Use (inv)logcdf in tests + update CI

[devmotion]

Alternative to #11. Let's see if it is sufficient to perform computations in log space.

[davibarreira]

If you don't want to sample from mu, another alternative is to have a larger value for the variance in order to guarantee that the sampled point is not too far way.

[davibarreira]

Perhaps try using ν = Normal(randn(), rand()+1) and the same for mu.

[devmotion]

Sure, let's try it.

[devmotion]

Seems to work 🎉

[coveralls]

Pull Request Test Coverage Report for Build 1396553910

• 0 of 0 changed or added relevant lines in 0 files are covered.
• No unchanged relevant lines lost coverage.
• Overall coverage increased (+3.7%) to 96.835%

---

Totals
Change from base Build 1199181442:	3.7%
Covered Lines:	153
Relevant Lines:	158

---

💛 - Coveralls

[devmotion]

Any additional comments or suggestions @davibarreira?

[davibarreira]

Any additional comments or suggestions @davibarreira?

Yeah, I forgot to write in the review, but I think we should sample more than one point. Like:

x = randn(10)
γ.(x) \approx invlogcdf(ν, logcdf(μ, x))

[devmotion]

Done.

[davibarreira]

All good.

[coveralls]

Pull Request Test Coverage Report for Build 1396494963

Warning: This coverage report may be inaccurate.

This pull request's base commit is no longer the HEAD commit of its target branch. This means it includes changes from outside the original pull request, including, potentially, unrelated coverage changes.

• For more information on this, see Tracking coverage changes with pull request builds.
• To avoid this issue with future PRs, see these Recommended CI Configurations.
• For a quick fix, rebase this PR at GitHub. Your next report should be accurate.

Details

• 0 of 0 changed or added relevant lines in 0 files are covered.
• No unchanged relevant lines lost coverage.
• Overall coverage increased (+3.7%) to 96.835%

---

Totals
Change from base Build 1199181442:	3.7%
Covered Lines:	153
Relevant Lines:	158

---

💛 - Coveralls