def training(result_filename, training_data_filename, num_epochs, column):
path = "./result/" + result_filename
if not os.path.isdir(path):
os.mkdir(path)
newdirs = []
for root, dirs, files in walk(training_data_filename):
if files is not None:
for i in range(len(files)):
newdirs.append(root + '/' + files[i])
print(root + '/' + files[i])
x_train, y_train = read_excel(training_data_filename + '/' + files[0])
for i in range(1, len(newdirs)):
a1, b1 = read_excel(newdirs[i])
x_train = np.vstack((x_train, a1))
y_train = np.hstack((y_train, b1))
print(newdirs[i])
x_tem_train = []
y_tem_train = []
x_tem_validation = []
y_tem_validation = []
#x_newtest = np.array(x_test,dtype=float)
#y_newtest = np.array(y_test,dtype=float)
for i in range(len(x_train)):
if i % 5 == 0:
x_tem_validation.append(x_train[i])
y_tem_validation.append(y_train[i])
else:
x_tem_train.append(x_train[i])
y_tem_train.append(y_train[i])
x_new_train = np.array(x_tem_train, dtype=float)
y_new_train = np.array(y_tem_train, dtype=float)
x_new_validation = np.array(x_tem_validation, dtype=float)
y_new_validation = np.array(y_tem_validation, dtype=float)
#LeNet_Regressor(train_inputs, train_labels, val_inputs, val_labels,
# filepath, num_epoch, SensorType):
seqModel = LeNet_Regressor(x_new_train, y_new_train, x_new_validation,
y_new_validation, path + '/best_weight.hdf5',
num_epochs)
hist_df = pd.DataFrame(seqModel.history)
# or save to csv:
hist_csv_file = path + '/history.csv'
with open(hist_csv_file, mode='w') as f:
hist_df.to_csv(f)
def json_to_csv():
data = read_json(path, 'Modian_Summary.json')
with open("test.csv", "w") as csvfile:
writer = csv.writer(csvfile)
writer.writerow(
['Team', 'Name', 'Total Amount', 'Headcount', 'Average'])
htt, lyt = read_excel()
for team in data:
for name in data[team]:
if name == '黄婷婷':
total_amount = round(data[team][name]['总计金额'] + htt, 2)
else:
if name == '李艺彤':
total_amount = round(data[team][name]['总计金额'] + lyt, 2)
else:
total_amount = data[team][name]['总计金额']
headcount = data[team][name]['集资人数']
average = data[team][name]['人均集资']
line = [team, name, total_amount, headcount, average]
writer.writerow(line)
print('已更新test.csv文件')
def main():
excel_file = 'data_origin.xls'
result_file = 'result.xlsx'
data_frame = read_excel(excel_file=excel_file, sheet_name='Sheet1')
companies = set_company(data_frame)
write_excel(companies, file_name=result_file)
def load_model(file_path,
wol_ratio=0.1,
scale_obj_rxn_id='BIOMASS_SCALED',
scale_obj_mtb_id='Bio_unscaled_w',
verbose=False):
"""Loads the model from file_path, and then adjusts the wolbachia. The ratio of wolbachia / worm biomass in the objective function is set to wol_ratio, and all wolbachia reactions have their bounds scaled by the same ratio. If the model has been previously scaled, it is adjusted to the given ratio."""
