samzer · HackerTrans

samzer

no profile record

Submissions

You shouldn't skimp on your design phase?

blog.saitamasolutions.com

2 pointsby samzervor 6 Jahren0 comments

Dark Series: Network Graph of the Family Tree

dark.saitama.solutions

2 pointsby samzervor 6 Jahren1 comments

Answering Monty Hall Problem with Monte Carlo

towardsdatascience.com

1 pointsby samzervor 7 Jahren0 comments

Ask HN: Any open source drag drop tool for tensorflow/pytorch?

4 pointsby samzervor 7 Jahren2 comments

Show HN: Open-source experiment tracking for machine and deep learning

github.com

4 pointsby samzervor 7 Jahren5 comments

Show HN: ModelChimp – Experiment tracking for machine and deep learning

github.com

4 pointsby samzervor 8 Jahren0 comments

Show HN: ModelChimp, Track and Collaborate on Machine Learning Models

modelchimp.com

6 pointsby samzervor 8 Jahren2 comments

How an AI project is helping to make Brazilian politicians more accountable?

samirmadhavan.com

2 pointsby samzervor 8 Jahren0 comments

Data Hacking through Hollywood

blog.samirmadhavan.com

10 pointsby samzervor 12 Jahren0 comments

Data Hacking through Hollywood

6 pointsby samzervor 13 Jahren0 comments

comments

letztes Jahrdiscuss

There is definitely hype around AI but there is also value in it.

The realistic scenario for a well functioning organization is not to replace their entire DevOps team with AI agents and then realize the hard reality that their entire tech infra is on fire but to actually use these as a tool that makes their best DevOps person more efficient and make their DevOps resources much more leaner. This increases the productivity for the same cost. Even though this sounds grim for the future of work, this is what I feel is going to happen.

vor 5 Jahrendiscuss

Has anyone here, made anything with the public beta api?

vor 5 Jahrendiscuss

This holds true for ML models too when the data is large. I've seen people(myself included) when the data is small, the iteration is faster when selecting the features but when the data is large then you have to think about the features that you want to select a bit more deeply, test it out on a sample and then submit it to the larger data set.

vor 5 Jahrendiscuss

Big company + Obscure stable internal tool + Maintenance of that tool = thumbs up

vor 5 Jahrendiscuss

What makes you say that they are in decline?

vor 5 Jahrendiscuss

Kickfurther | Full Stack Engineer - Laravel | Remote - North and South America | Full-Time | https://www.kickfurther.com/

Kickfurther is creating a marketplace that serves users who want to earn money by participating in the retail economy, and consumer product businesses who need financing for inventory to meet their own growing demand and production cycles. We use 3rd-party data integrations from a variety of sources to vet co-op applications and automate business processes. We regularly close crowd-sourced funding campaigns in just a few minutes as our user base rushes to fund co-ops, meaning that we simultaneously process many transactions in a seamless way. Kickfurther itself just closed our largest investment round; there is a lot to accomplish and we have the means to accomplish it.

To that end, Kickfurther is looking to add a new member to our dev team. The team is still fairly small, so all of your contributions will be noticed. All of your work will be deployed once it is reviewed, so you will be an essential part of building our platform. We are an honest and hard-working group of people who take our work seriously, but don’t take ourselves too seriously. We lean towards simple solutions to complex problems, and complicate things only when absolutely necessary. We have strong opinions weakly held and are always open to new ideas. We need someone who shares our values and is excited to solve the challenges ahead.

https://github.com/Kickfurther/careers/blob/main/full-stack-...

vor 5 Jahrendiscuss

I guess it's subjective. I found brave quite decent and there were some moving moments.

Worst for me would be Cars 2 and in the context of general films, it's still okay.

vor 5 Jahrendiscuss

Would it be possible to view personal private respository code with this?

I'm not sure what it would take to make that happen.

vor 6 Jahrendiscuss

For the 1st, a critical mass has to be reached in order for it to become a problem. Even if it does and slows it down, then the places where the speeds are fast will become attractive for the same people. This is just a hypothesis, it will be interesting to see the dynamics created by such service.

W.r.t 2nd, Yep!

vor 6 Jahrendiscuss

Yep. Totally agree.

I appreciate apple's attempt in the privacy space as well as loathe their 30% mandatory in app purchase integration which EPIC is fighting against. You need someone who is big like Epic to stand against apple as they the resource to battle it out legally.

One of my clients is a small player and they have stripe integration for their service. The android platform uses the stripe integration where as the apple platform has forced in app purchases down their throat. In app purchases should be an option and it should be upto the customer if they want to use it or not.

vor 6 Jahrendiscuss

They weren't engineers. It was shows attempt to humanize the algorithm.

I agree it was cringy but the shows intent wasn't bad.

vor 6 Jahrendiscuss

This nerdwriter1 video in kintsugi is also really nice.

https://youtu.be/lT55_u8URU0

vor 6 Jahrendiscuss

Hey guys,

I just finished watching Dark and out of curiosity, I wanted to see how the family members are linked to each other since its not the typical family tree. I used a network graph visualization for this and each node is clickable and opens up the character detail from the wiki. [Warning] If you haven't seen season 3 then I would advise not to click on the link as it contains spoilers

vor 7 Jahrendiscuss

For a moment I thought Bajaj and TVS came together.

vor 7 Jahrendiscuss

Would these suggestions apply for Django?

I'm assuming Django also faces similar issues

vor 7 Jahrendiscuss

Thanks!

vor 7 Jahrendiscuss

Yeah, I'll be adding that. There will be an option to toggle between list and thumbnail view.

vor 7 Jahrendiscuss

I started off having strong integration with ML frameworks but eventually, I made it more generic to track any kind of metrics and parameters especially if one is using a custom algorithm.

I'm working on adding data versioning where one can keep version their datasets and pull it anywhere using a line of code in python or command line.

vor 7 Jahrendiscuss

Hi,

The author here.

ModelChimp is an experiment tracker for Deep Learning and Machine Learning experiments.

ModelChimp provides the following features:

  - Real-time tracking of parameters and metrics
  - Realtime charts for experiment metrics at epoch level
  - Code used for the experiment
  - Experiment comparison
  - Collaborate and share experiments with team members
  - Python objects storage such as data objects and model 
    objects which can be used pulled for other experiments
  - Storage of test and validation images for computer vision 
    use cases. Useful for post experiment forensics of deep 
    learning models
  - Server based solution with user registration and 
    authentication

vor 7 Jahrendiscuss

This is like an app store of ml models which is pretty cool. There are a couple of tooling around ML that more of compliments the above than being redundant.

-- Open Source --

Mlflow : https://github.com/mlflow/mlflow

Polyaxon : https://github.com/polyaxon/polyaxon

Modelchimp(mine): https://github.com/ModelChimp/modelchimp

ModelDB : https://github.com/mitdbg/modeldb

Sacred : https://github.com/IDSIA/sacred

-- Non open source --

Cometml : https://www.comet.ml/

Weights and Biases : https://www.wandb.com/

MissinglinkAi : https://missinglink.ai/