-
Notifications
You must be signed in to change notification settings - Fork 1
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Commented mnist_igor_two.py plus removed double parens in Ca_HVA.mod #6
base: master
Are you sure you want to change the base?
Conversation
…nto models_igr
…red); marking of true/predicted classes (record.py), randomization of starting weights (ebner_agent.py)
…T, EPSILON); mnist_igor_two.py chooses one output"
agents/ebner_agent.py
Outdated
delay=1, netcon_weight=WEIGHT, ach_weight=1, da_weight=1, rule='all', | ||
ACh_tau=50, Da_tau=50) | ||
delay=1, netcon_weight=np.max((0, WEIGHT + np.random.normal(loc=0.0, scale=EPSILON))), | ||
ach_weight=1, da_weight=1, rule='all', ACh_tau=50, Da_tau=50) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I have doubt about netcon_weight to make it random - let's talk on the call
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
As of now, reject these modifications. It needs deeper tests.
mnist_igor_two.py
Outdated
sys.exit(0) | ||
|
||
|
||
class EbOlAg(EbnerOlfactoryAgent): |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Naming should be consistent with other Agents. So name it Ebner...Agent
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Oh, I did not require the inclusion of mnist_igor_two.py into the master. It was only designed for you commenting on my comments whether I understood it correctly. And try some very minor changes into the architecture.
Nonetheless, the EbnerOlfactoryAgent() has some constants in it (numbers of hidden and inhibitory neurons) and it makes it harder to extend it easily.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
OK, that's fine for now. but if you want me to comment on something - maybe the best way is to use Github Gist in the future: https://gist.github.com
mnist_igor_two.py
Outdated
sys.exit(0) | ||
|
||
|
||
class EbOlAg(EbnerOlfactoryAgent): |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
OK, that's fine for now. but if you want me to comment on something - maybe the best way is to use Github Gist in the future: https://gist.github.com
# Conflicts: # agents/ebner_agent.py
break | ||
if not found: | ||
print("NEURON-7.7 path not found in sys.path; exiting") | ||
sys.exit(0) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This not works for me (on Ubuntu)
mnist_igor_two.py
Outdated
# info left AGENT_STEPSIZE / DT for additional steps on the left of display | ||
agent.rec_output.plot(animate=True, | ||
steps=int(AGENT_STEPSIZE / DT + 2 * display_interval * AGENT_STEPSIZE / DT), | ||
true_class=last_true, pred_class=last_predicted, stepsize=AGENT_STEPSIZE, dt=DT) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This code requires changes to neuron++ so before we merge them - I can't merge this one to the gym repo
…gor_run is the latest try-on version of mnist
# Conflicts: # agents/agent.py
# Conflicts: # agents/agent.py # agents/ebner_agent.py
# Conflicts: # mnist_run.py # simple_run.py
# Conflicts: # agents/ebner_agent.py # mnist_run.py
…ent.py + mnist_igor_two.py
…nt: corrected input field calculation; ebner_olfactory_agent: extended init parameters; mnist_igor_two: rewritten + clear
…nt: corrected input field calculation; ebner_olfactory_agent: extended init parameters; mnist_igor_two: rewritten + clear
# Conflicts: # agents/agent.py # agents/ebner_agent.py # mnist_igor_two.py
…nt: corrected input field calculation; ebner_olfactory_agent: extended init parameters; mnist_igor_two: rewritten + clear
…nt: corrected input field calculation; ebner_olfactory_agent: extended init parameters; mnist_igor_two: rewritten + clear
Hello @igorpodolak! Thanks for updating this PR. We checked the lines you've touched for PEP 8 issues, and found:
|
UPDATE:
In agents/agent.py and agents/ebner_agent.py corrected:
input_syn_per_cell computed as an area resulting from the kernel ratio,
corrected upper ranges for x and y to constitute for the case when stride < kernel_width