Token Engineering Commons Lab!, 8 Jan 2021

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: W5 TEC Lab!: Visualizing Economic Templates with Holoviews

Description

This week we review hvplot and panel and use them to create hatch param classes!

🙏 Thank you for watching! Hit 👍 and subscribe 🚩 to support this work

🌱Join the Community🌱
on Discord https://discord.gg/DDr5kYU
or say hello on Telegram http://t.me/CommonsStack
Join the conversation https://forum.tecommons.org/
Follow us on Twitter: http://twitter.com/CommonsStack
Learn more http://tecommons.org/

A

What's up dude, hey jake.

A

uh How are you jake.

B

A

Good nice, how are you, why did you yeah I'm doing real well doing good had a good week yeah, I can tell you, I see you super active.

A

A

Yeah, it was good, I was in georgia and we were just kind of hunkered down with my girlfriend's parents and it was really chill.

B

uh Georgia, like the city wait, is that a state.

A

Or country uh atlanta's the city but uh yeah georgia is the is the state?

A

Were you in atlanta uh close to it's a suburb outside of um atlanta? No one. A lot of people don't actually live in atlanta, very pissed people. A lot of people live in atlanta, but it's a massive like suburb kind of vibe.

A

A

A

Okay, I'll share my screen. We need to talk at some point. I don't know when it's happened.

A

I'll uh I'll put a small agenda on.

B

Okay cool, so I am streaming we're live and recording so we'll give it another couple minutes in case there's any late stragglers.

B

um Anyone else want to give. How was hey sam how's, your how's, your new year's.

C

B

It was nice to just get away from the computer for a.

B

While hi santiago.

C

Hi everyone nice to have you here.

B

Okay, so if you guys want to get uh started, um checking out what we have planned for today, I just dropped the link to the notion workspace. I also pinned it in the channel, so it should be easily accessible from now on. I'm also streaming on on twitch here and I've been streaming quite a bit lately. Whenever I'm hacking on open source stuff, I tend to throw on the stream. It's pretty easy, so keep an eye on that. Follow me on twitch um yeah.

B

So I guess everyone found the notion page, so you can go ahead and open up lab 5 here and one thing to note: oh yeah, let's start this, so let's keep this attendance going. We can use this as a I plan on sort of using this later as a data sheet to do some analysis on, but so I'm just gonna insert a column.

B

Okay and everyone can punch in their favorite emoji, not sure why this one got copied over, but that's fine.

A

Okay, so let's see.

B

Penguin for linux.

B

All right, let's see who can find the spreadsheet first, the lab is meant to be interactive, so I recommend everyone to follow. Follow along. If you can it's? Okay, if you just watch too that's fine.

B

Usually tonga's the first one.

A

B

So pkk, oh it's in general,.

B

um I'm not sure how to send a link to that, but so can you guys uh follow the notion page and the attendance sheet just chime in if you guys ever face any issues or any questions. Anything like that. I just want to see that people are tracking their there's tonga, okay, good, okay, so things are working, okay, so I'll. Let the stragglers fill in their emojis here just want to make people make sure people can track these resources.

B

And when you get a chance, when you're finished with that, oh nice, someone's coming in with the robot santiago, good, okay, so here's the lab 5 visualizing economic templates with hollow views. So I just put a rough schedule here: we're gonna, I'm gonna open up some resources. I'm gonna show you guys the workflow of actually discovering these libraries and how to navigate the documentation and how to run the examples. So that's going to be really helpful.

B

If you want to kind of carry this work forward that we're going to get started on today, it's going to be really important. I'm actually planning this to be a two-part series. We started it a bit before the new years.

B

We were taking a look at the hatch parameters, we're going to look at those again we're going to put them into a param class, which is a really it's a way of parameterizing python classes, a really neat library that we're going to check out that enables us to make visualizations with holo views and put some panels together with widgets and visualizations using the panel library.

B

um So we're going to do a bit of an introduction into that ecosystem and then we're going to dive in we're going to take a look at the new hatch.js parameters and start plugging them into these libraries and just see how that works: the basics, how to get the widgets going, how to get the visualizations going and then we're going to do sort of a free form we're just going to build from there and see how far we can get and answer any questions along the way and if you guys can get your um workflows going open up a terminal.

B

If you want to follow along if you're, just gonna watch, that's fine too, and any questions so far.

