2 replies
2 replies
2 replies
VWMultiClassifier.fit only supports a 1-d y
5 replies
4 replies
Link - VowpalWabbit/vowpal_wabbit#3352
I would like to work on this issue and it would be great if anyone can point to a direction and some files.
Thanks :)
Hello everyone. I have problems with convergence for --cv_explore_adf on simple synthetic dataset of 3 classes and 5 features, all informative (the data was simulated via scikit-learn's make_classification)
Here's a gist and an output plot above. Basically, it doesnt learn the difference b/w classes and oerforms no better than baseline (predict one class each time)/ tried w/ softmax exploration w/ different lambdas (since it's not bound by epsilon-soft convergence bound like e-greedy policy), same results
FYI, sklearn's softmax multiclass classifier gets 80% mean accuracy. So I really don't understand what might be the problem with VW (maybe smaller learning rate needed?). The documantation and the lack of --audit support in Python certainly don't help.
https://gist.github.com/ColdTeapot273K/76a91a0416cb9b8ecb114e625f88f4a0
4 replies
- Does cb_explore’s online cover algorithm support arguments like nounif, epsilon, etc? for cb_explore_adf, I know it's supported as shown here https://github.com/VowpalWabbit/vowpal_wabbit/blob/1c5bbb3d3726d72fd8de8f22a5fa8590768408bd/vowpalwabbit/cb_explore_adf_cover.cc#L54-L56
- Is there a way to change the exploration algorithm and its hyperparameter after a cb_explore model is created? I think adding a different command line arg related to exploration algo and its hyperparameter throws an error that starts with Disagreeing option values for
6 replies
7 replies
Good first issue label and se if anything interests you
19 replies
may you help understand why I have bug on Windows 10 and vowpalwabbit 8.10.2 or I do something wrong:
this code runs correctly and it takes few seconds
from vowpalwabbit import pyvw
ridge_lambda = 0.001
loss_function = 'squared'
works
VWmodel_simple = pyvw.vw(
learning_rate = 0.1,
l2 = ridge_lambda,
loss_function = loss_function,
)
but this code not running correctly
VWmodel_simple = pyvw.vw(
cache_file = 'my_VW_2.cache',
passes = 3,
learning_rate = 0.1,
l2 = ridge_lambda,
loss_function = loss_function,
)
code run never finished
what is wrong with creating this file 'my_VW_2.cache' ??
command line output is
using l2 regularization = 0.001
Num weight bits = 18
learning rate = 0.1
initial_t = 0
power_t = 0.5
decay_learning_rate = 1
creating cache_file = my_VW_2.cache
Reading datafile =
num sources = 1
Enabled reductions: gd, scorer
average since example example current current current
loss last counter weight label predict features
2 replies
HI. I am using https://vowpalwabbit.org/tutorials/cb_simulation.html. I need to save the model and use it to later as the reward(feedback comes to the system few hours later). vw1 = pyvw.vw("--cb_explore_adf -q UA --quiet --epsilon 0.2 save_resume=True")
num_iterations = 5000
ctr = run_simulation(vw1, num_iterations, users, times_of_day, actions, get_cost)
vw2 = pyvw.vw("--cb_explore_adf -q UA --quiet --epsilon 0.8 i=data/saved_model.model").
1 reply
Hello, can anyone tell how to make cb_explore_adf agent respond to requests in daemon mode properly? i send the multiline commands via echo ... | netcat ... like in documentation and get no response.
If i launch w/ --audit flag i recieve a bunch of info with unintuitive formatting (see attachment). i assume the 1st value in each line is action probability, and the very last line is some combined gradients or whatnot. Very different from a pmf output, like in python example on website.
Sorry, couldn't attach the img to my prev message thread
A problem:
for some reasons i get n+1 size of pmf for data with n distinct action.
Details:
When i do training on cb_explore_adf for datapoints only with 3 actions (no features apart from shared|...) and supply one of these examples for testing (action:cost:proba removed obviously), i get 4 actions in the output file. Why it might be?
