Documentation
Search…
W&B for Julia
For those running machine learning experiments in the Julia programming language, a community contributor has created an unofficial set of Julia bindings called wandb.jl that you can use.

Example

You can find examples in the documentation on the wandb.jl repository. Their "Getting Started" example is here:
1
using Wandb, Dates, Logging
2
3
# Start a new run, tracking hyperparameters in config
4
lg = WandbLogger(project = "Wandb.jl",
5
name = "wandbjl-demo-$(now())",
6
config = Dict("learning_rate" => 0.01,
7
"dropout" => 0.2,
8
"architecture" => "CNN",
9
"dataset" => "CIFAR-100"))
10
11
# Use LoggingExtras.jl to log to multiple loggers together
12
global_logger(lg)
13
14
# Simulating the training or evaluation loop
15
for x ∈ 1:50
16
acc = log(1 + x + rand() * get_config(lg, "learning_rate") + rand() + get_config(lg, "dropout"))
17
loss = 10 - log(1 + x + rand() + x * get_config(lg, "learning_rate") + rand() + get_config(lg, "dropout"))
18
# Log metrics from your script to W&B
19
@info "metrics" accuracy=acc loss=loss
20
end
21
22
# Finish the run
23
close(lg)
Copied!
Copy link
Contents
Example