Comments (3)
So you are diving into the source code. Yes, the output units are constrained using ICs and BCs.
the
train_x
concludes all the BC and initial points and domain points and the shape of the each output will be train_x
Can you paste the link to the relevant code?
So how does component work and why the initial and BCs of each component exist in the batch_size for both outputs
Again, a link to the particular code might be helpful.
How does the bc_points work at the beginning of the
train_x
?
What do you mean by beginning of train_x
? train_x
is a variable containing the training dataset.
from deepxde.
I read the explanations of train_x and it said that the train_x is ordered by bc_points containing BC and initial points and domain points. For example when there is a system of PDE with two outputs y[:,0:1], y[:,1:2]. If the training data has 2000 BC and initial points AND 3000 domain points, the train_x would have 5000 training points. My question is that why both outputs have the same shape since their initial and boundary conditions are different?
Better explanation: consider there are two boundary conditions and two initial conditions for each PDE in the system. For each initial conditions 100 points were used total of 200. Why does the shape of the each output contains the 200 points.
If from our 2000 bc_points the last 200 is the initial points, what will be the y[-200:,0,1] and y[-200:,1:2]?
Ps: I use anchors thus I know my training points and I examined the explanation but struggling to understand the value of outputs on initial points.
from deepxde.
My question is that why both outputs have the same shape since their initial and boundary conditions are different? Better explanation: consider there are two boundary conditions and two initial conditions for each PDE in the system. For each initial conditions 100 points were used total of 200. Why does the shape of the each output contains the 200 points. If from our 2000 bc_points the last 200 is the initial points, what will be the y[-200:,0,1] and y[-200:,1:2]?
I don't really understand you question and what you are confused about. An easy solution is that you can use a simple 1D example as a test case, and print train_x
.
from deepxde.
Related Issues (20)
- FNN class constructor does not contain regularization implementation for PyTorch backend HOT 1
- batch_size not fully implemented in DataSet class HOT 1
- PIDeepONet, aux_vars on BC
- PINNs energy method HOT 11
- Export data for plot HOT 4
- Cannot convert a symbolic tf.Tensor (Placeholder_6:0) to a numpy array HOT 2
- Hi, I read the changes you made for deepxde and I have some similar questions I would like to ask you
- Does PINN training require a reference solution to the training data points? HOT 1
- How to give different inputs for subnetworks in PFNN
- Polygon PeriodicBC Problem
- Model doesnot converge even after trying everything HOT 2
- possible bug: 'targets' variable not used in PDE losses() method?
- Hyperelasticity HOT 2
- Boundary condition as input for branch net in PI DeepONet
- ImportError: cannot import name 'check_pandas_support' from 'sklearn.utils' HOT 1
- ImportError check_pandas_support HOT 4
- Operator prediction becomes constant
- PDE error evaluated on bc points in PDE.losses but not used by following codes
- Is there a demo/code/model available for Fourier-MIONet?
- PDE with additional parameters
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from deepxde.