karpathy / char-rnn

Multi-layer Recurrent Neural Networks (LSTM, GRU, RNN) for character-level language models in Torch
11.58k stars 2.58k forks source link

Error when attempting to use -data_dir to point to another input.txt file... #193

Open danindiana opened 7 years ago

danindiana commented 7 years ago

NOTE: When I place the training file in the /data/tinyshakespear directory and run just: th train.lua -gpuid -0 -num_layers 3 -rnn_size 600 -eval_val_every 30000

the program runs.

But when I attempt to use -data_dir and point to a different input file in a different directory I get:

speed@speed-desktop:~/Downloads/char-rnn-master$ th train.lua -data_dir /data/KJ -gpuid -0 -num_layers 3 -rnn_size 600 -eval_val_every 30000 using CUDA on GPU 0...
vocab.t7 and data.t7 do not exist. Running preprocessing... one-time setup: preprocessing input text file /data/KJ/input.txt... loading text file...
/home/speed/torch/install/bin/luajit: ./util/CharSplitLMMinibatchLoader.lua:133: /data/KJ/input.txt: No such file or directory stack traceback: [C]: in function 'assert' ./util/CharSplitLMMinibatchLoader.lua:133: in function 'text_to_tensor' ./util/CharSplitLMMinibatchLoader.lua:38: in function 'create' train.lua:111: in main chunk [C]: in function 'dofile' ...peed/torch/install/lib/luarocks/rocks/trepl/scm-1/bin/th:150: in main chunk [C]: at 0x00405e40

Luxter77 commented 6 years ago

this just happened to me too