model = read_excel(file_path, verbose=verbose)
set_model_wolbachia(model, wol_ratio, scale_obj_rxn_id, scale_obj_mtb_id)
return model
def __init__(self):
self.read_pdf = read_pdf()
self.read_excel = read_excel()
self.read_web = read_web()
self.create_folder = create_folder
self.choices = {
"1": self.create_folder,
"2": self.read_pdf.read_folder,
"3": self.read_excel.read_folder,
"4": self.read_web.read_folder,
"9": self.display_help,
"0": self.quit
}
def hello_world():
form = QueForm()
if request.method == 'GET':
db = ExeMysql()
mac = db.query()
all_coa = []
for ma in mac:
all_coa.append(ma[0])
return render_template('line-stack.html', form=form, mac=all_coa)
else:
f = request.files['file']
exl_data = read_excel(f)
db = ExeMysql()
db.insert(exl_data)
return render_template('line-stack.html', form=form)
if not i.endswith(".mat"):
continue
m = cobra.io.load_matlab_model(os.path.join("mat", i))
m.id = i[:-4]
if m.id in open_boundaries:
open_exchanges(m)
models.append(m)
# ### Some models are only available as Microsoft Excel files
# In[5]:
m = read_excel("xls/iJS747.xls",
verbose=False, rxn_sheet_header=7)
models.append(m)
# In[6]:
m = read_excel("xls/iRM588.xls",
verbose=False, rxn_sheet_header=5)
models.append(m)
# In[7]:
m = read_excel("xls/iSO783.xls", verbose=False, rxn_sheet_header=2)
models.append(m)
import pandas
from cobra import Model, Metabolite, Reaction
from read_excel import read_excel, write_excel
elegans_file = 'iCEL1273.xlsx'
model_in_file = 'model_b_mal.xlsx'
model_out_file = 'model_b_mal_2-wip.xlsx'
reversible_arrow = '<==>'
irreversible_arrow = ('-->', '<--')
l_bound, u_bound = -1000, 1000
old_m = read_excel(model_in_file)
pio = pandas.io.excel.ExcelFile(elegans_file)
rxn_frame = pandas.read_excel(elegans_file, 'Reactions')
def use_directionalities(old_m, disagreements):
unresolved = []
for r_id, cel_dir, m_dir, cel_products in disagreements:
updated = False
if cel_dir == 'reversible':
rxn = old_m.reactions.get_by_id(r_id)
rxn.bounds = (l_bound, u_bound)
updated = True
elif cel_dir == 'irreversible':
if ',' not in r_id: # combined reactions are handled manually
cel_prods = set()
for m in cel_products:
met = m.replace('M', 'C').replace('E', 'C')
reaction.upper_bound = max(reaction.upper_bound, 10)
else:
reaction.lower_bound = min(reaction.lower_bound, -10)
models.append(m)
#### Some models are only available as Microsoft Excel files
# In[6]:
from read_excel import read_excel
# In[7]:
m = read_excel("xls/iJS747.xls", verbose=False, rxn_sheet_header=7)
m.change_objective("agg_GS13m_2")
models.append(m)
# In[8]:
m = read_excel("xls/iRM588.xls", verbose=False, rxn_sheet_header=5)
m.change_objective("agg_GS13m")
models.append(m)
# In[9]:
m = read_excel("xls/iSO783.xls", verbose=False, rxn_sheet_header=2)
m.change_objective("Biomass")
'partial_wOv_manual_4.xlsx', 'partial_wBm_manual_4.xlsx'