5 replies
I trained the contextual bandit as vw1 = pyvw.vw("-d data/cb_load.dat --cb_explore_adf -q UA -P 1 --invert_hash mymodel.inverted") on https://github.com/VowpalWabbit/vowpal_wabbit/blob/master/test/train-sets/cb_load.dat with the --inverted-hash and I got the mymodel.inverted
I could understand the user and action features. What does 18107:0.137426 . means in "User^time_of_day=afternoonAction^article=politics:18107:0.137426" ?
I think 18107 is the hash value for "User^time_of_day=afternoonAction^article=politics" and 0.137426 is weight. I don't know if this is correct?
How can I get the probability corresponding to the user and action features from the weights?
Version 8.11.0
Id
Min label:-1
Max label:0
bits:18
lda:0
0 ngram:
0 skip:
options: --cb_adf --cb_explore_adf --cb_type mtr --csoaa_ldf multiline --csoaa_rank --quadratic UA
Checksum: 2033437909
event_sum 113
action_sum 791
:0
User^time_of_day=afternoonAction^article=politics:18107:0.137426
User^user=Tom:32581:-0.0636371
User^user=TomAction^article=politics:38087:-0.0636749
Action^article=politics:52568:-0.110663
User^time_of_day=morningAction^article=music:58967:0.224528
User^user=AnnaAction^article=politics:62875:0.0165196
User^time_of_day=afternoon:65137:-0.0253498
Action^article=music:67569:-0.0505464
User^time_of_day=afternoonAction^article=food:67793:0.121444
User^time_of_day=morningAction^article=politics:77054:-0.192732
User^user=AnnaAction^article=music:81714:0.297336
Action^article=sports:86811:0.0540273
User^user=TomAction^article=music:89710:-0.101787
User^user=AnnaAction^article=sports:93144:0.0540273
Action^article=food:99122:0.121444
User^time_of_day=afternoonAction^article=music:101394:-0.190187
User^user=AnnaAction^article=food:113649:0.0457554
Constant:116060:-0.0947514
User^time_of_day=afternoonAction^article=sports:121080:0.0540273
User^user=TomAction^article=food:121517:0.109427
Action^article=camping:134640:0.0742112
User^user=Anna:141841:0.0966574
User^time_of_day=afternoonAction^article=health:144204:0.0344906
User^user=TomAction^article=camping:152687:0.0742112
User^user=AnnaAction^article=health:163948:0.0344906
Action^article=health:178351:0.0971796
User^user=TomAction^article=health:188720:0.09161
User^time_of_day=morningAction^article=health:219401:0.09161
User^time_of_day=morning:243586:0.0320462
User^time_of_day=morning*Action^article=camping:257110:0.0742112
2 replies
Hi All,Is there any way to add importance weight to off line training of the contextual bandits? Similar to linear regression where we specify importance weight as 2 in the training example " 1 2 second_house | price:.18 sqft:.15 age:.35 1976".
This will help in reducing the training time of contextual bandits as the training data points are in billions. But we get good reduction if we use importance weight as data points are repeated.
6 replies
3 replies
--lambda or --epsilon value and then load with another?
10 replies
1 reply
Hello all,
I'm NG Sai, final year UG @ IIIT Sri City. I got to know about Microsoft RLOS programme via LinkedIn. My experience with open source includes contributing to C++ ML libraries such as: shogun, tensorflow-lite support and, mlpack where I've done GSoC'21 and currently serve as a member. My github.
I came across the Safe Contextual Bandits. Is this topic taken for this summer or will it be available for the year 2022. My forte is implementing algorithms from research papers so I wanted to inquire about this.
Thanks in advance!
1 reply
1 reply
1 reply
vw -h) but didn't find and it's neither in the documentation. I'm interesting in knowing the values for gamma_scale (I believe I saw in the presentation that it's set to 1000 but would be good to confirm) and gamma_exponent.Using
vw.get_arguments() on the model also doesn't show the default values.Thanks
4 replies