hhexiy / opponent Goto Github PK
View Code? Open in Web Editor NEWImplementation for ICML 16 paper "Deep reinforcement learning with opponent modeling"
Implementation for ICML 16 paper "Deep reinforcement learning with opponent modeling"
Hi,
Thank you for sharing the code, I just run it, and get an error:
module 'setup' not found:No LuaRocks module found for setup
I am new to torch. So I cannot figure out whether I did not install some package/part in torch for this setup module or the code just missed this part? And, as to the function env_setup()
, I cannot find its definition in the code.
Could you please give me some hint on solving this?
By the way, I found the argument setting in makefile is still about run_qb
, If I want to run_soccer
, which part of the arguments should I change?
Sorry to be a bother. Any reply would be really appreciated.
When run make train_content
. I receive this error:
loading data from text file dat/qb-all/content_data.txt
loading dat/qb-all/content_data.txt.mapping.t7
maximum sequence length: 266
vocab size: 22052
# answers/classes: 1046
# users: 0
# categories: 54
creating mappings: 0.22
loading dat/qb-all/content_data.txt.tensor.t7
/home/ethereal/torch/install/bin/luajit: ...aduationproject/opponent-master/util/MinibatchLoader.lua:442: bad argument #1 to 'ipairs' (table expected, got nil)
stack traceback:
[C]: in function 'ipairs'
...aduationproject/opponent-master/util/MinibatchLoader.lua:442: in function 'text_to_tensor'
...aduationproject/opponent-master/util/MinibatchLoader.lua:21: in function 'load_data'
train_content.lua:106: in main chunk
[C]: in function 'dofile'
...real/torch/install/lib/luarocks/rocks/trepl/scm-1/bin/th:150: in main chunk
[C]: at 0x561f089bb570
Makefile:60: recipe for target 'train_content' failed
make: *** [train_content] Error 1
How can I solve this problem?
when I run make, error happen:
opponent-master: make
if [ ! -d dat/glove ]; then mkdir -p dat/glove; fi
wget http://www-nlp.stanford.edu/data/glove.840B.300d.txt.gz -P dat/glove
--2017-06-06 12:03:45-- http://www-nlp.stanford.edu/data/glove.840B.300d.txt.gz
Resolving www-nlp.stanford.edu... 171.64.67.140
Connecting to www-nlp.stanford.edu|171.64.67.140|:80... connected.
HTTP request sent, awaiting response... 302 Found
Location: https://nlp.stanford.edu/data/glove.840B.300d.txt.gz [following]
--2017-06-06 12:03:45-- https://nlp.stanford.edu/data/glove.840B.300d.txt.gz
Resolving nlp.stanford.edu... 171.64.67.140
Connecting to nlp.stanford.edu|171.64.67.140|:443... connected.
HTTP request sent, awaiting response... 404 Not Found
2017-06-06 12:03:46 ERROR 404: Not Found.
make: *** [dat/glove/glove.840B.300d.txt] Error 8
When run make run_soccer
I receive this error. Do you have any suggestion?
th train_soccer_agent.lua -agent "QBNeuralQLearner" -agent_params "lr="0.0005",ep="0.3",ep_end="0.1",ep_endt="500000",discount="0.9",learn_start="1000",update_freq="4",minibatch_size=64,rescale_r=1,bufferSize=512,valid_size=500,target_q=10000,clip_delta=1,replay_memory=100000,n_experts=3,model='concat '" -opponent RandomAgent -games 5000 -eval_games 5000 -hist_len 2 -eval_freq 0 -prog_freq 500 -gpuid -1 -threads 8 -test 0 -network "" -best -savefile "" -height 6 -width 9 -defend 0.5 -seed 1
defensive agent: 0.5
state dim: 15
/home/novin/torch/install/bin/lua: ./setup.lua:130: attempt to call field '?' (a nil value)
stack traceback:
./setup.lua:130: in function 'soccer_dqn_setup'
train_soccer_agent.lua:51: in main chunk
[C]: in function 'dofile'
...ovin/torch/install/lib/luarocks/rocks/trepl/scm-1/bin/th:150: in main chunk
[C]: in ?
Makefile:108: recipe for target 'run_soccer' failed
make: *** [run_soccer] Error 1
When run make dat/qb/buzz_data.txt
, I receive this error.
make: *** No rule to make target 'dat/qb/naqt_questions.csv', needed by 'dat/qb/buzz_data.txt'. Stop.
I have downloaded glove.840B.300d.txt.gz and qb_emnlp_2012.tar.gz, and obtained vec.t7, vocab.txt, buzzes.csv and questions,csv.but i can't find dat/qb/naqt_questions.csv.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.