A

A

B

Okay, so let's take a look at these libraries, so we have python.

B

Now the name, it's all in the name right. What are we doing here? We're trying to figure out the hatch, params and visualize them? So this seems like a good place to start so um param is a library providing parameters.

B

Python attributes extended to have features such as type and range checking dynamically generated, values, documentation, strings default values, each of which is inherited from parent classes, if not specified.

B

So we can see here we're able to create a python class and it just inherits this param.parameterized class, and then we can give it these param attributes, like numbers, booleans, etc, and we can have bounds on those parameters. We can have default value values and we can have documentation.

B

So this is just good practice in general software engineering to have basically types and bounds. It means anytime. Your system starts to go outside of these ranges or types it'll notify you. It can either crash with an error or it can put out a warning in your logs things like that. So this is a really good practice in software engineering. But it's going to come. It's so useful for data visualization, because once your data is annotated like this, you don't really ever have to think about how to use a visualization library again.

B

It should just all come for free, because nowadays there's an ecosystem of data, visualization libraries that given annotated data, they just know what to do with it and the main library that implements that is called holoviews, and these are all part of an ecosystem.

B

This is actually a really cool kind of story to be told, it's actually made all made by this one guy, this one genius guy who's like a neuroscience researcher um and then a whole is supported by a whole team through anaconda and um the whole sort of pi viz uh ecosystem and it's all open source code. It's a really good example of a well-built, well-documented open source ecosystem, so with param.

B

We can then extend that to this hollow views, data, visualization library and the slogan here, like I said, stop plotting your data, annotate your data and let it visualize itself. So that's what we're going to do with these libraries and with the hatch parameters?

B

Holoviews is a really powerful and extensive visualization library. You can do a ton with it, but it's it has a bit of a learning curve. So on top of that, there's a library called hvplot, which makes it really simple. It's essentially an interface into the hall of use, plotting library, and um it gives a really simple api and we're going to play around with this today. This is what we're going to use we're actually going to go through this. This is cool because this is a jupiter notebook.

B

Actually, all of the documentation here is our jupyter notebooks that are rendered to html and so you'll see what I'm talking about when I, when I say workflow, on learning these libraries we're actually going to clone this repo and work directly on this tutorial, so that's hvplot and now the last library I want you to start to think about that fits into this ecosystem is called panel again, it's made by the same team, and um this is a high level app and dashboarding solution for python and essentially what it lets you do.

B

um If you have these sort of holoviews objects, then they're actually composable. This is one of the cool features of this library. Is, you can add two plots together using the plus operator, which will literally just put them side by side or on top of each other, and you can use the multiplication operator or list the star operator to actually put two plots on top of each other. So if they have the same axes, then it'll just work, no problem um yeah.

B

So the multiplication operator essentially like joins plots by axes and um allows us to easily compose these uh dashboards and and have these widgets and connectivity between our charts.

B

So this is the sort of stack that we're going to use today to visualize our economies that we want to deploy- or in this case the tech hatch.

B

So the workflow here all of these um have a github repo associated with them and we're first going to take a look. I'm going to use this hv plot as the example. So I'm just gonna again, google hp plot and we get the github repo here.

B

Okay- and this is the guy- I think we should all find him on twitter. If he's on twitter, philip he's uh basically started all of these libraries and is the primary uh maintainer.

B

I don't know if he's on twitter either I followed him before or not. I kind of forget.

B

Yes, everyone should follow this guy he's, like a um genius, he's an absolute genius: okay, uh senior software engineer at anaconda neuroinformatics phd creator of panel hall of views, geoviews core developer bokeh and data shader uh he's in berlin, germany.

B

So a little shout out there, phillip he's quite active on github. If you ever have a question, you can just go post issues on these uh on these organizations and he'll usually respond within a day.

B

Okay. So let's try this out now. I want to jump back a little bit. We're going to open up our I've consolidated the lab resources into a single repository, which is called the tech lab and so right now it just has a running. I actually couldn't remember where lab 2 and lab 3 were so that's a little. We can put that as a housekeeping note in the lab organization.

B

So, uh where are lab two and lab three files, uh someone could go watch the recordings they're, probably in some repository that I I just don't recall um so so we got lab one lab four and lab five.

