Closed cversteeg closed 6 years ago
The directory containing LFADS is most likely not in your PATH environment variable
Yup, that’s likely it.
I should probably add a check in the bash script to
make sure which run_lfads.py
evaluates to something to make this clearer.
Yep that was it. Found the error... PATH=$PATH:/path/to/models/research/lfads That's what I get for copying and pasting willy-nilly Thanks!
Danny,
Thanks for this awesome code! I've gotten everything working except for one thing; getting the final .sh files to run from the terminal. I an encountering an error with my generated .sh returns an error 'Unknown option: -- usage: python [option] ... [-c cmd | -m mod | file | -] [arg] ... Try `python -h' for more information.'
This is what the generated script looks like.
!/bin/bash
source activate tensorflow (DISPLAY=:50 python $(which run_lfads.py) --data_dir=/home/chris/Documents/MATLAB/exampleRun/param_eTl5Fm/single_LFADSChipsTest/lfadsInput --data_filename_stem=lfads --lfads_save_dir=/home/chris/Documents/MATLAB/exampleRun/param_eTl5Fm/single_LFADSChipsTest/lfadsOutput --cell_clip_value=5.000000 --factors_dim=8 --ic_enc_dim=64 --ci_enc_dim=128 --gen_dim=64 --keep_prob=0.950000 --learning_rate_decay_factor=0.980000 --device=/gpu:0 --co_dim=0 --do_causal_controller=false --do_feed_factors_to_controller=true --feedback_factors_or_rates=factors --controller_input_lag=1 --l2_gen_scale=500.000000 --l2_con_scale=500.000000 --batch_size=70 --kl_increase_steps=900 --l2_increase_steps=900 --ic_dim=64 --con_dim=128 --learning_rate_stop=0.001000 --temporal_spike_jitter_width=0 --allow_gpu_growth=true --kl_ic_weight=1.000000 --kl_co_weight=1.000000 --inject_ext_input_to_gen=false && python $(which run_lfads.py) --data_dir=/home/chris/Documents/MATLAB/exampleRun/param_eTl5Fm/single_LFADSChipsTest/lfadsInput --data_filename_stem=lfads --lfads_save_dir=/home/chris/Documents/MATLAB/exampleRun/param_eTl5Fm/single_LFADSChipsTest/lfadsOutput --kind=posterior_sample_and_average --batch_size=512 --checkpoint_pb_load_name=checkpoint_lve --cell_clip_value=5.000000 --factors_dim=8 --ic_enc_dim=64 --ci_enc_dim=128 --gen_dim=64 --keep_prob=0.950000 --learning_rate_decay_factor=0.980000 --device=/gpu:0 --co_dim=0 --do_causal_controller=false --do_feed_factors_to_controller=true --feedback_factors_or_rates=factors --controller_input_lag=1 --l2_gen_scale=500.000000 --l2_con_scale=500.000000 --kl_increase_steps=900 --l2_increase_steps=900 --ic_dim=64 --con_dim=128 --learning_rate_stop=0.001000 --allow_gpu_growth=true --kl_ic_weight=1.000000 --kl_co_weight=1.000000 --inject_ext_input_to_gen=false && python $(which run_lfads.py) --data_dir=/home/chris/Documents/MATLAB/exampleRun/param_eTl5Fm/single_LFADSChipsTest/lfadsInput --data_filename_stem=lfads --lfads_save_dir=/home/chris/Documents/MATLAB/exampleRun/param_eTl5Fm/single_LFADSChipsTest/lfadsOutput --kind=write_model_params --checkpoint_pb_load_name=checkpoint_lve --cell_clip_value=5.000000 --factors_dim=8 --ic_enc_dim=64 --ci_enc_dim=128 --gen_dim=64 --keep_prob=0.950000 --learning_rate_decay_factor=0.980000 --device=/gpu:0 --co_dim=0 --do_causal_controller=false --do_feed_factors_to_controller=true --feedback_factors_or_rates=factors --controller_input_lag=1 --l2_gen_scale=500.000000 --l2_con_scale=500.000000 --batch_size=70 --kl_increase_steps=900 --l2_increase_steps=900 --ic_dim=64 --con_dim=128 --learning_rate_stop=0.001000 --allow_gpu_growth=true --kl_ic_weight=1.000000 --kl_co_weight=1.000000 --inject_ext_input_to_gen=false) 2>&1 | tee /home/chris/Documents/MATLAB/exampleRun/param_eTl5Fm/single_LFADSChipsTest/lfads.out
Thanks! Chris Versteeg