def q5b2():
logger.log('Analyzing Q5(b2)...')
compare = {
'acc_converge': {
3: acc_converge_epoch(histories['Q4']['optimal']['val_accuracy']),
4: acc_converge_epoch(histories['Q5']['val_accuracy'])
},
'loss_converge': {
3: loss_converge_epoch(histories['Q4']['optimal']['val_loss']),
4: loss_converge_epoch(histories['Q5']['val_loss'])
},
'final_acc': {
3: training_result(histories['Q4']['optimal']['val_accuracy']),
4: training_result(histories['Q5']['val_accuracy'])
},
'final_loss': {
3: training_result(histories['Q4']['optimal']['val_loss'], mode='loss'),
4: training_result(histories['Q5']['val_loss'], mode='loss')
}
}
compare_df = pd.DataFrame(compare)
compare_df.to_csv('result/AQ5_result.csv')
logger.log('Saved result to \"result/AQ5_result.csv\"')
def compare():
logger.log('Compare all features, 6 features and 5 features')
mse_all = hist1['Q1']['a']['val_mse']
best_6 = val_mse[[c for c in df.columns
if len(c.split(",")) == 1]].idxmin()
mse_6 = hist2['Q2'][best_6]['val_mse']
best_5 = val_mse[[c for c in df.columns
if len(c.split(",")) == 2]].idxmin()
mse_5 = hist2['Q2'][best_5]['val_mse']
f, ax = plt.subplots(figsize=(10, 5))
plt.plot(mse_all, label='all features')
plt.plot(mse_6, label='6 features')
plt.plot(mse_5, label='5 features')
plt.xlabel('epoch', fontsize=18)
plt.ylabel('mean squared error', fontsize=18)
plt.legend(loc='upper right', fontsize=18)
plt.title('MSE for Different Numbers of Input Features',
fontsize=20,
pad=20)
plt.xlim(1000, 5000)
plt.ylim(0.0058, 0.010)
plt.xticks(fontsize=12, wrap=True)
plt.yticks(fontsize=12, wrap=True)
plt.tight_layout()
plt.savefig('result/BQ2_compare.png')
logger.log('Saved result to \"result/BQ2_compare.png\"')
final = {
'mse_converge': {
"all_features": loss_converge_epoch(mse_all),
"6_features": loss_converge_epoch(mse_6),
"5_features": loss_converge_epoch(mse_5)
},
'final_mse': {
"all_features": training_result(mse_all, mode='loss'),
"6_features": training_result(mse_6, mode='loss'),
"5_features": training_result(mse_5, mode='loss')
}
}
final_df = pd.DataFrame(final)
final_df.to_csv('result/BQ2_result.csv')
logger.log('Saved result to \"result/BQ2_result.csv\"')
# update hyperparameter
result = min(final['final_mse'].keys(),
key=lambda x: final['final_mse'][x])
if result == 'all_features':
removed = []
elif result == '6_features':
removed = list(map(int, best_6.split(",")))
elif result == '5_features':
removed = list(map(int, best_5.split(",")))
hyperparameters['input_shape'] = (len(columns) - len(removed), )
hyperparameters['removed'] = removed
write_dict_to_json(hyperparameters, args.params)
def compare():
df = pd.DataFrame(histories['Q3'])
val_mse = df[df.index == 'val_mse'].apply(
lambda x: x.explode()).reset_index(drop=True)
f, ax = plt.subplots(1, 1, figsize=(15, 5))
val_mse.plot()
plt.xlabel('epoch', fontsize=15)
plt.ylabel('mean squared error', fontsize=15)
plt.legend(loc='upper right', fontsize=12)
plt.title('MSE for Different Models', fontsize=15, pad=20)
plt.xlim(500, 10000)
plt.ylim(0.0055, 0.0115)
plt.xticks(fontsize=12)
plt.yticks(fontsize=12)
plt.tight_layout()
plt.savefig('result/BQ3_compare.png')
logger.log('Saved result to \"result/BQ3_compare.png\"')
df.applymap(lambda x: training_result(x, mode='loss')).to_csv(
'result/BQ3_compare.csv')
logger.log('Saved result to \"result/BQ3_compare.csv\"')
df.applymap(lambda x: loss_converge_epoch(x)).to_csv(
'result/BQ3_converge_epoch.csv')
logger.log('Saved result to \"result/BQ3_converge_epoch.csv\"')
def q1a():
logger.log('Analyzing Q1(a)...')
training_error = histories['Q1']['a']['mse']
testing_error = histories['Q1']['a']['val_mse']
f, ax = plt.subplots(figsize=(10, 5))
plt.plot(training_error, label='training')
plt.plot(testing_error, label='testing')
plt.xlabel('epoch', fontsize=18)
plt.ylabel('mean squared error', fontsize=18)
plt.xticks(fontsize=12)
plt.yticks(fontsize=12)
plt.legend(loc='upper right', fontsize=18)
plt.title('Training and Testing Error against Training Epochs', fontsize=20)
ax.axvline(
x=loss_converge_epoch(testing_error),
linestyle='--', linewidth=1, color='r'
)
plt.tight_layout()
plt.savefig('result/BQ1_error.png')
logger.log('Saved result to \"result/BQ1_error.png\"')
def q1b():
logger.log('Analyzing Q1(b)...')
training_loss = histories['Q1']['loss']
testing_loss = histories['Q1']['val_loss']
f, ax = plt.subplots(figsize=(10, 5))
plt.plot(training_loss, label='training')
plt.plot(testing_loss, label='testing')
plt.xlabel('epoch', fontsize=18)
plt.ylabel('loss', fontsize=18)
plt.xticks(fontsize=18)
plt.yticks(fontsize=18)
plt.legend(loc='upper right', fontsize=18)
plt.title('Training and Testing Losses against Training Epochs',
fontsize=20)
ax.axvline(x=loss_converge_epoch(testing_loss, mean_thres=0.01),
linestyle='--',
linewidth=1,
color='r')
plt.tight_layout()
plt.savefig('result/AQ1_loss.png')
logger.log('Saved result to \"result/AQ1_loss.png\"')