February is the month for data science before Open Data Event on March 4th. After the first session about data mining tools on Feb 11th, Coderbunker will host practical deep learning meeting on Saturday Feb 18th.
What is it?
This group is for everyone interested in Machine Learning with special focus on Deep Nets using TensorFlow.
If you’ve been trying to understand DL theory and write a working code, you might have noticed that some problems have steep learning curve. If you want to make faster progress, bring it to our group and let’s brainstorm the solution together! It does not matter if you’re learning, stuck on a problem or just working on something interesting and want to share, just bring it to meeting.
Structure of the meeting
The meeting will start with participants pitching their problem (max 5 minutes). The group votes which problem they find most interesting and everyone start working on it. Basing on number of participants, if the group is too big to work together on single problem, we will divide into smaller groups, brainstorm/live code and share our findings between groups.
So come prepared and bring your problem. As mentioned before, it can be anything you’re currently working on. The more interesting problem you bring, the higher chances everyone will help you to solve it.
Requirements for developers
at least basic knowledge in Python
at least basic understanding of machine learning and deep neural networks
good if you know at least read and understood TensorFlow basics from “MNIST FOR ML Beginners” (https://www.tensorflow.org/tutorials/mnist/beginners/) and TensorFlow 101 (https://www.tensorflow.org/tutorials/mnist/tf/)
Requirements for domain experts
Clean (consistent, structured) dataset importable into a database
Well defined measurable desired end result
There are many ML frameworks available but we chose one to ease the communication between group participants. It’s just easier to explain concept with real code and TensorFlow is already the most trending DeepLearning framework.
Organizer: Bartek Grasza
Bart has 12+ years experience working in the IT industry taking on several roles ranging from Project/Account Manager, through Developer to System Administrator. Likes to broaden his knowledge by playing with various, but not closely related technologies like docker, semantic web, microservices. Big fan of Scrum/Agile. His adventure with Data Science is mainly focused on Deep Learning for NLP using Python. Currently at specifiko.com, happily working with on backend using Ruby on Rails and Python.
Coderbunker: Shanghai Xuhui, 1199 Panyu Road, Building 3, underground. Take Line 3/4/9 to Yishan Road.
First meetup: Saturday February 18th 2017
Every 2 Saturdays afterwards
Also welcome to join Coderbunker Co-learning sessions weekly on Saturday
How much is it?
Free for Agora Space and Coderbunker members.
2 hour session is 50 RMB (includes 1 drink)
160 RMB for all-day access pass for non-members
Ask us about discount long-term packages!
When everyone, together tries to solve single problem, no matter what problem we choose, you will learn something new. Whether it’s new debugging method, optimization, some cool function, data preprocessing or something else.
If you’re experienced user and might think that you won’t learn anything new, remember that best way of solidifying your knowledge is by explaining problem to others so that they understand it. It’s very important step taken from (Richard) Feynman technique.
Scan QR code to join our "Coderbunker Data Sciences ML DL" Wechat group chat
We just sent you an email. Please click the link in the email to confirm your subscription!
OKSubscriptions powered by Strikingly