]
in_wol_common_file = 'partial_wol_common_4.xlsx'
wol_model_names = ['model_wOv_4', 'model_wBm_4']
out_nem_model_names = ['model_o_vol_4', 'model_b_mal_4']
nem_dont_delete = set(['R00104', 'R00161', 'R05332'])
nem_xfer_to_wol = set(['R01195', 'R05636'])
# # # Run-time options
save_wol_models = True
save_nem_models = True
perform_gap_filling = 5 # False to not, otherwise a number indicating how many iterations. Performs gap-filling using all nematode reactions not already added to the wol model.
test_removed_wol_rxns = False # Check the effect of removing the wolbachia-only reactions
# # # Run steps
in_model_files = [os.path.join(files_dir, m_file) for m_file in in_model_files]
in_models = [read_excel(m_file, verbose=False) for m_file in in_model_files]
in_wol_unique_files = [
os.path.join(files_dir, m_file) for m_file in in_wol_unique_files
]
wol_unique_models = [
read_excel(m_file, verbose=False) for m_file in in_wol_unique_files
]
wol_common_model = read_excel(os.path.join(files_dir, in_wol_common_file))
fvas = []
for m in in_models:
optimize_minimal_flux(m)
fvas.append(flux_variability_analysis(m))
wol_models, unused_models, unique_wol_ids = separate_wol_rxns(
in_models, wol_unique_models, wol_common_model, wol_model_names,
test_removed_wol_rxns)
if rxn.id.startswith(rxn_pref_remove):
rxns_to_remove.append(rxn.id)
worm_m.remove_reactions(rxns_to_remove)
worm_m.id = out_name
write_excel(worm_m, out_file)
print('%s saved to %s' % (worm_m, out_file))
# # # Inputs
files_dir = '/mnt/hgfs/win_projects/brugia_project'
worm_model_files = ['model_b_mal_4-sep.xlsx', 'model_o_vol_4-sep.xlsx']
wol_model_files = ['model_wBm_4-sep.xlsx', 'model_wOv_4-sep.xlsx']
out_model_names = ['model_b_mal_4.5', 'model_o_vol_4.5']
# # # Run-time options
# # # Run steps
worm_model_files = [
os.path.join(files_dir, m_file) for m_file in worm_model_files
]
wol_model_files = [
os.path.join(files_dir, m_file) for m_file in wol_model_files
]
worm_models = [
read_excel(m_file, verbose=False) for m_file in worm_model_files
]
wol_models = [read_excel(m_file, verbose=False) for m_file in wol_model_files]
for worm_m, wol_m, out_name in zip(worm_models, wol_models, out_model_names):
out_file = os.path.join(files_dir, out_name + '-wip.xlsx')
merge_models(worm_m, wol_m, out_name, out_file)
print('Originally %s yielded %.2f' % (test_m, orig_f))
for rxn in good_m.reactions:
r_id = rxn.id
t_rxn = test_m.reactions.get_by_id(r_id)
good_bounds, test_bounds = rxn.bounds, t_rxn.bounds
if good_bounds != test_bounds:
t_rxn.bounds = good_bounds
new_f = test_m.optimize().f
if new_f - orig_f > 1.0:
print r_id, new_f, good_bounds, test_bounds
print_biomass_impact(r_id, test_m, test_fva, good_f)
t_rxn.bounds = test_bounds
if test_biomass_impact:
diffs = id_bottleneck_metabolites(test_m, orig_f, 'BIOMASS', threshold=1.0)
