Comments (4)
I am providing in the next release an utility for inference on a single instance.
In the mean time either you do this yourself (take the model pertained and pass only one example per time) or, having all of them together, take the node and edge prediction all together and using number of nodes and edges per each graph as indices to correctly retrieve the prediction separately.
I close the issue.
Please provide descriptions or screenshots to leave better documentation for other users too.
from doc2graph.
Hi, is there a timeline on when this inference feature will be added?
from doc2graph.
Hi @ranzmigrod,
thanks for your patience 🙌🏼 (this should turn handy also for you @NaveenVinayakS ).
In the meantime I develop the novel feature, try this out!
In training/funsd.py
:
- Line 177: use only one graph
gi
instead of batching them all. You can access an item either usingtest_data.graphs[gi]
ortest_data[gi]
- Line 184:
n
ande
should be the nodes and edges predictions, respectively. If you want theentity labelling
predictions, just usen -> _, classes = torch.max(n, dim=1)
.
Predictions are "one to one" with the nodes (entities) of the graph: you can access them through your_graph.ndata['geom']
to print them on the image along with their predicted class (remember to scale them back! 🧐)
from doc2graph.
Thanks for your replay @andreagemelli
from doc2graph.
Related Issues (19)
- FUNSD test result is very low! HOT 7
- Facing some Issue when i try to run main.py HOT 2
- RunTime error: float division by zero in main.py HOT 5
- python src/main.py giving KeyError: 'feat' HOT 4
- dont know where the output /prediciton is saved.. HOT 1
- what does this actually tell by this prediciton .? HOT 2
- get RuntimeError: Error(s) in loading state_dict for E2E: when doing test on model from training HOT 2
- Training on DocVQA? HOT 3
- Passing linking for test images . HOT 1
- Environment installation problem HOT 3
- e2e-funsd-best.pt Error(s) in loading state_dict HOT 3
- RuntimeError: CUDA out of memory HOT 1
- How to set tg batch_size HOT 3
- OSError: [E050] Can't find model 'en_core_web_lg'.
- Pydantic problems with dependency mismatches HOT 6
- RuntimeError: weight tensor should be defined either for all 4 classes or no classes but got weight tensor of shape: [3] HOT 4
- Anyone successfully had this run on M1 Mac CPU?
- How to train my data ?
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 doc2graph.