B

Now what I've done here, if you're familiar with git, you can use this get ignore and it will ignore all these files. So it's just nice for development, sometimes there's different artifacts that are created through the development process, but I've added this folder called scratch, so that's going to get ignored by git and so what I do in my local development environment.

B

So I just bring up a terminal here.

B

Okay, so in the repository here now locally, I have this folder called scratch and I can do whatever I want in there and it's not going to pollute the git repository. It's not going to be tracked by git, so you can run all sorts of your local projects and research and experiments and you can mess around in there.

B

You can put any files in there that you want and not have to worry about if they're going to get pushed to get or not so we're going to go into there into that scratch folder and I'm going to clone this hv plot.

B

Sweet um and while I'm at it there's so we're gonna play around with hp plot and also panel.

A

B

Here we go oh yep panel.

B

If anyone's following along, if you get stuck or anything, let me know we can slow down a bit.

B

Come on internet okay! So now we have these two libraries that we're going to use and we'll notice if we activate our virtual environment and open up jupiter lab.

B

Then we can check them out and they follow this standard, which is really useful. So we go into the repository and they always have this examples folder. So this full stack of parameterized classes and visualization. All of the repositories have this in common that if you go into them, they're gonna have this examples directory, and this is actually all the documentation that we were seeing earlier in the browser. So for the hvplot one we can open up the home page and now we have an actual jupyter notebook that corresponds to their documentation.

B

So we could go through this. Let's see high level plotting api for the pi data ecosystem built on holoviews, very cool. This whole ecosystem is very nice. I've seen, for example, block science uses this network x quite a bit to do. Network modeling, x, arrays really come in handy for modeling.

B

N-Dimensional data sets it's a sort of generalization of pandas, um and then we get this nice ecosystem and data shader is for like scalable visualizations, so no matter how large your data is, it allows you to sort of zoom in to one piece of it at a time. So you can visualize infinitely large data sets and then here's some more on the ecosystem, and I guess I recommend everyone to go through this at your own leisure.

B

As you have time, we won't go through all of it now, of course, but let's just check it out a bit so they're importing pandas, they're, creating a sort of date, a date range and then they're.

B

Creating a random data set a thousand points uh by four, so they have a b c d here and they load that up into a data frame and how this works is when you import hv plot it, what it does something called monkey patching, which is a design pattern in software engineering, in which case it actually overwrites the pandas built in plotting function, because I think we could do this.

B

Pandas already have this plot function. So this is the default that you just get with stock pandas and when you import hp plot.pandas it actually monkey patches. Your data frame to now have these hv plot functions, and so we get something like this rather than just a static image. This is actually interactive. We can zoom in and we can pan and we could save this as a png, etc, etc.

B

So we get all this functionality just with a simple import, so there's all sorts of cool stuff here: violin, charts, they're, examining crime data, and then this is where they talk about this composition that I mentioned earlier so using this star plot.

B

So so they superimpose two plots here on top of each other and then they add a plot beside it, which is just the table itself, and then you can even use this streams library to have streaming data frames, and it's you get this streaming visualization with almost no effort at all using this same hp, plot interface, um and then they do some geospatial stuff.

B

I believe this I had to you guys might get stuck here on installing the dependencies and don't worry about it too much car to pi was the one that took some time but anyways. This is the hp plot library check that out. I don't want to get too. I don't want to get behind schedule here. So let's look at panel now same so this is a different repository panel. It's got this examples, we're going to go into user guide and we're going to check out this one called param.

B

All of these are really cool, different ways like components we can. Actually, I didn't look at this before. Let's just see if we could run it really quick, what do we get? Oh yeah, these are just all the different components that we can. So this is where we're going to get our widgets from, and this is cool. But let's take a look at this one called param.

B

uh This is what I want to do today, so this is kind of an upgrade of our last lab we're going to use these parameterized classes. Last week or last month we were just making individual widgets and then seeing what we could do with them, but today we're going to do something much cooler. So this is the way of combining this param class with the panel library that allows us to build dashboards, and so what we do is we parameterize a python class and we define all of its attributes using the param uh syntax.

B

So we you see here what we get access to. We can have just a general parameter, that's like your any type or we can have strings integers floating point numbers ranges, dictionaries, boolean, values, colors dates, data frames, objects, lists files, multiple files and actions, so you could re, get very creative with this and compose it into some complex functionality. So this is really, if you just read one documentation page out of all this.