# producing_r_id = mtb.id.upper() # used to go a level deeper in biomass rxn.
files_dir = '/mnt/hgfs/win_projects/brugia_project'
good_model = 'model_b_mal_4.5-wip.xlsx'
test_model = 'model_b_mal_5_M30.xlsx'
test_biomass_impact = True
good_m = read_excel(os.path.join(files_dir, good_model))
test_m = read_excel(os.path.join(files_dir, test_model))
test_fva = cobra.flux_analysis.flux_variability_analysis(test_m)
compare_constraints(good_m, test_m, test_fva, test_biomass_impact)
from read_excel import read_excel
import random
dataset = input("file to be opened:")
sheet = input(f'sheet inside {dataset+".xlsx"} to be opened:')
dirt, max_q = read_excel(dataset + ".xlsx", sheet)
q_num = input(
f'There are {max_q} words, how many words you want to challenge? ')
score = 0
while (True):
ran_int = random.randint(0, int(q_num) - 1)
print(ran_int)
print(f'what is the hiragana of... {dirt[ran_int][0]}')
answer = input('Answer: ')
if answer == dirt[ran_int][1]:
print("Correct!")
score += 1
else:
print(f'Correct answer is {dirt[ran_int][1]}')
# task to do:
# 1. no repeat question
# 2. program end when gone through all the questions.
from matplotlib import pyplot as plt
x, height = [item[0].strftime('%m%d')
for item in data], [item[1] for item in data]
plt.figure(dpi=200)
plt.bar(x, height, width=0.35)
plt.title("Bar")
plt.xlabel("month")
plt.ylabel("month count")
plt.tick_params(axis='x', labelsize=9, rotation=50)
plt.show()
if __name__ == '__main__':
import read_excel
file = '/Users/admin/PycharmProjects/learn-python/python_coder_share/数据来源.xlsx'
# excel 数据
data = read_excel.read_excel(file, sheet='Sheet1')
print(data)
# data = [[datetime.datetime(2018, 12, 1, 0, 0), 227526, 16251], [datetime.datetime(2019, 1, 1, 0, 0), 263218, 22689],
# [datetime.datetime(2019, 2, 1, 0, 0), 189339, 9585], [datetime.datetime(2019, 3, 1, 0, 0), 116730, 9425],
# [datetime.datetime(2019, 4, 1, 0, 0), 383429, 21349], [datetime.datetime(2019, 5, 1, 0, 0), 247331, 12531],
# [datetime.datetime(2019, 6, 1, 0, 0), 181440, 14579], [datetime.datetime(2019, 7, 1, 0, 0), 236901, 16256],
# [datetime.datetime(2019, 8, 1, 0, 0), 200990, 15150]]
# print(data)
# x = [item[0].strftime('%m%d') for item in data]
# y = [item[1] for item in data]
# print(x, y)
data_plot(data)
pass
def run():
# # # Parameters
min_flux = -1000
max_flux = 1000
min_colour = '#ff0000'
zero_colour = '#ffff00'
max_colour = '#00ff00'
colour_range = (min_colour, zero_colour, max_colour)
# # # Run-time options
wolbachia_percent = 10.0
files_dir = '/mnt/hgfs/win_projects/brugia_project'
#model_files = ['model_bm_5_L3.xlsx', 'model_bm_5_L3D6.xlsx', 'model_bm_5_L3D9.xlsx', 'model_bm_5_L4.xlsx']
#model_files = ['model_bm_5_M30.xlsx', 'model_bm_5_F30.xlsx', 'model_bm_5_M42.xlsx', 'model_bm_5_F42.xlsx', 'model_bm_5_M120.xlsx', 'model_bm_5_F120.xlsx']
#model_files = ['model_bm_5_L3.xlsx', 'model_bm_5_M30.xlsx', 'model_bm_5_F30.xlsx', 'model_bm_5_M42.xlsx', 'model_bm_5_F42.xlsx', 'model_bm_5_M120.xlsx', 'model_bm_5_F120.xlsx']
#model_files = ['model_bm_6.xlsx', 'model_bm_6_lohg.xlsx', 'model_bm_6_holg.xlsx', 'model_bm_6_lolg.xlsx']
#model_files = ['model_bm_6_L3.xlsx', 'model_bm_6_L3D6.xlsx', 'model_bm_6_L3D9.xlsx', 'model_bm_6_L4.xlsx']
#model_files = ['model_bm_6_M30.xlsx', 'model_bm_6_F30.xlsx', 'model_bm_6_M42.xlsx', 'model_bm_6_F42.xlsx', 'model_bm_6_M120.xlsx', 'model_bm_6_F120.xlsx']
model_files = ['model_bm_6.xlsx']
mtb_cmp_str = '%s_wip.xlsx'
verbose = True
topology_analysis = False
fba_analysis = False
fva_analysis = True
save_visualizations = False
do_reaction_mtb_comparison = None # 'C00080', or None
test_nutrient_combos = False # False is don't. 2 tests all combos of length 2, etc.
show_wol_transports = False
model_files = [os.path.join(files_dir, m_file) for m_file in model_files]
models = [
load_model(m_file,
wol_ratio=wolbachia_percent / 100.0,
verbose=verbose) for m_file in model_files
]
set_nutrients = False
if set_nutrients:
#nutrients = [('CARBON_SOURCE', (0, 250)), ('DIFFUSION_2', (0, 580))] # HOHG
#nutrients = [('CARBON_SOURCE', (0, 45)), ('DIFFUSION_2', (0, 580))] # HOLG
#nutrients = [('CARBON_SOURCE', (0, 250)), ('DIFFUSION_2', (0, 90))] # LOHG
nutrients = [('CARBON_SOURCE', (0, 45)),
('DIFFUSION_2', (0, 90))] # LOLG
for model in models:
for name, bounds in nutrients:
model.reactions.get_by_id(name).bounds = bounds
if do_reaction_mtb_comparison:
new_model_files, new_models = [], []
cel_m = read_excel(os.path.join(files_dir, 'iCEL1273.xlsx'),
verbose=False)
cel_m.reactions.BIO0103.objective_coefficient = 1.0
for m_file, model in zip(model_files, models):
mtb_outfile = mtb_cmp_str % m_file.rpartition('.')[0]
m = model.copy()
compare_reaction_mtbs(cel_m, m, do_reaction_mtb_comparison,
mtb_outfile)
new_model_files.append(mtb_outfile)
new_models.append(read_excel(mtb_outfile))
model_files += new_model_files
models += new_models
viz_strs = [
os.path.split(m_file.rpartition('.')[0] + '_%s.txt')[1]
for m_file in model_files
]
for m in models:
cobra.flux_analysis.parsimonious.optimize_minimal_flux(m)
#m.optimize()
