continue
else:
score.append(False)
return np.sum(score) / len(score), score
if __name__ == "__main__":
header = ["word", "iteration", "rt", "freq", "cycles"]
results = []
random.seed(44)
path = "../../corpora/lexicon_projects/elp-items.csv"
words = np.array(list(read_elp_format(path, lengths=[4])))
freqs = [x['frequency'] + 1 for x in words]
freqs = np.log10(freqs)
sampler = BinnedSampler(words, freqs)
np.random.seed(44)
n_cyc = 1000
for idx in tqdm(range(100)):
w = deepcopy(sampler.sample(1000))
rt = np.array([x['rt'] for x in w])
inputs = ('letters-features',)
continue
else:
score.append(False)
return np.sum(score) / len(score), score
if __name__ == "__main__":
header = ["word", "iteration", "rt", "freq", "cycles"]
results = []
random.seed(44)
path = "../../corpora/lexicon_projects/elp-items.csv"
w = read_elp_format(path, lengths=[4])
np.random.seed(44)
n_cyc = 1000
rt = np.array([x['rt'] for x in w])
inputs = ('letters-features',)
w = process_data(w,
decomposable=('orthography',),
decomposable_names=('letters',),
feature_layers=('letters',),
feature_sets=('fourteen',),
negative_features=True,
length_adaptation=True)
score.append(True)
continue
else:
score.append(False)
return np.sum(score) / len(score), score
if __name__ == "__main__":
header = ["word", "rt", "freq", "cycles"]
results = []
path = "../../corpora/lexicon_projects/elp-items.csv"
words = read_elp_format(path, lengths=list(range(3, 11)))
for x in words:
x['frequency'] += 1
x['log_frequency'] = np.log10(x['frequency'])
n_cyc = 1000
rt = np.array([x['rt'] for x in words])
w = process_data(words,
decomposable=('orthography', ),
decomposable_names=('letters', ),
feature_layers=('letters', ),
feature_sets=('fourteen', ),
negative_features=True,
length_adaptation=True)
score.append(False)
return np.sum(score), score
if __name__ == "__main__":
header = ['word', 'iteration', 'rt', 'freq', 'cycles', 'le', 'ne', 'spa']
results = []
random.seed(44)
threshold = .7
path = "../../corpora/lexicon_projects/elp-items.csv"
words = np.array(list(read_elp_format(path, lengths=list(range(3, 11)))))
num_to_sample = len(words) // 4
freqs = [x['frequency'] + 1 for x in words]
freqs = np.log10(freqs)
sampler = BinnedSampler(words, freqs)
total = (2**3) * 100
n_cyc = 350
for idx, (le, ne, spa) in enumerate(
product([True, False], [True, False], [True, False])):
length_adaptation = le
negative_evidence = ne
space_character = spa