I want to use LightGBM to fit a function curve,but in the examples of LightGBM's dataset,every record has a label column.
I don't know how to create my training dataset and testset.
Related
I want to train my custom dataset using the yolov4 algorithm.
The difference between my data and the coco dataset is that my objects label's are as follows:
label annotation: <object-class> <x_center> <y_center> <width> <height> <d>
We always use pre-trained files to object detection, but given that my data has different labels, Do I need to change pre-trained data?
how should I train the dataset?
I am trying to build a Keras model to implement to approach explained in this paper.
Context of my implementation:
I have two different kinds of data representing the same set of classes(labels) that needs to be classified. The 1st kind is Image data, and the second kind is EEG data (a time series sequence).
I know that to classify image data we can use CNN models like this:
model.add(Conv2D(filters=256, kernel_size=(11,11), strides=(1,1), padding='valid'))
model.add(Activation('relu'))
model.add(Dense(1000))
model.add(Activation('relu'))
model.add(Dropout(0.4))
# Batch Normalisation
model.add(BatchNormalization())
# Output Layer
model.add(Dense(40))
model.add(Activation('softmax'))
And to classify sequence data we can use LSTM models like this:
model.add(LSTM(units = 50, return_sequences = True))
model.add(Dropout(0.2))
model.add(Flatten())
model.add(Dense(32, activation='relu'))
model.add(Dense(128, activation='relu'))
model.add(Dropout(0.5))
model.add(Dense(40, activation='softmax'))
But the approach of the paper above shows that EEG feature vectors can be mapped with image vectors through regression like this:
The first approach is to train a CNN to map images to corresponding
EEG feature vectors. Typically, the first layers of CNN attempt to
learn the general (global) features of the images, which are common
between many tasks, thus we initialize the weights of these layers
using pre-trained models, and then learn the weights of the last
layers from scratch in an end-to-end setting. In particular, we used
the pre-trained AlexNet CNN, and modified it by replacing the
softmax classification layer with a regression layer (containing as
many neurons as the dimensionality of the EEG feature vectors),
using Euclidean loss as the objective function.
The second approach consists of extracting image features using
pre-trained CNN models and then employ regression methods to map
image features to EEG feature vectors. We used our fine-tuned
AlexNet as feature extractors by
reading the output of the last fully connected layer, and then
applied several regression methods (namely, k-NN regression, ridge
regression, random forest regression) to obtain the predicted
feature vectors
I am not able to comprehend how to code the above two approaches. I have never used a regressor for feature mapping and then do classification. Any leads on this are much appreciated.
In my understanding the training data consists of (eeg_signal,image,class_label) triplets.
Train the LSTM model with input=eeg_signal, output=class_label. Loss is crossentropy.
Peel off the last layer of the LSTM model. Let's say the pre-last layer's output is a vector of size 20. Let's call it eeg_representation.
Run this truncated model on all your eeg_signal inputs, save the output of eeg_representation. You will get a tensor of [batch, 20]
Take that AlexNet mentioned in the paper (or any other image classifier), peel off the last layer. Let's say the pre-last layer's output is a vector of size 30. Let's call it image_representation.
Stich a linear layer to the end of the previous layer. This layer will convert image_representation to eeg_representation. It has 20 x 30 weight.
Train the stiched model on (image, eeg_representation) pairs. Loss is the Euclidean distance.
And now the fun part: Stich together model trained in step 7. and the peeled off part of model trained in step 1. If you input an image, you will get class predictions.
This sound like not a big deal (because we do image classification all the time), but if this is really working, it means that this is a "prediction that is running through our brains" :)
Thank you bringing up this question and linking the paper.
I feel I just repeated what's in your question and in the the paper.
I would be beneficial to have some toy dataset to be able to provide code examples.
Here's a Tensorflow tutorial on how to "peel off" the last layer of a pretrained image classification model.
I have a dataset in the form (input_text, embedding_of_input_text), where embedding_of_input_text is an embedding of dimension 512 produced by another model (DistilBERT) when given as input input_text.
I would like to fine-tune BERT on this dataset such that it learns to produce similar embeddings (i.e. a kind of mimicking).
Furthermore, by default BERT returns embeddings of dimension 768, while here embedding_of_input_text are embeddings of dimension 512.
Which is the correct way to to that within the HuggingFace library?
you can get the tokenizer of the dataset
and add the neural network to get embedding of dimension 512.
However,what is the meaning of this operation.
To whom may it concern,
Is it possible to plot an exploratory variable versus the target in h2o? I want to know whether it is possible to carry out basic data exploration in h2o, or whether it is not designed for that.
Many thanks in advance,
Kere
the main plotting functionality for an H2O frame is for histograms (hist() in python and h2o.hist() in R).
Within Flow you can do basic data exploration if you import your dataframe, then click on inspect and then, next to the hyperlinked columns, you'll see a plot button which will let you get bar charts of counts for example and other plot types.
You can also easily convert single columns you want to plot into a pandas or R dataframe with
H2OFrame.as_data_frame() in python
as.data.frame.H2OFrame in R and then use the native python and R plotting methods
I extract 2 edge features (Hog feature and sobel operator) from a single image.
How can i create an image feature dataset in Scikit-learn python, like iris_dataset ?
In the library there are csv files which represent datasets. A csv file containing only numbers. How were generate these numbers? feature extraction?
unfortunately i saw only a java tutorial here http://www.coccidia.icb.usp.br/coccimorph/tutorials/Tutorial-2-Creating-..., at 5 point talk about generating the training matrices (average and co-variance matrices)?
There is any function in Scikit who generate these training arrays?
You don't need to wrap your data as a CSV file to load it as a dataset. scikit-learn models have a fit method that expects:
as first argument that is a regular numpy array (or scipy.sparse matrices) with shape (n_samples, n_features) (most often with dtype=numpy.float64) to encode the features vector for each sample in the training set,
and for supervised classification models, a second argument with shape (n_samples,) and dtype=numpy.int32 to encode the class label assignments encoded as integer values for each sample of the training set.
If you don't know the basic numpy datastructure and what shape and dtype mean, I stongly advise you to have a look at a tutorial such as SciPy Lecture Notes.
Edit: If you really need to read / write numerical CSV to / from numpy arrays, you can use numpy.loadtxt / numpy.savetxt