B

This is the one you're going to want to find to start utilizing these parameterized python classes and here's the example of all the widgets that you can. You can put together, there's even a file selector here, so you can load up files dynamically.

B

So this is good for, like loading data sets and stuff like that, and they go through this. They show you how to they compose like geometric things and they get this map explorer. So some inspiration for what we could do uh in from like a data science approach when it comes to analyzing all of our tech data now and down the road.

B

So that's um kind of a whirlwind introduction: it's really just showing you how to explore these libraries um so start with the documentation and then go to the github repo and find the contributors on twitter and then clone the repos and run the examples, and so now we're going to put it to use. So we've got our lab here. I can probably close some of these for now and see. Oh nice blair, yay, okay, so so, let's jump out of this documentation.

B

Okay, so I need to go back here and I was in the scratch. So I'm just going to go up a directory into the lab and I'm going to run jupyter lab from.

B

Here: okay, so let's take a look at lab five um yeah and we have access to our documentation, so we can open up any of those examples that we wanted maybe I'll leave. The panel example open.

B

So, let's take a look at lab.

B

5., so I did some impact hours modeling. We might bring that in with what we're doing, but let's start out here with the hatch parameters. So this is what we had. Last month, we had copied in parameters from newdow.js and maybe I'll give a shout out to griff and sam. Is there any particular changes that we should be mindful of moving from that newdow.js to newhatch.js, because I believe these were copied from new dao.

A

The hatch, the impact hours for sure, that's that wasn't in there what else wasn't in well new doubt the big new dow js is totally different. It has bonding curve, it has everything right. So we you cut out all the conviction, voting stuff, you cut out all the bonding curve stuff and you include the impact hour stuff and a few other little pieces. Okay,.

B

So we probably want to just should I just blast these and start from scratch with.

A

I think so: okay.

C

The thing is that last time we we kind of erased already the things that we didn't need, but it's okay! Let's, let's do that again! Okay,.

B

So now we have new hatch uh yep. So let's um just stop me on every section: that's okay! I know we did this last time. It's too bad, we're doing it again here, but okay, I'm just going to start fresh. Do we need this hatch oracle ratio.

C

Yeah, because this is the cstk radio, okay, you had you had those already on top okay, so it's different, because there is a contribution.

C

There is a new variable there. Maybe it it wasn't needed, but the last one in that specific block contributions, proceeded per transaction yeah that one we don't need it actually. No, no all right.

B

So we have the hatch oracle ratio. Let's start here right with the dandelion voting settings.

B

Okay: this is perfect. uh This this code's actually cleaner than the new dow as well.

B

So that's good. Okay,.

A

B

Okay, so let's actually just start with these.

B

Hey mishka, okay,.

B

Okay, so what we're gonna do now we're going to create a class which is a.

B

Dandelion voting and it's going to be a pram dot. Parameterized.

B

And we'll just go through these, so let's think about support required. Okay, so we're going to use this, so it's going to be a number param dot number and um the default's gonna be 0.6.

B

Now what are the round the bounds of this, and we can make these constants as well for any of these, uh some of them might make sense to be constants. um I guess this could be from.

A

Zero point one to one: zero to one yeah.

C

Yeah it got it probably zero that one um or even less 0.01- I don't know if we can add us a strict widow, yeah.

B

Yeah there is a way to do it. um Let's come back to that in a bit, so I think this is good. um We can set a step, I believe, to be like 0.1 or no 0.01.

B

um Okay. So, let's see if that works.

A

B

We can instantiate it, so we can say: dandylion equals dandelion voting and.

B

Okay, so now we have something.

B

Okay, so let's go: let's: let's keep rolling on these, it's nice to actually type them out, because it forces us to read them and think about them.

B

Min acceptance quorum equals param dot number, uh so the default 0.02 bounds bounds equals um so, let's think zero. Okay, minimum acceptance.

C

Yeah, I I'm not sure if there is a relation between super required and minimal acceptance. Maybe you need minimal, minimal acceptance to be lower than required. I am going to check out in the code. Meanwhile, you continue if you want yeah one day, but let's see if there is any other restriction.

B

Okay: let's, let's try that for now.

B

Actually so duration blocks has to be greater than zero. That's for sure.

B

um Vote buffer blocks.

B

So default to eight- and this is ours-.

