2,993,227 events, 1,466,199 push events, 2,301,598 commit messages, 168,771,070 characters
completely reverted that last pull request.
Fuck you baljeet you fucking pedophile.
Update config.yml
Analyze Phase UdaPeople handles some private information like social security numbers, salary amount, etc. It would be a shame if a package with a known vulnerability left a security hole in our application, giving hackers access to that information! That’s why we should include a job that checks for known vulnerabilities every time we check in new code.
Find the jobs named scan-frontend and scan-backend in the config file. For both jobs, select a Docker image that is compatible with NodeJS. Write code to check for security vulnerabilities in the packages used in the application. Use npm to “audit” the code to check for known package vulnerabilities. Just cd into the directory of front-end and back-end and run the following: npm audit --audit-level=critical
ML3M Rebalancing
Originally, the ML3M was intended to be a little bit stronger. The healing values for the Brute, Burn, Omni, Toxin, and Antirad cells, have been adjusted according to what was initially planned. Base Tier's now heal 10 per shot, for a total of 40 damage healed, up from 20. Second Tier's now heal 20, instead of 10. Third Tier's heal 40, instead of 20.
Omni 1 heals 5's for brute, burn, and toxin, and 30 for oxy, up from half those values. Omni 2, is 10's and 60, and Omni 3 is 20's and 120.
Antirad Cell had its toxin heal bumped up from 2.5 to 5, and its radiation-heal from 150, to 350. I wasn't originally going to touch this one, but the text said "It's 150 because that's equal to five units of arithrazine", and... it wasn't. So, I bumped it up to actually be equal to 5u of Arithrazine's worth of radiation healing.
Stabilizer Cell, CorpseMend, Resistance, Haste, and the size-changing cells, were not touched. The reasons for this are as follows:
- The healing laser was simply not strong enough to ever tend to more than one patient. Healing 20-40 damage per cell just isn't enough, even if you had several clips full of cells, it just wasn't viable.
- As a result of this, it was seldom if ever used. Field Medics rarely if ever took it with them. In nearly every round I've played, it was left exactly where it was spawned, collecting dust.
- I'd like to make it a more proper treatment option. It wouldn't be equal to reagents used in your usual Medbay setting, or surgical repair, but if it's modified to actually fulfill its niche as a rapid, contact-free treatment option, it could be relied on more by Field Medics. There is so much potential for this device as an "in the field" healing tool, and I'd like to see it done well. We have no shortage of people who would like to see non-chemical-based treatment options. We might as well make it useful enough to actually use.
Here are a few issues people brought up in Discord when I floated this idea. "But, Tempest, you're doubling everything? That's a huge change! Won't this result in the ML3M overshadowing other treatment options?" Certainly not. Currently, with tier-2 cells that only heal up to 40 damage, this is actually worse than basic first-aid in terms of raw healing ability, as the cutoff for first aid healing is 50 damage on a limb. First aid and reagents will still be the most convenient, most readily available, and overall most-used manner of treatment for injuries that can be treated using those methods. All this change would do is allow the ML3M to also have its uses, and stop being overshadowed by everything else.
"Okay, but what if it DOES actually become The New Meta and becomes the favored tool for healing?" I wouldn't see it as an issue if Chemistry gets dethroned, in all honesty. They'd both take quite a bit of prep-work to use. You have to make multiples of every chem if you want a solid stock, and you'd need multiples of every cell if you want proper coverage and healing-ability, since even with these changes you're expending 1-3 cells per patient. And, unless you get multiple ML3M's made, only folks who are gonna have one are the CMO, and whoever nabs the spawned one first.
"But wouldn't this result in less departmental cooperation between Medical and Science now that Medical will require fewer cells?" Most likely, no. If anything, I'd think there would be an increase, round-to-round, if this becomes a more viable and wanted tool, more people will ask for cells. Personally, my list of requested cells would stay the same, but I'd be sending that request in more often since I know it'll actually be useful rather than having it just to have it. Tier 2 cells are still going to be desirable, and I'm still probably not going to order Tier 3's because of the cost, but that's just me. It's still going to be incredibly weak at roundstart, and better cells will be that much nicer to obtain.
"But Tempest! If you get the highest tiers of cells, and a lot of them, you could fix anything that doesn't require surgery!" Well, if Science had the time and mats, then sure, you could gather up a large enough collection of cells, and you probably could indeed do this for one or two badly injured patients... ... ...and I could accomplish the same goal if I'm given 20 seconds in a chem-lab. The ML3M is more costly to use, the cells cost resources, have to be recharged after 4 shots, and you only start with a base tier Burn, Brute, and Stabilizer cell. Each shot from a base level burn/brute cell is the equivalent of 1.25u of bicaridine or kelotane. This change would make it 2.5. Base level cells are still going to be next to useless for anything other than the most minor of injuries. Chemistry will still be the dominating powerhouse it always has been, but at least you can use this tool, provided the PR goes through, instead, more often.
"Tempest, this thing just wasn't intended to be used to fully heal people..." A couple people have said this, and to be honest, it's silly. In our medical system, there's very little distinction between what can "heal a little bit" and what can "heal fully", because it all comes down to the fact that everything works off of damage values. If Option A heals 5 damage, and Option B heals 50... Both are capable of "healing fully" if you have enough Option A, it's just a question as to whether or not that's practical. Currently, the ML3M's healing values can't really even be used to even partially heal injuries, unless you have a large number of cells. You'd expend all of your charges on one patient and then you've got a paperweight taking up storage space until you can get to a recharger.
