-
Autoencoder + Weights
Hi, I'm training a deep learning model for a classification task with a weights column. I also want to use a pretrained autoencoder to initialize the model. I don't really care about the weights for the autoencoder, but it seems if I don't use a weights column in the autoencoder, I cannot do so later in the classifier. But…
-
Guide on Generative AI for Personalization
Generative AI is a game-changer, especially when it comes to making things personal. Think of the last time you got awesome song recommendations on a music app or spot-on movie suggestions. That's Generative AI at work, tailoring your experience based on what it knows about you. It's like having a virtual assistant that…
-
h2o automl hyperparameters
Hi , I see there is a getgrid api to get the grid used for hyperparameter search https://docs.h2o.ai/h2o/latest-stable/h2o-docs/grid-search.html Is it also possible to get programmatically the grid of values used for hyperparameters when using the automl class https://docs.h2o.ai/h2o/latest-stable/h2o-docs/automl.html We…
-
Using H2O Automl model in a constraint for optimization
Hello! I am working on an optimization problem using NeverGrad with a H2O AutoMl model. The objective function is a linear function of decision variables which is an array of 6 scalars. The constraints for the objective function are upper and lower bounds on the value of output variables. The output variables are obtained…
-
Single column model usage
Having had AutoML() reccomend a model after I experiemented with a single column of data and needed to identify a target as a step-1 of the single data to train the succesful model I now want to use the model on test (live) data. The model however wants a 2 column input and when I use a padded secondnd column "target" =…
-
Training XGBoost in a distributed environment
I've read in the docs that XGBoost supports multiprocessing. I have successfully run it in a single node environment, where I verified that it was using all the cores. However, when I try it in a distributed setting, it does not give an error, it just gets stuck at 0%. I'm not sure if it currently does not support…
-
Natural Language Processing with Transformers (Hugging Face Study Group)
Hello I am Prabhav. I was following the study group here. I hope using the discussion forums one can form a collection of doubts about concepts and implementation for effective solutions and brainstorming as well as future reference.
-
how to dynamically scale autoencoder with different input features
I am trying to use one code and create autoencoder with following configuration hidden=c(6,6,6) as shown below. Question: 1) How can I use the same code for multiple input dataframe such that one can have 20 features and other can have 400 and 1000 and so on. Goal is to change the hidden vector 2) What does hidden=c(6,6,6)…
-
How do I generate a formula from best DNN model?
Hello all, I am using DeepLearning to predict salinity values in a local bay. These predicted values are at sample stations throughout the bay. I need to generate a formula from the best model generated to apply to other areas of the bay. I have attempted to use variable importance to do so, but it did not yield accurate…
-
Why classification model is not selected ?
Hello ! I'm trying to carry out a classification model on data set with malignant and benign tumors. The first time, I was able to select classification model. I lost the model because of an internet bug. Thus, I carried out a second tryied but this time, regression model is selected and I cannot select the classification…
-
Pull Requests Tracker for Deep Learning Book Study Group
Hi, Kindly use this discussion to discuss the PRs that you are working on or raising for the study group to help with coordination. YouTube Study Group Link GitHub Repository
-
Help me in H2o Timeout error
Hello h2o experts! I am running a deep learning model (binary classification) with following grid search code # hyper_params <- list( activation = c("Rectifier", "Maxout", "Tanh", "RectifierWithDropout", "MaxoutWithDropout", "TanhWithDropout"), hidden = list(c(5, 5, 5, 5, 5), c(10, 10, 10, 10), c(50, 50, 50), c(100, 100,…