B

B

B

Vote execution delay blocks equals okay same. Oh, these are actually integers.

B

So let's make this an integer and make this an integer and param dot integer and this defaults to 24.

B

B

And okay tool gate fee set the fee paid to the org to create an administrative.

B

C

Effectively, the minimum acceptance quorum must be less or equal to support required.

B

Okay, so I'm gonna, I'm gonna, we'll think about that in a bit we'll see if that we can uh impose that constraint, um because I know last time we tried that where we put in the minimum bound to actually be the value of this support required and uh I'm pretty sure, that's possible to do, but I hadn't uh looked into it- haven't looked into it yet it might be a homework assignment.

B

So let's say tolgay fee.

B

Okay, let's just see what this looks.

C

Like it doesn't have to be an integer, it could be decimal because it's you know like.

C

It could be 0.5 if we.

A

A

B

So it looks like it gives us these sort of selector widgets on the integers. um I know you can customize the integers, so we could look how to do that. But this is our our base. uh Dandelion voting template, so we've got a support uh required 0.6 minimum acceptance, quorum, 2 uh vote duration blocks is three uh vote. Buffer blocks is eight vote, execution delay blocks is 24 and the toll gate fee is three good, so we can do a dandelion voting. um Let's keep going. I guess. Let's get the hatch settings.

A

C

They will, at least this is going to be very, very equal to what we did last time, and I think that it's better to recover what we did well.

B

We were doing it in a different way. We actually did it a different way. We were building independent. um So what do I have here? So we have some here. We actually have a template. We can work with yep good point, so let's call this the hatch and so how many collateral tokens so hatch minimum goal. Oh yeah, we have that uh set here and here it's set to five and um okay. So the bounds, let's say a minimum of so this is for the test. Hatch. I guess that's why it's.

A

C

Okay, yeah: it's that's right.

B

Okay, so and then a target goal uh max school should it be max goal or target goal.

C

That's interesting and right now what we have is a max goal in which, if we reach that the hatch ends and it ends, but we are also talking about having a kind of target goal or something like that. So in the front end it isn't is also going to be.

C

It may be something but not for the smart contract.

B

Okay, let's call it max goal for now: that's what we have.

B

So max goal is set to a thousand and uh okay. Let's see.

B

Okay, so next is the hatch exchange.

B

B

B

So that is what should this be oak times fundraising token. So uh fundraising, one token: what is.

A

This value, actually, actually I'm really glad you're asking some this, because when I look at that, like I don't see the right results in the app when it happens, you know seems like you should get 10 million tokens, but instead you only get 10 000 or something for one raptex die.

A

What's up with that, sam.

C

Let me let me check it out and the hatch.

A

Exchange rate yeah.

C

I actually made a forum.

A

Post about this, that looks that will show you where it's.

C

A

Jake, I think jake made it. I don't remember who actually authored it, but I reviewed it and it's weird.

C

Yeah, that's good, and I think that we already kind of solve it, because in a previous version- and the thing is that you have to multiply by fundraising one token and divide it by the it's- it's speaker of the decimals that it's talking, ha I I'm not sure. I I think that we have.

C

We understood that uh and documented it well, but probably now that changes has been lost in the transition to the new hatch and script.

C

A

C

Response right now.

A

Basically, I put it for the forum post in the in the te labs and basically the hatch rate when it's set to uh three six. Oh man, I can't tell percentages like seven zeros and a one point.

A

Basically one satoshi.

A

When it's set to one satoshi, uh then one wrapped x, die, gives you 10 000 test tc, which is just really weird. I don't know.

C

Yeah yeah we'll have to check it out.

A

At least you can write that in the comments uh you can, if you want ygg, because I I had to test this manually- and I just I assumed that this new hatch.js was the thing that was used to deploy right like it didn't get changed.

C

We have to check it out for, for the for the moment we can, we can use it as it is and and then come back and fix it.

B

Okay, so we're getting into a new another example here, so um any sort of dependencies where we have dependent um variables, they can be encoded in this way or really any function. This is how we're going to do our visualizations and stuff later too, as we tweak these parameters.

B

So we use this param decorator, param dot depends because I see I'm getting towards the vesting cliff period here and it's actually a function of the hatch period. So I'm going to do that, let's see def. um So I would call this update vesting cliff period and.