Now, as for issues that I personally see? Biggest one is probably the Toxin cell. If I remlember correctly, the base Toxin cell is the same tech requirement as the Second Tier Brute/Burn cells. This PR doesn't fix that, and I don't really see it as a huge priority since I don't think anyone uses the toxin cells as it is, but yeah. Might need to be fixed at some point if this PR goes through.
And, absolute worst case scenario, let's say that this PR screws everything up, turns the Tether upside down, and now Medbay is dominated by a bunch of blue laser beams flying everywhere like it's the new Star Wars Trilogy (but actually better because those movies kinda set a low bar)... this PR was really easy to do. The only long part about it was typing all of this up. We can just bump the numbers down if this ends up sucking really bad, ezpz.
Thank you for coming to my ted talk.
Merge branch 'master' of https://github.com/Sicario7297/jenkins
Fuck you
Kinda start to work still trouble with Z-buffer and other shit
Update README.md
My "Building AI course" project An attempt to create an AI tool to fight against bias in media articles, social network posts, web pages descriptions....
We all have some unconscious gender biases which we can not notice when we speak or write up. This also happens to journalists, media people or influenceres who inadvertently contribute to perpetuate these prejudices. To break out this harmful chain, the idea is to create an AI tool they can use to revise their texts before posting them in order to identify these unconscious biases.
In our daily life we can find instances of unconscious male chauvinism such as the football coach shouting "stop running like a girl" to those players who are not making the most,or a newspaper article about the last World Economic Forum describing females outfits but not their males partners. Most of times those are unconscious but they are harmful anyways. They are contributing to perpetuate gender inequalty.To help us to avoid those, I'd like to create an AI tool.
It works as a single grammar checker or writing assistant.The app Grammarly (helping non English speakers to write properly in English by checking the grammar of the texts written) is a good reference. It can be added as a new keyboard in your smartphone to make it all even easier. It will be useful for anyone writing up and posting articles, master cases, end of degree works... or even novels. In other words, whoever wanting to be readen.
Internal Affairs (IA) pretends to be an AI tool that uses machine learning to help anyone to avoid unconscious gender biases.
Using Natural Language Processing and neural network for classification will classify the texts in gender balanced or unbalanced, and will propose corrections for the last one ones.
The model will be trained with data sets from accredited authors and organizations such as UN Women, Women Faculty Forum (yale university), or some other relevant institutions in this area in different countries and languages
The main problem is that expanding across languages When it is time to expand to another countries with differente languages, we´ll need to seek a solution on how to utilize our existing model in a new market because the tool has been trained with Spanish or English or Deustch.. data and as such it could not be applied to incoming cases in another different language.
To make it able to work in any language. I've been reading aboputlanguage-agnostic word embeddings and I guess they could be a good way to explore through.
“Building AI course project” by University of Helsinki Reaktor https://somebuddy.com/
Add files via upload
This data was downloaded from the Hass Avocado Board website in May of 2018 & compiled into a single CSV. Here's how the Hass Avocado Board describes the data on their website:
The table below represents weekly 2018 retail scan data for National retail volume (units) and price. Retail scan data comes directly from retailers’ cash registers based on actual retail sales of Hass avocados. Starting in 2013, the table below reflects an expanded, multi-outlet retail data set. Multi-outlet reporting includes an aggregation of the following channels: grocery, mass, club, drug, dollar and military. The Average Price (of avocados) in the table reflects a per unit (per avocado) cost, even when multiple units (avocados) are sold in bags. The Product Lookup codes (PLU’s) in the table are only for Hass avocados. Other varieties of avocados (e.g. greenskins) are not included in this table.
Some relevant columns in the dataset:
Date - The date of the observation AveragePrice - the average price of a single avocado type - conventional or organic year - the year Region - the city or region of the observation Total Volume - Total number of avocados sold 4046 - Total number of avocados with PLU 4046 sold 4225 - Total number of avocados with PLU 4225 sold 4770 - Total number of avocados with PLU 4770 sold
Inspiration /Label
The dataset can be seen in two angles to find the city or region and find the average price .
Task: Classification /Regression
Do both tasks in the same .ipynb file and submit at [email protected].
Note: This original dataset has 2000 rows. Here we reduce it to 1500 rows. In the dataset, you may find blank values after 1510 rows. Hence, you may delete all blank rows in your CSV files explicitly.
Data Source:- https://github.com/dsrscientist/Data-Science-ML-Capstone-Projects
Hopefully this push works or else i will tearout my hair. I have trying to push this to github since last nyt fuck you git
wezterm: improve shaping of emoji
This is one of those massive time sinks that I almost regret... As part of recent changes to dust-off the allsorts shaper, I noticed that the harfbuzz shaper wasn't shaping as well as the allsorts one.
This commit:
-
Adds emoji-test.txt, a text file you can
catto see how well the emoji are shaped and rendered. -
Fixes (or at least, improves) the column width calculation for combining sequences such as "deaf man" which was previously calculated at 3 cells in width when it should have just been 2 cells wide, which resulted in a weird "prismatic" effect during rendering where the glyph would be rendered with an extra RHS portion of the glyph across 3 cells.
-
Improved/simplified the clustering logic used to compute fallbacks. Previously we could end up with some wonky/disjoint sequence of undefined glyphs which wouldn't be successfully resolved from a fallback font. We now make a better effort to consolidate runs of undefined glyphs for fallback.
-
For sequences such as "woman with veil: dark skin tone" that occupy a single cell, the shaper may return 3 clusters with 3 glyphs in the case that the font doesn't fully support this grapheme. At render time we'd just take the last glyph from that sequence and render it, resulting in eg: a female symbol in this particular case. It is generally a bit more useful to show the first glyph in the sequence (eg: person with veil) rather than the gender or skin tone, so the renderer now checks for this kind of overlapping sequence and renders only the first glyph from the sequence.