Comments (4)
Hello pliz,
To fix the first problem - make sure cudamat is on your LD_LIBRARY_PATH. See INSTALL.txt for more details.
It is failing on data import because it can't find the data. Did you download the data and run setup_examples.py ? If not, see INSTALL.txt for how to do this.
from deepnet.
Thanks Nitish! cudamat is on my path as I said if I run python anywhere else and just import cudamat - works smoothly, which does not solve the problem.
With the second issue - I've resolved it by looking into mnist.pbtxt and observing relative paths there, but relative to the examples directory. After changing the relative path to the absolute path - I can run the example (using my trainer.py copy method rather than your runall.sh) and it works (well, still running).
Thanks!
Sergey
from deepnet.
So I've figured out the problems:
-
Before reading INSTALL.txt I've shadowed cudamat.py with a different file in the deepnet directory. Removing it - solved the first problem.
-
When running setup_examples.py I gave the script relative paths :) That created problems in issue 2. An easy fix would be:
diff --git a/deepnet/examples/setup_examples.py b/deepnet/examples/setup_examples.py
index e23fee5..136ba43 100644
--- a/deepnet/examples/setup_examples.py
+++ b/deepnet/examples/setup_examples.py
@@ -37,8 +37,8 @@ def MakeModels(model_file, output_path):
util.WritePbtxt(model_file, model)def main():
- data_path = sys.argv[1] # Path to mnist data directory.
- output_path = sys.argv[2] # Path where learned models will be written.
- data_path = os.path.abspath(sys.argv[1]) # Path to mnist data directory.
- output_path = os.path.abspath(sys.argv[2]) # Path where learned models will be written.
data_pbtxt_file = os.path.join(data_path, 'mnist.pbtxt')
MakeDataPbtxt(data_pbtxt_file, data_path)
Thanks!
Sergey
from deepnet.
Thanks for the suggestion. I have added abspath in a new commit.
from deepnet.
Related Issues (20)
- How to set gaussian and binary layers in RBM? HOT 2
- Error 127 while 'sudo make' cudamat HOT 2
- CUDAMatException: Incompatible matrix dimensions
- close
- About label prediction in evaluation of DBM HOT 3
- Can deepnet choose gpu ID?
- Error when COMPILING CUDAMAT AND CUDAMAT_CONV on windows
- a question about multimodal data output
- No module named google.probuf HOT 1
- attribure Error:nonetype object has no attribute seed
- find this trouble on run 段错误 (核心已转储)
- missing data imputation
- Error while 'make' cudamat HOT 3
- Error 127 while 'make' cudamat HOT 1
- For Multi-label Classification
- What's the meaning of this error?
- Output model for DBN model_type classifier
- Using the trained DBN
- reproduce multimodal dbm result HOT 3
- Any clue or example how to process superpixels
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 deepnet.