C

Wait a second, the best thing cliff and the best incomplete period are, are not going to be parameters that we are going to touch in the in the hatch because they are going to be applied in the in the new down and they are here because we need to best until the hatch is ended.

C

So the token cannot be moved, but they are not. Actually, they are not actual parameters of the hatch, so yeah.

B

Okay, so we just okay they're, just part of they're needed for the smart contract, but they really.

C

B

Not part of the hatch, okay, okay, good to know um vesting cliff period, vesting this one as well, and then what percentage of hatch contributions, hatch percent supply offered equals fundraising, 100 percent, uh what percentage of hatch contributions should go to the funding pool and therefore be non-refundable?

B

So we have this hatch percent funding for beneficiary uh yeah.

C

Yeah yeah, the third one, which is one hundred percent, is it's not even a parameter because it it will mean pc tokens for the funding pool, and we don't want that.

C

And so it's 100 and it's set up as it is. Maybe we have to remove that the parameter, because.

B

This one is the hatch tribute.

C

Yeah yeah: let's.

B

Call that hatch.

C

Which splits the dice into the hunting pool and the resort.

A

I sort of think that it's more like.

A

Because the hatch tribute it is going to probably be more than it will be determined. The final hatch tribute will be: maybe it's like initial hatch tribute or um forced donation, or um you know, because there will be another hatch tribute which will actually split between the bonding curve and the funding.

C

No, no, that's the one. Maybe you are talking about the culture. The cultural attribute.

A

No, no! No, but when I mean when, when the dow upgrades to a commons, it will be able to change this, it will be able to add more to the funding pool if it wanted to.

C

If we do the split at the at this point, when we, when we do the hatch, we finish, the hatch, the dice are already splitted and the natural thing is to move the fans as they are.

C

The dandelion boating would be able to move the front inside of the balls so applying a different height to it. But it's not the expected outcome. The expected outcome is that the migration roses from that result to that reserve and from that funding pool to that funding.

A

I would say I don't think it makes sense to determine the hatch tribute without understanding the rest of the bonding curve and the in the conviction, voting parameters.

A

It would be it's like incomplete and unnecessarily restrictive like when, and I think what will happen is like the this hatch tribute whatever it is, it'll probably stay small or even be zero, potentially, and it will say, like all of that, will go to the funding pool, but then of the hard, larger vault one transaction would go to the funding pool and one transaction would observe.

C

uh I I don't know if, if that's we discussed it in in legal uh two or three weeks ago, if the first donation is important or not, and that affects that that discussion.

C

I think that it's, we should have the discussion in another moment, but that would affect how we do the the hatch and if we just need one bolt or two, which could be an interesting discussion.

C

But I I don't want to to take time from ycg and we can just.

A

I was in that discussion. I was pretty sure that the idea was oh, we might be setting this to zero.

A

It was part of the options right so like the initial hatch tribute might be zero and then, but um but anyway yeah we can keep moving.

B

Good discussion guys, this is a big. I think. That's awesome, that's uh great, that it's coming out of the labs, because sometimes it takes this sort of visual stimulation to just think. Oh yeah, that you know I thought this was this or I thought this was this, so I'm actually stoked that this is happening um but anyways. So we have the hatch here we have the dandelion voting um and then- and that's that's really- I realize that's the end of the params here.

C

um Other than I know we yeah.

A

B

B

Okay, so let's make.

B

B

Max impact hour rate equals ram dot number.

B

B

So max impact our rate max theory: oh, is this r from the equation.

C

Yes, it is, it could be, no, it could be much more. It could be like that everyone.

C

It was the maximum that somebody could get in die for for the job in one hour,.

B

This is awesome, we'll see if we can. uh I think we can visualize that equation in 12 minutes. Hopefully I think we can do it. Okay, 0.012 um times 1 token divided by 10 times, okay, okay, so this should do it so.

B

Okay, so now, let's see if we can so we have the data here, we can come back to that, but let's just see if we can move towards rendering the equation. I have it here as a.

B

Resource oh impact hours, this one.

C

A

Awesome: okay,.

A

A

B

Okay, so let's think here, let's just plot this first, so we have, we should have. Let's say: if we go, I equals impact hours.

B

Then we have I dot max impact hour rate and we have I dot expected impact hour rate, so we can use those in this equation.

B

B

X, greater than zero okay, I think we can do this so similar as we did last week, so we can import numpy um as np numpy.linspace. Then we can create our x axes so from zero to let's see what is uh x.

