Questions in Science & Mathematics | VIEWS | POSTED | BOOKMARK | |
---|---|---|---|---|

How do I coalesce data from various sources? | Today | Bookmark It | ||

How do I setup an environment where I can coalesce data from various sources, like mysql databases and csv files, on my macbook? Is there some opensource tool like informatica? or... ?
This question has been viewed |
||||

ML models: average of all versus average of averages? | Today | Bookmark It | ||

I have data for 20 different people and am training a model (e.g. a neural network with the same hyperparameters) on the data from each person; so this gives me 20 models.
I chose to use RMSE to assess the performance. However, since the training data is shuffled, the computed RMSE is nondeterministic and so oscillates. So I thought running each model 10 times and averaging the results, i.e. the RMSE's, would give me a better estimate of performance. But this is for a single person/model. How do I combine the performance of everything, i.e.
This question has been viewed |
||||

Tensorflow: can not convert float into a tensor? | Today | Bookmark It | ||

I'm working on a multi-digit image recognition project, but I'm stumbling across a certain error no matter what I do.
Here's a sample of my code:
graph = tf. Graph()
with graph. as_default():
x = tf. placeholder(tf. float32, shape = (None, 66, 66, 1), name = 'x')
y = tf. placeholder(tf. int64, shape = (None, 5), name = 'y')
keep_prob = tf. placeholder(tf. float32, name = 'keep_prob')
...
with tf. Session(graph = graph) as sess:
sess. run(tf. global_variables_initializer())
for step in range(num_steps):
offset = (step * batch_size) % (train_labels.
This question has been viewed |
||||

What is a large enough sample size? | Today | Bookmark It | ||

I have 600 customers per week. How many do I need to survey to get a good sample to work with?
I'm very new to stats.
This question has been viewed |
||||

Categorical, nominal or continuos variable? | Today | Bookmark It | ||

I'm trying to understand what level of measurement is best for describing 'number of rooms in a flat' feature.
First of all, I think it's not a continuous feature because rational values like for example 1.42 doesn't make sense.
To decide whether feature is categorical or nominal we should try to find a meaning of ordering between values. And here is my question... should we look for an order with respect to the response feature( in my case 'Price of a property')?
We can say '1 room flat' is cheaper than '2 rooms flat' and so on. But that is not always a true. In general maybe it's true but there are cases when '1 room flat' in the city centre is way more expensive than farther one.
This question has been viewed |
||||

Any “rules of thumb” on number of features versus number of instances? (small data sets) | Today | Bookmark It | ||

I am wondering, if there are any heuristics on number of features versus number of observations. Obviously, if a number of features is equal to the number of observations, the model will overfit. By using sparse methods (LASSO, elastic net) we can remove several features to reduce the model.
My question is (theoretically): before we use metrics to assess the model selection are there any empirical observations which relate the optimal number of features to the number of observations?
For example: for a binary classification problem with 20 instances in each class, is there any upper limit on the number of features to use?
This question has been viewed |
||||

is there any mathetmatical formula, paper or anything that shows the probabilty of a meteor hitting a particular spot of moon's surface? | Today | Bookmark It | ||

According to this article http://news. nationalgeographic. com/2016/02/160209-meteorite-death-india-probability-odds/ a science professor at Tulane University made a paper to calculate the probability of a person being hit by a meteor on Earth . And also an astronomer made a calculation (not mentioning the procedure) with a similar figure. My question is, since the moon has no atmosphere and more meteors reach the surface, is there any mathematical formula, paper, etc. to calculate how likely a spot in moon's surface it's likely to be hit by a meteor?
This question has been viewed |
||||

Is the bright red body a star? | Today | Bookmark It | ||

I was helping classify galaxies at Galaxy Zoo when I came across this picture of a galaxy. The dull red body looks huge compared to the galaxy. Is it because it is closer to the earth? It would be interesting to know what is going on here.
This question has been viewed |
||||

Strength of gravity during the big bang? | Today | Bookmark It | ||

I was watching a documentary on the big bang, one of the astronomers said that initially the four fundamental forces were combined as one, they then emerged to become: The strong and weak nuclear forces, the electromagnetic force and the gravitational force. He claimed that the first force to emerge was the gravitational force, he went on to say that had the gravitational force been weaker than it is everything would fly apart so no galaxies would form and if gravity was too large than we would end up with black holes everywhere, so the force of gravity needed to be just right, as it is now.
This question has been viewed |
||||

How should I find a telescope for research time? | Today | Bookmark It | ||

I'm an undergrad in physics and astronomy deciding to look into applying for time on a research telescope, and I'm not sure how to proceed with locating/finding telescopes that I can apply for time on. Specifically, I'm interested in looking at the chemical composition of protostellar disk(s). Any applicable information would be nice.
This question has been viewed |

Get access to full search & filters, create alerts, bookmarks and more!