df_high_awe = df_sub[df_sub['awareness'] == 'Aware'].sort_values(['sub','variable','Rating',])
df_low_awe = df_sub[df_sub['awareness'] == 'Unaware'].sort_values(['sub','variable','Rating',])
ps = utils.resample_ttest_2sample(df_high_awe['value'].values,
df_low_awe['value'].values,
n_ps = 100,
n_permutation = 5000)
results['dprime_type'].append(dpr)
results['experiment'].append(exp)
results['ps_mean'].append(ps.mean())
results['ps_std'].append(ps.std())
results['high_aware'].append(df_high_awe['value'].values.mean())
results['low_aware'].append(df_low_awe['value'].values.mean())
results = pd.DataFrame(results)
results = results.sort_values('ps_mean')
converter = utils.MCPConverter(pvals = results['ps_mean'].values)
d = converter.adjust_many()
results['p_corrected'] = d['bonferroni'].values
results['stars'] = results['p_corrected'].apply(utils.stars)
for experiment in pd.unique(df_plot['Experiment']):
results_sub = results[results['experiment'] == experiment]
g = sns.catplot(x = 'awareness',
y = 'value',
hue = 'Rating',
col = 'dprime_type',
# row = 'Experiment',
data = df_plot[df_plot['Experiment'] == experiment],
aspect = 2,
df_sub
anova = ols('mean_variance ~ model_name',data = df_sub).fit()
aov_table = sm.stats.anova_lm(anova,typ=2)
print(aov_table)
temp['F'].append(aov_table['F']['model_name'])
temp['df_nonimator'].append(aov_table['df']['model_name'])
temp['df_denominator'].append(aov_table['df']['Residual'])
temp['p'].append(aov_table['PR(>F)']['model_name'])
temp['model'].append(model)
temp['condition'].append(condition)
temp['roi'].append(roi)
anova_results = pd.DataFrame(temp)
temp = []
for roi,df_sub in anova_results.groupby('condition'):
df_sub = df_sub.sort_values('p')
converter = utils.MCPConverter(pvals = df_sub['p'].values)
d = converter.adjust_many()
df_sub['p_corrected'] = d['bonferroni'].values
temp.append(df_sub)
anova_results = pd.concat(temp)
anova_results['stars'] = anova_results['p_corrected'].apply(utils.stars)
anova_results = anova_results.sort_values(['roi','condition','model'])
g = sns.catplot(x = 'roi',
y = 'mean_variance',
hue = 'model_name',
hue_order = ['vgg19',
'densenet121',
'mobilenetv2_1',
'Fast_Text',
0.5,
n_ps=100,
n_permutation=int(1e4),
one_tail=True)
df_post['model'].append(model_name)
df_post['window'].append(window)
df_post['train'].append(train)
df_post['score_mean'].append(df_sub['score'].values.mean())
df_post['score_std'].append(df_sub['score'].values.std())
df_post['pval'].append(ps.mean())
df_post = pd.DataFrame(df_post)
temp = []
for (train, model_name), df_sub in df_post.groupby(['train', 'model']):
df_sub = df_sub.sort_values(['pval'])
ps = df_sub['pval'].values
converter = utils.MCPConverter(pvals=ps)
d = converter.adjust_many()
df_sub['p_corrected'] = d['bonferroni'].values
temp.append(df_sub)
df_post = pd.concat(temp)
df_post['star'] = df_post['p_corrected'].apply(utils.stars)
df_post = df_post.sort_values(['window', 'train', 'model'])
df_post = df_post[df_post['window'] > 0]
df_plot['experiment_train'] = df_plot['experiment_train'].map({
'POS':
"Trained on Exp.1/ Tested on Exp.2",
'ATT':
"Trained on Exp.2/ Tested on Exp.1"
})
one_tail=True)
ps42 = utils.resample_ttest_2sample(m4['value'].values,
m2['value'].values,
n_ps=200,
n_permutation=5000)
for var_name, name in zip(['model', 'experiment', 'window'], attri):
results[var_name].append(name)
results['ps21'].append(ps21.mean())
results['ps32'].append(ps32.mean())
results['ps42'].append(ps42.mean())
results = pd.DataFrame(results)
corrected = []
for attri, df_sub in results.groupby(['model', 'experiment']):
df_sub = df_sub.iloc[np.argsort(df_sub['ps21'].values), :]
converter = utils.MCPConverter(pvals=df_sub['ps21'].values)
d = converter.adjust('bonferroni')
df_sub['ps21_corrected'] = d
df_sub = df_sub.iloc[np.argsort(df_sub['ps32'].values), :]
converter = utils.MCPConverter(pvals=df_sub['ps32'].values)
d = converter.adjust('bonferroni')
df_sub['ps32_corrected'] = d
df_sub = df_sub.iloc[np.argsort(df_sub['ps42'].values), :]
converter = utils.MCPConverter(pvals=df_sub['ps42'].values)
d = converter.adjust('bonferroni')
df_sub['ps42_corrected'] = d
corrected.append(df_sub)
corrected = pd.concat(corrected)