 4
}
for each_dir in os.listdir('.'):
if os.path.isdir(each_dir):
for each_file in os.listdir(each_dir):
X.append(cv2.imread(os.path.join(each_dir, each_file), -1).reshape(1, -1))
Y.append(feature_map[os.path.basename(each_file).split('-')[0]])
X = np.squeeze(X)
X = X / 255.0 # normalize the training data
Y = np.array(Y)
Y = np.eye(5)[Y]
print (X.shape)
print (Y.shape)
This is printing (244, 40000) and (244, 5). Here is my model:
model = Sequential()
model.add(Dense(8000, input_dim = 40000, activation = 'relu'))
model.add(Dense(1200, activation = 'relu'))
model.add(Dense(700, activation = 'relu'))
model.add(Dense(100, activation = 'relu'))
model.add(Dense(5, activation = 'softmax'))
# Compile model
model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])
# Fit the model
model.fit(X, Y, epochs=25, batch_size=15)
When I train the model, It stuck at the accuracy 0.2172, which is almost the same as random predictions (0.20).
I've also tried to train mode with grayscale images but still not getting expected accuracy. Also tried with different network architectures by changing the number of hidden layers and neurons in hidden layers.
What am I missing here? Is my dataset too small? or am I missing any other technical detail?
For more details of code, here is my notebook: https://colab.research.google.com/drive/1hSVirKYO5NFH3VWtXfr1h6y0sxHjI5Ey

Two suggestions I can make:
Your data set is probably too small. If you are splitting training and validation at 80/20, that means you are only training on 200 images, which is probably too small. Try increasing your data set to see if results improve.
I would recommend adding Dropout to each layer of your network as your training set is so small. Your network is most likely over-fitting your training data set since it is so small, and Dropout is an easy way to help avoid this problem.
Let me know if these suggestions make a difference!

I agree that the dataset is too small, 50 instances of each person is probably not enough. You can use data augmentation with the keras ImageDataGenerator method to increase the number of images, and rewrite your numpy reshaping code as a pre-processing function for the generator. I also noticed that you haven't shuffled the data, so the network is likely predicting the first class for everything (which is maybe why the accuracy is near random chance).
If increasing the dataset size doesn't help, you'll probably have to play around with the learning rate for the Adam optimizer.

CNN on small dataset is overfiting

I want to classify pattern on image. My original image shape are 200 000*200 000 i reshape it to 96*96, pattern are still recognizable with human eyes. Pixel value are 0 or 1.
i'm using the following neural network.
train_X, test_X, train_Y, test_Y = train_test_split(cnn_mat, img_bin["Classification"], test_size = 0.2, random_state = 0)
class_weights = class_weight.compute_class_weight('balanced',
np.unique(train_Y),
train_Y)
train_Y_one_hot = to_categorical(train_Y)
test_Y_one_hot = to_categorical(test_Y)
train_X,valid_X,train_label,valid_label = train_test_split(train_X, train_Y_one_hot, test_size=0.2, random_state=13)
model = Sequential()
model.add(Conv2D(24,kernel_size=3,padding='same',activation='relu',
input_shape=(96,96,1)))
model.add(MaxPool2D())
model.add(Conv2D(48,kernel_size=3,padding='same',activation='relu'))
model.add(MaxPool2D())
model.add(Conv2D(64,kernel_size=3,padding='same',activation='relu'))
model.add(MaxPool2D())
model.add(Flatten())
model.add(Dense(128, activation='relu'))
model.add(Dense(256, activation='relu'))
model.add(Dense(16, activation='softmax'))
model.compile(optimizer="adam", loss="categorical_crossentropy", metrics=["accuracy"])
train = model.fit(train_X, train_label, batch_size=80,epochs=20,verbose=1,validation_data=(valid_X, valid_label),class_weight=class_weights)
I have already run some experiment to find a "good" number of hidden layer and fully connected layer. it's probably not the most optimal architecture since my computer is slow, i just ran different model once and selected best one with matrix confusion, i didn't use cross validation,I didn't try more complex architecture since my number of data is small, i have read small architecture are the best, is it worth to try more complex architecture?
here the result with 5 and 12 epoch, bach size 80. This is the confusion matrix for my test set
As you can see it's look like i'm overfiting. When i only run 5 epoch, most of the class are assigned to class 0; With more epoch, class 0 is less important but classification is still bad
I added 0.8 dropout after each convolutional layer
e.g
model.add(Conv2D(48,kernel_size=3,padding='same',activation='relu'))
model.add(MaxPool2D())
model.add(Dropout(0.8))
model.add(Conv2D(64,kernel_size=3,padding='same',activation='relu'))
model.add(MaxPool2D())
model.add(Dropout(0.8))
With drop out, 95% of my image are classified in class 0.
I tryed image augmentation; i made rotation of all my training image, still used weighted activation function, result didnt improve. Should i try to augment only class with small number of image? Most of the thing i read says to augment all the dataset...
To resume my question are:
Should i try more complex model?
Is it usefull to do image augmentation only on unrepresented class? then should i still use weight class (i guess no)?
Should i have hope to find a "good" model with cnn when we see the size of my dataset?

