Comments (2)
Hi @stamatisvas ,
It seems that you did not specify num_class
parameter, which should be equal to the number of classes of your dataset.
from fedtree.
Yes, this was the problem and it is now working. Thank you! However, having a trained model I faced the following problems
- I want to use the Python interface in order to use the trained model and predict the test data (csv). What format should the test data follow? i.e during the training I used a csv file: [id, y, x1, x2, x3, x4, x5, x6, x7, x8, x9]. I tried the usual approaches for the test data like dropping the label and others but I am getting always the error
predictions = federatedcsvmodel.predict(X_test)
2023-05-26 12:06:21,520 INFO [default] #instances = 10845, #features = 11
2023-05-26 12:06:21,526 INFO Performance checkpoint [init trees] for block [predict] : [1 ms]
2023-05-26 12:06:21,527 INFO Performance checkpoint [copy data] for block [predict] : [2 ms ([0 ms] from checkpoint 'init trees')]
2023-05-26 12:06:21,683 INFO Executed [predict] in [158 ms]
free(): invalid pointer
- When I run from the terminal ./build/bin/FedTree-predict ./predict.conf where predict.conf points to the csv test data the output in the terminal is this:
2023-05-26 12:09:06,024 INFO dataset.cpp:396 : loading csv dataset from file ## /media/sf_Shared_Ubuntu_Windows/fedtree_data/test_data.csv ##
2023-05-26 12:09:06,142 INFO dataset.cpp:567 : #instances = 10845, #features = 10
2023-05-26 12:09:06,142 INFO dataset.cpp:581 : Load dataset using time: 0.117392 s
2023-05-26 12:09:06,161 INFO fedtree_predict.cpp:75 : multi-class accuracy = 0.984417
which makes sense (I am getting 99% accuracy with centralized models). However, the prediciton.txt file that is saved in the folder has totally random predictions most of which are 0, and this results in 0% accuracy when checking it manually.
from fedtree.
Related Issues (20)
- 安装问题 HOT 1
- python horizontal 模型训练自动killed问题 HOT 4
- How to train a random forests model? HOT 4
- Specify Data Partition in Horizontal Federated Tree Training HOT 1
- A Problem in Build on Linux HOT 2
- Install issue on MacOS HOT 5
- Is there any solution to bind `FedTree-train` program to specific cpu cores in standalone simulation? HOT 2
- Horizontal Federated Random Forest - Errors for Multiclass Classification HOT 1
- Error imputing more than 10 different data paths (files - .csv) HOT 2
- Is it an alternative way to install NTL? HOT 2
- documentation issue HOT 2
- why does it still require nvidia thrust cuda after set USE_CUDA to OFF HOT 1
- not working with grpc-1.53.0 & server still waiting after finish training HOT 8
- Error_when building with -DDISTRIBUTED=ON HOT 6
- A question about data partition in vertical FL in standalone simulation HOT 4
- Stack smashing when handling multiple csv files. HOT 3
- Feature Request: add support for "wait-for-ready"
- 可以在每一轮后输出测试集的评价指标吗? HOT 2
- prediction issue on vertical federated learning with he
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 fedtree.