5
votes
Accepted
MLOps for beginner
a. For a beginner I would suggest the fullstackdeeplearning course, it's a modern overview of tools and best practices for ML in production. As you can see below, there are a lot of moving pieces.
b. ...
- 196
5
votes
MLOps for beginner
You can do live learning but most models don't require it, because many businesses don't need to learn directly from new input.
Nevertheless, you can apply an automated task every time range (day, ...
- 2,168
1
vote
In Gradient descent, Why the gradient of cost function do not have to be normalized into unit vector
The most obvious reason is that a gradient of the norm of 1 is expected to be at learning_rate * 1 away from the loss function minimum. That is averaged of course.
...
- 186
1
vote
In Gradient descent, Why the gradient of cost function do not have to be normalized into unit vector
In a gradient descent algorithm, the algorithm proceeds by finding a direction along which you can find the optimal solution. The optimal direction turns out to be the gradient. However, since we are ...
- 11
1
vote
Outputs from models (trained on different data) as inputs to another model?
The characteristics you mention are typically called features. My interpretation of your description is that each of the models performs feature engineering. You'll still need to put all the features ...
- 16.5k
1
vote
How to test the confidence for a rule based system?
The confidence interval is the main indicator to test confidence, and it is measurable thanks to the data volume indeed, but also whether or not the endpoints are reliable enough.
Therefore, testing ...
- 2,168
1
vote
Manipulating noise to get some data in right format and apply it to task using PPO
From my understanding of your question, you are looking to implement a learning-to-sort algorithm.
There are current learning-to-sort machine learning solutions that do not require reinforcement ...
- 16.5k
1
vote
How to reduce the size of Bert model(checkpoint/model_state.bin) using pytorch
Are you reusing an existing Bert model or are you training it from scratch?
In all the cases, you can apply several solutions to your model:
Use distill Bert instead
Pruning
Freezing
Convert it to ...
- 2,168
1
vote
How to avoid numerous Hyperparameter tuning in ML?
A good hyperparameter may be considered as a random variable with some variation. I wouldn't worry to much not to find the best parameter for one specific test set. In case you are sure there should ...
1
vote
Accepted
Grid Searching seed in randomized machine learning
It's definitely an error to select an "optimal" random seed.
If performance depends a lot on the random seed, it means that the the model always overfits, i.e. the patterns used by the model ...
- 21.8k
1
vote
Gradient descent implementation of logistic regression
I think your implementation is correct and the answer provided is just wrong.
Just for reference, the below figure represents the theory / math we are using here to implement Logistic Regression with ...
- 124
Only top scored, non community-wiki answers of a minimum length are eligible