Piero Molino

w4nderlust ·

@desikan95 thank you really much for your availability! Really appreciated.

Regarding data soruces, in the recent v0.3 version of Ludwig we added many new supported formats already, and are still working on adding even more and improve the preprocessing ...

thank you really much for your interest and your willingness to get involved!I believe the best way to get started without the need to get all the nitty gritty details of Ludwig internals is helping out with tests. It's a great way to gradually get acqua...

Hi Aris,

there could be more than one way to approach this problem.In my opnion the most traightforward way would be to split the abstract in sentences and feeding each sentence as input with the label as output.This may end up working fine, but does no...

Hi Charlie,
Posting here the conversation we were having on twitter for future reference: https://twitter.com/w4nderlus7/status/1319371195653840896?s=20
Let me know if yo ustill have doubts or issues!
Cheers,
Piero

Hey Lucas,funny enough I'm the person giving that talk :)And actually COTA was the first model developed with Ludwig, so definitely yes, you can reproduce exactly that architecture in Ludwig.In your model definition you would need to have the inp...

Hi Sutirtha,a couple things: you are getting a few erors when loading the data, so I'm not really sure in which order you are running the cells and what actually is the problem. if train is not none this should work without problems.Also, the output seems ...

Hi Shyam,Ludwig is a toolbox for deep learning models. You can obtain something similar to arima using a specific combiner after a timeseries input feature with a passthrough encoder, but that's not really the intent of the library. If you want to use arim...

Hi Sutirtha,yes you can definitely do time series classification with Ludwig.Assuming your 621 columns are numbers at different times, you'll have to combine them in a single column with space delimiters. Like three columns with values 4, 2, 1 will become ...

Hi, the message means that Ludwig is not available in your environment.
There could be a multitude of reasons for that. one could be that you are starting the jupyter notebook from a different virtual environment from the one where you installed Ludwig.

Pl...

Python 3.8 has only very recently been supported for TF2, TF1 still needs python 3.7.
The next release of Ludwig, due in a little bit, will support TF2 and thus python 3.8 but up until then, you have to use TF1, meaning you need to have python 3.6 or 3.7.

Which version of python are you using?It may be that some packages are not available with some versions of python.Anyway this isn't really a problem with Ludwig, but with TensorFlow.Let's try to figure out if it has to do with the specifics of your install...

Hi Jari,

glad to hear things are working!
I welcome your suggestions, thay are very valuable to me. I will add them as things to do in this board that tracks how to improve the documentation: https://github.com/uber/ludwig/projects/2

For me to understan...

Hi Jari,

a couple things here:

1. I should probably make it clear somewhere in the docs, but some sequence models, because of they architectures, work only on sequences longer than a certain amount. The cnnrnn is one of those, because the first cn...

Please post what you want to achieve, you tried and what errors you got, otherwise it is difficult for anyone to try to help.

That's great news :) Let me know if there are other problems.

As you are using python 3.7, typing is not needed. You should be ok by just removing it from the virtualenv before creating the neuropod.

Hi Peng, this looks like a problem with the typing package.
Check this out: https://github.com/ethereum/eth-abi/issues/131

After looking at the alexa repo you linked, in their case they were explicitly having as a dependency, which is not the case in Lu...

Hi SJF,I’ll try to improve the doc on that. Anyway, the intuition is that the more penalty you add, the more you expect your predictions to be entropic and less sharp. This may be useful when you have noisy labels you don’t trust too much.As per the spec...

you can use the tensorboard for that, just point it to the model/log directory

The JSON file contains all the mappings (both classes and text) that are needed ;)

I think the main problem with this is the fact that at the moment there is a 1:1 mapping between input columns and placeholders in tf, if one adds some process that would map from one input to potentially several inputs, that would require some work.

Anyw...

Got it, will take a closer look then.
Anyway, the preprocessing and postprocessing are still issues, you would still have to perform them with Ludwig or to reimplement them, so I'm not 100% sure it would be a viable easy solution anyway.

Reading the example, this part:

inp = g.get_tensor_by_name("real_A_and_B_images:0")
out = g.get_tensor_by_name("generator/Tanh:0")

sigs[signature_constants.DEFAULT_SERVING_SIGNATURE_DEF_KEY] = \
    tf.saved_model.signature_def_utils.predict_signature_...

Hi Jad, thank you for asking about this. I'm not familiar with that interface to be honest, i have to do some digging myself to figura out what one should do in order to make make it work.My concern by the way is always the same: preprocessing and postproc...

Ludwig already vectorizes your text and looks up embeddings and does everything for you. You can specify pretrained embeddings too. Please look at the sequential encoders for more details.