--- name: "vw" section: "1" platform: "linux" package: "vowpal-wabbit" version: "7.3-1.1+b1" release: "Debian 8 (Jessie)" platforms: ["linux"] url: "https://new.carta.tech/vw" --- # vw(1) Vowpal Wabbit -- fast online learning tool ## NAME vw - Vowpal Wabbit -- fast online learning tool ## DESCRIPTION ### VW options: - **-h** \[ **--help** \] Look here: http://hunch.net/~vw/ and click on Tutorial. - **--active\_learning** active learning mode - **--active\_simulation** active learning simulation mode - **--active\_mellowness** arg active learning mellowness parameter c\_0. Default 8 - **--binary** report loss as binary classification on **-1**,1 - **--autolink** arg create link function with polynomial d - **--sgd** use regular stochastic gradient descent update. - **--adaptive** use adaptive, individual learning rates. - **--invariant** use safe/importance aware updates. - **--normalized** use per feature normalized updates - **--exact\_adaptive\_norm** use current default invariant normalized adaptive update rule - **-a** \[ **--audit** \] print weights of features - **-b** \[ **--bit\_precision** \] arg number of bits in the feature table - **--bfgs** use bfgs optimization - **-c** \[ **--cache** \] Use a cache. The default is \.cache - **--cache\_file** arg The location(s) of cache\_file. - **--compressed** use gzip format whenever possible. If a cache file is being created, this option creates a compressed cache file. A mixture of raw-text & compressed inputs are supported with autodetection. - **--no\_stdin** do not default to reading from stdin - **--conjugate\_gradient** use conjugate gradient based optimization - **--csoaa** arg Use one-against-all multiclass learning with \ costs - **--wap** arg Use weighted all-pairs multiclass learning with \ costs - **--csoaa\_ldf** arg Use one-against-all multiclass learning with label dependent features. Specify singleline or multiline. - **--wap\_ldf** arg Use weighted all-pairs multiclass learning with label dependent features. - Specify singleline or multiline. - **--cb** arg Use contextual bandit learning with \ costs - **--l1** arg l\_1 lambda - **--l2** arg l\_2 lambda - **-d** \[ **--data** \] arg Example Set - **--daemon** persistent daemon mode on port 26542 - **--num\_children** arg number of children for persistent daemon mode - **--pid\_file** arg Write pid file in persistent daemon mode - **--decay\_learning\_rate** arg Set Decay factor for learning\_rate between passes - **--input\_feature\_regularizer** arg Per feature regularization input file - **-f** \[ **--final\_regressor** \] arg Final regressor - **--readable\_model** arg Output human-readable final regressor - **--hash** arg how to hash the features. Available options: strings, all - **--hessian\_on** use second derivative in line search - **--version** Version information - **--ignore** arg ignore namespaces beginning with character \ - **--keep** arg keep namespaces beginning with character \ - **-k** \[ **--kill\_cache** \] do not reuse existing cache: create a new one always - **--initial\_weight** arg Set all weights to an initial value of 1. - **-i** \[ **--initial\_regressor** \] arg Initial regressor(s) - **--initial\_pass\_length** arg initial number of examples per pass - **--initial\_t** arg initial t value - **--lda** arg Run lda with \ topics - **--span\_server** arg Location of server for setting up spanning tree - **--min\_prediction** arg Smallest prediction to output - **--max\_prediction** arg Largest prediction to output - **--mem** arg memory in bfgs - **--nn** arg Use sigmoidal feedforward network with \ hidden units - **--noconstant** Don't add a constant feature - **--noop** do no learning - **--oaa** arg Use one-against-all multiclass learning with \ labels - **--ect** arg Use error correcting tournament with \ labels - **--output\_feature\_regularizer\_binary** arg Per feature regularization output file - **--output\_feature\_regularizer\_text** arg Per feature regularization output file, in text - **--port** arg port to listen on - **--power\_t** arg t power value - **-l** \[ **--learning\_rate** \] arg Set Learning Rate - **--passes** arg Number of Training Passes - **--termination** arg Termination threshold - **-p** \[ **--predictions** \] arg File to output predictions to - **-q** \[ **--quadratic** \] arg Create and use quadratic features - **--cubic** arg Create and use cubic features - **--quiet** Don't output diagnostics - **--rank** arg rank for matrix factorization. - **--random\_weights** arg make initial weights random - **--random\_seed** arg seed random number generator - **-r** \[ **--raw\_predictions** \] arg File to output unnormalized predictions to - **--ring\_size** arg size of example ring - **--examples** arg number of examples to parse - **--save\_per\_pass** Save the model after every pass over data - **--save\_resume** save extra state so learning can be resumed later with new data - **--sendto** arg send examples to \ - **--searn** arg use searn, argument=maximum action id - **--searnimp** arg use searn, argument=maximum action id or 0 for LDF - **-t** \[ **--testonly** \] Ignore label information and just test - **--loss\_function** arg (=squared) Specify the loss function to be used, uses squared by default. Currently available ones are squared, classic, hinge, logistic and quantile. - **--quantile\_tau** arg (=0.5) Parameter \\tau associated with Quantile loss. Defaults to 0.5 - **--unique\_id** arg unique id used for cluster parallel jobs - **--total** arg total number of nodes used in cluster parallel job - **--node** arg node number in cluster parallel job - **--sort\_features** turn this on to disregard order in which features have been defined. This will lead to smaller cache sizes - **--ngram** arg Generate N grams - **--skips** arg Generate skips in N grams. This in conjunction with the ngram tag can be used to generate generalized n-skip-k-gram.