C

Millions yeah and eggs are totally raised. Total fans raised okay.

B

So let's go from oh, it should be. We could use our hatch, so we could say h equals our hatch and then we.

A

B

Like h, dot min goal and h dot dot max goal. Okay, so we can go from h, dot, mingle to h, dot max goal. Okay, so this is our x axis.

B

Now we want to apply our equation, which is essentially this one so x over x, plus m h. So let's just define r m and h in here so um r is actually equal to um our impact, might be something like that m and h. um So what's this expected raise? Oh that's h. I think right or.

C

C

B

So then we can say y equals r.

B

A

A

Oh actually, we need that say: x, equals.

B

X, okay, so then we should be able to make a data frame, so data frame equals pandas.dataframe, um maybe xy, something like this: um okay, zip.

C

Yeah another one.

B

Transpose okay, so now we get our hp plot, probably just like that, um uh that's not quite right! We need columns here, uh columns equals x y. For now we could probably give it better names. Let's see.

A

Oh uh df, dot columns equals.

B

A

X, equals x.

C

Wow yeah awesome.

B

So we have six minutes good. I think we're ahead on time, so I think we should be able to make this interactive.

B

So, let's see just gotta put this all together now, so this is just an import put this up here.

A

B

So we have these okay, so let's take.

A

C

The other one, why is that expected? Probably it serves yourself right.

A

A

B

All right, let's give that a try.

A

um Okay, so we need to go.

A

I equals impact.

B

B

Oh, I have this uh this. Oh it's this.

A

B

So now we can play with these. So that's that's good. We got to a good point here. um Maybe I'll make this a.

C

Yeah, it could be also slider yeah.

B

Expected raise per impact hours. What do you think.

C

It could be 0 to 100. You know.

C

Okay, it's too much!

C

Actually, maybe it's because the max in well! It depends. It depends, then, but with the current one, it's too much because it it flat so so fast. You know.

B

Okay, awesome, so this is a good start. I think we're. Oh my clock's glitching out. That's weird: do you guys.

A

C

B

So I have no idea what time it is, but I feel like this. This is a good session. I this is actually a really good place to be. I'm gonna push all of these updates everything. I've got here, I'm gonna push it to github and it'll be all accessible through the central repository uh which you can find here and under lab five and uh yeah. I do recommend people try out that little, the little git hack where um the scratch directory is get ignored, so you can do any sorts of experiments in there.

B

You can clone any sort of repositories and you'll see that in my jupyter lab I have it all available here, so I have all the labs and then I have my scratch and I have my references. I can go into the plot examples and check out all of the network x, for example, and.

C

B

I want to continue this next week, we'll pick it up where we left off I'll. I think I'll make some progress on this actually throughout the week, and I hope that everyone else does as well. These notebooks are good to run. You shouldn't have any problem running them. I've got a requirements file in the repository.

B

So let me just show that so there's a instructions here on how to create a virtual environment and install all the requirements, and then you should be able to just run jupyter lab on your system and message me or post in the tech labs channel. If you have any difficulties with that and by the end of next week, we should have a nice sort of hatch template that we can, by the end of next week's lab.

B

We should be all cloning it and making our own copies of the files, and then we can bring that to the whole tech community and get everyone involved. At that point.

C

It has been super cool to to check out the holoviews.

B

Oh, it's awesome, it's a whole rabbit hole. I I this stuff is so great. It's a it's a good community. It's actually a good model of of open source developers, but they have no idea of about tokenomics and cryptocurrency, so, like bridging those worlds is, is huge. I think.

B

Anyone know the time.

C

It is, it is 59 cool.

B

Any anything else, any anyone have questions or anything.

C

No that maybe you said that we follow you on twitter, but you didn't, or I didn't manage, to see the twitter. So, okay, okay,.

A

um I'll just already.

B

A

Igg underscore anderson on twitter yeah on twitch also also on twitch.

B

And that that is pinned in the lab channel.

C

B

Okay, well thanks all for coming out. I really appreciate it it's nice to see so many faces in here. I hope you guys learned some good stuff and I'll see everyone next week and stay active in the tech labs. uh I do check that channel so I'll try to answer any questions that are in there.

A

For the session thanks, everyone yeah thanks sean yep.

C

B