I think according to the imbalanced data, it is better to create a custom data generator for your model so that each of it's generated data batch, contains at least one sample from each class. And also it is better to use Dropout layer after each dense layer instead of conv layer. For data augmentation it is better to at least use combination of rotate, horizontal flip and vertical flip. there are some other approaches for data augmentation like using GAN network or random pixel replacement.
For Gan you can check This SO post
For using Gan as data augmenter you can read This Article.
For combination of pixel level augmentation and GAN pixel level data augmentation

What I used - in a different setting - was to upsample my data with ADASYN. This algorithm calculates the amount of new data required to balance your classes, and then takes available data to sample novel examples.
There is an implementation for Python. Otherwise, you also have very little data. SVMs are good performing even with little data. You might want to try them or other image classification algorithms depending where the expected pattern is always at the same position, or varies. Then you could also try the Viola–Jones object detection framework.

TimeDistributed layer and return sequences etc for LSTM in Keras

Sorry I am new to RNN. I have read this post on TimeDistributed layer.
I have reshaped my data in to Keras requried [samples, time_steps, features]: [140*50*19], which means I have 140 data points, each has 50 time steps, and 19 features. My output is shaped [140*50*1]. I care more about the last data point's accuracy. This is a regression problem.
My current code is :
x = Input((None, X_train.shape[-1]) , name='input')
lstm_kwargs = { 'dropout_W': 0.25, 'return_sequences': True, 'consume_less': 'gpu'}
lstm1 = LSTM(64, name='lstm1', **lstm_kwargs)(x)
output = Dense(1, activation='relu', name='output')(lstm1)
model = Model(input=x, output=output)
sgd = SGD(lr=0.00006, momentum=0.8, decay=0, nesterov=False)
optimizer = sgd
model.compile(optimizer=optimizer, loss='mean_squared_error')
My questions are:
My case is many-to-many, so I need to use return_sequences=True? How about if I only need the last time step's prediction, it would be many-to-one. So I need to my output to be [140*1*1] and return_sequences=False?
Is there anyway to enhance my last time points accuracy if I use many-to-many? I care more about it than the other points accuracy.
I have tried to use TimeDistributed layer as
output = TimeDistributed(Dense(1, activation='relu'), name='output')(lstm1)
the performance seems to be worse than without using TimeDistributed layer. Why is this so?
I tried to use optimizer=RMSprop(lr=0.001). I thought RMSprop is supposed to stabilize the NN. But I was never able to get good result using RMSprop.
How do I choose a good lr and momentum for SGD? I have been testing on different combinations manually. Is there a cross validation method in keras?

So:
Yes - return_sequences=False makes your network to output only a last element of sequence prediction.
You could define the output slicing using the Lambda layer. Here you could find an example on how to do this. Having the output sliced you can provide the additional output where you'll feed the values of the last timestep.
From the computational point of view these two approaches are equivalent. Maybe the problem lies in randomness introduced by weight sampling.
Actually - using RMSProp as a first choice for RNNs is a rule of thumb - not a general proved law. Moreover - it is strongly adviced not to change it's parameters. So this might cause the problems. Another thing is that LSTM needs a lot of time to stabalize. Maybe you need to leave it for more epochs. Last thing - is that maybe your data could favour another activation function.
You could use a keras.sklearnWrapper.