Skip to content

Commit

Permalink
Creates new run_all_trials for parameter finding
Browse files Browse the repository at this point in the history
  • Loading branch information
@tbohlen
tbohlen committed Dec 30, 2012
1 parent ffa03b9 commit ed6b278
Showing 1 changed file with 59 additions and 17 deletions.
76 changes: 59 additions & 17 deletions web/analysis.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -294,6 +294,29 @@ def compare_trial(trial_id, move_probability):
This method now runs each child through compare_trial with find_params_for_move, rather than move_probability.
"""
def run_all_trials():
# runs all the trials and writes out all data into a json string saved to results.json
trials = db.get_all_trials()
trialsData = {}
for trial in trials:
trial = compare_trial(trial._id, move_probability)
# find a results file that does not yet exist
base = "./results"
name = base
number = 0
while os.path.exists(name + ".json"):
name = base + str(number)
number += 1
f = open(name + ".json")
json.dump(trials, f)
f.close()

"""
Function: run_all_trials_for_params
Gathers all trial ids from the database and runs them all through compare_trial, saving the results out into a file. This function finds the best parameters for each move, rather than the likelihood of the human move.
This method now runs each child through compare_trial with find_params_for_move, rather than move_probability.
"""
def run_all_trials_for_params():
# runs all the trials and writes out all data into a json string saved to results.json
trials = db.get_all_trials()
trialsData = {}
Expand All @@ -310,13 +333,29 @@ def run_all_trials():
json.dump(trials, f)
f.close()

"""
Function: gibbs_prob
takes in a move probability, move time, and new group, and returns a probability that this set of move params is
"""
def gibbs_prob(group, time, prob):
maxVal = -inf
max_group = -1
for key, val in prob.iteritems():
if val > maxVal:
group = key
maxVal = val
if max_group != group:
return 0
else:
return var(prob)

"""
Gibbs Sampling Functions
These functions sample the various parameters of our particle filter model for the gibbs sampler.
"""

def sample_mu(current_partition, image_id, group, params):
def sample_mu(current_partition, image_id, group, time_elapsed, params):
#print "Sampling mu..."
# new random mu
new_mu = truncnorm.rvs((0.0-params[0])/10.0, (255.0-params[0])/10.0, params[0], 10.0) # pretty random...
Expand All @@ -333,7 +372,7 @@ def sample_mu(current_partition, image_id, group, params):
prob = new_mu_log_prob
return [params, prob]

def sample_mu_conf(current_partition, image_id, group, params):
def sample_mu_conf(current_partition, image_id, group, time_elapsed, params):
#print "Sampling mu_conf..."
# new random mu_conf
new_mu_conf = truncnorm.rvs((0.0-params[1])/0.25, (1.0-params[1])/0.25, params[1], 0.25) # pretty random...
Expand All @@ -350,7 +389,7 @@ def sample_mu_conf(current_partition, image_id, group, params):
prob = new_mu_conf_log_prob
return [params, prob]

def sample_var(current_partition, image_id, group, params):
def sample_var(current_partition, image_id, group, time_elapsed, params):
#print "Sampling variance..."
# new random var
new_var = truncnorm.rvs((0.0-params[2])/8.0, (256.0*256.0-params[2])/8.0, params[1], 8.0) # pretty random...
Expand All @@ -367,7 +406,7 @@ def sample_var(current_partition, image_id, group, params):
prob = new_var_log_prob
return [params, prob]

def sample_var_conf(current_partition, image_id, group, params):
def sample_var_conf(current_partition, image_id, group, time_elapsed, params):
#print "Sampling variance conf..."
# new random var_conf
new_var_conf = truncnorm.rvs((0.0-params[3])/0.25, (1.0-params[3])/0.25, params[3], 0.25) # pretty random...
Expand All @@ -384,7 +423,7 @@ def sample_var_conf(current_partition, image_id, group, params):
prob = new_var_conf_log_prob
return params, prob

def sample_disp(current_partition, image_id, group, params):
def sample_disp(current_partition, image_id, group, time_elapsed, params):
#print "Sampling distribution..."
# new random disp
new_disp = truncnorm.rvs((0.0-params[4])/4.0, (1000.0-params[4])/4.0, params[4], 4.0) # pretty random...
Expand Down Expand Up @@ -422,7 +461,7 @@ def sample_disp(current_partition, image_id, group, params):
Returns:
The mean of the samples taken
"""
def find_params_for_move(current_partition, image_id, group):
def find_params_for_move(current_partition, move):
# first generate a random start point
mu = random.normal(255.0/2.0, 256.0/4.0)
mu_conf = random.uniform(0.0, 1.0)
Expand All @@ -433,32 +472,36 @@ def find_params_for_move(current_partition, image_id, group):
params = [mu, mu_conf, sig, var_conf, disp]
sample_params = []

image_id = move['image_id']
group = move['new_group']
time_elapsed = move['time_elapsed']

# iterate across the variables, testing each new suggestion in turn
print "Walking in...."
for i in range(walk_in):
# sample mu
params, prob = sample_mu(current_partition, image_id, group, params)
params, prob = sample_mu(current_partition, image_id, group, time_elapsed, params)
# sample mu_conf
params, prob = sample_mu_conf(current_partition, image_id, group, params)
params, prob = sample_mu_conf(current_partition, image_id, group, time_elapsed, params)
# sample var
params, prob = sample_var(current_partition, image_id, group, params)
params, prob = sample_var(current_partition, image_id, group, time_elapsed, params)
# sample var_conf
params, prob = sample_var_conf(current_partition, image_id, group, params)
params, prob = sample_var_conf(current_partition, image_id, group, time_elapsed, params)
# sample disp
params, prob = sample_disp(current_partition, image_id, group, params)
params, prob = sample_disp(current_partition, image_id, group, time_elapsed, params)

print "Sampling..."
for j in range(samples):
# sample mu
params, prob = sample_mu(current_partition, image_id, group, params)
params, prob = sample_mu(current_partition, image_id, group, time_elapsed, params)
# sample mu_conf
params, prob = sample_mu_conf(current_partition, image_id, group, params)
params, prob = sample_mu_conf(current_partition, image_id, group, time_elapsed, params)
# sample var
params, prob = sample_var(current_partition, image_id, group, params)
params, prob = sample_var(current_partition, image_id, group, time_elapsed, params)
# sample var_conf
params, prob = sample_var_conf(current_partition, image_id, group, params)
params, prob = sample_var_conf(current_partition, image_id, group, time_elapsed, params)
# sample disp
params, prob = sample_disp(current_partition, image_id, group, params)
params, prob = sample_disp(current_partition, image_id, group, time_elapsed, params)
print "New sample: " + str(params)

# save the sample we just generated
Expand Down Expand Up @@ -498,7 +541,6 @@ def randomize(l):
def decide_group(partition, image_id):
probabilities = move_probability(partition, image_id)
print 'Probabilities:', probabilities
sorted_probabilities = sorted(probabilities.iteritems(), key=operator.itemgetter(1))
maxVal = -inf
group = -1
for key, val in probabilities.iteritems():
Expand Down

0 comments on commit ed6b278

Please sign in to comment.