# rather than m.optimize(); returns optimal FBA with minimum total flux through network.
if topology_analysis:
for m in models:
basic_stats(m)
# # compare_models(m1, m2)
if verbose:
for m in models:
print('\n%s summary:' % m)
m.summary()
if fba_analysis:
for m, viz_str in zip(models, viz_strs):
analyze_shadows(m, 5)
if save_visualizations:
visualize_fba_reactions(m, (min_flux, max_flux),
colour_range,
to_file_str=viz_str)
# # compare_objective_functions(m1.reactions.get_by_id('BIOMASS'), m2.reactions.get_by_id('BIOMASS'))
fvas = []
if fva_analysis:
for m, viz_str in zip(models, viz_strs):
m_fva = cobra.flux_analysis.flux_variability_analysis(m)
fvas.append(m_fva)
if save_visualizations:
visualize_fva_reactions(m_fva, (min_flux, max_flux),
colour_range, viz_str)
active_rxn_mtb_analysis(models, fvas)
pathway_analysis(models, fvas, pathways_for_analysis)
if show_wol_transports:
wol_transport_analysis(models, fvas)
#assess_metabolites_impact(models[0], models[0].metabolites)
if test_nutrient_combos:
desc_str = '\nNutrient imports'
print('%s\n%s' % (desc_str, '-' * len(desc_str.strip())))
m, orig_fva = models[2], fvas[2]
bounds_deltas = (0, 60)
rxn_ids = ['NUTRIENTS_%i' % i for i in range(1, 21)]
rxn_combs = list(itertools.combinations(rxn_ids, test_nutrient_combos))
diffs = [
test_changed_constraints(r_ids, m, orig_fva, bounds_deltas)
for r_ids in rxn_combs
]
diffs = [d for d in diffs if d]
diffs.sort(key=lambda d: -d[0])
for d in diffs:
print(d)
return models, fvas
rxn.name = cel_rxn.name
rxn.bounds = cel_rxn.bounds
rxn.subsystem = cel_rxn.subsystem
rxn.enzyme_commission = cel_rxn.enzyme_commission
model.add_reaction(rxn)
rxn.build_reaction_from_string(cel_rxn.reaction)
return True
# # # Options
files_dir = '/mnt/hgfs/win_projects/brugia_project'
model_files = ['model_o_vol_3.xlsx', 'model_b_mal_3.xlsx']
out_files = ['model_o_vol_3.5-wip.xlsx', 'model_b_mal_3.5-wip.xlsx']
do_deletions = True
cel_m = read_excel(os.path.join(files_dir, 'iCEL1273.xlsx'), verbose=False)
#cel_m.reactions.BIO0101.objective_coefficient = 1.0 # # # TESTING ONLY
#cobra.flux_analysis.parsimonious.optimize_minimal_flux(cel_m)
models = [
read_excel(os.path.join(files_dir, m_file), verbose=False)
for m_file in model_files
]
for m, out_file in zip(models, out_files):
if do_deletions:
modify_model(m, cel_m, do_deletions)
write_excel(m, os.path.join(files_dir, out_file))
else:
print('Obj before modifications: %.1f' % (m.optimize().f))
rxns_to_delete = modify_model(m, cel_m, do_deletions)
diffs = []
"""Run on the models_V3.5. Adds mtb compartments, ensures names are transferred, and renames the GXXXXX ids.
"""
import os
from read_excel import read_excel, write_excel
files_dir = '/mnt/hgfs/win_projects/brugia_project'
in_model_names = ['model_o_vol_3.5', 'model_b_mal_3.5']
in_model_files = [os.path.join(files_dir, m_file+'.xlsx') for m_file in in_model_names]
out_model_files = [os.path.join(files_dir, m_file+'-wip.xlsx') for m_file in in_model_names]
in_models = [read_excel(m_file, verbose=False) for m_file in in_model_files]
for m in in_models:
for mtb in m.metabolites:
if mtb.id == 'FA_storage_mix': continue
elif mtb.id[0] == 'C':
mtb.compartment = 'c'
if mtb.id[1] == 'C': continue
elif mtb.id[0] == 'M':
mtb.compartment = 'm'
if mtb.id[1] == 'C' or mtb.name: continue
c_id = 'C' + mtb.id[1:]
if c_id in m.metabolites and m.metabolites.get_by_id(c_id).name:
mtb.name = m.metabolites.get_by_id(c_id).name
elif mtb.id[0] == 'G':
pass
for m, out_file in zip(in_models, out_model_files):
write_excel(m, out_file)
for i in sorted(os.listdir("mat")):
if not i.endswith(".mat"):
continue
m = cobra.io.load_matlab_model(os.path.join("mat", i))
m.id = i[:-4]
if m.id in open_boundaries:
open_exchanges(m)
models.append(m)
# ### Some models are only available as Microsoft Excel files
# In[5]:
m = read_excel("xls/iJS747.xls", verbose=False, rxn_sheet_header=7)
models.append(m)
# In[6]:
m = read_excel("xls/iRM588.xls", verbose=False, rxn_sheet_header=5)
models.append(m)
# In[7]:
m = read_excel("xls/iSO783.xls", verbose=False, rxn_sheet_header=2)
models.append(m)
# In[8]:
m = read_excel("xls/iCR744.xls", rxn_sheet_header=4, verbose=False)
print("please input excel file and sheet name, for example:")
print("python signal_fft.py \"imu_sample.xlsx\" \"static\" 5")
sys.exit(1)
print(sys.argv[0])
print(sys.argv[1])
print(sys.argv[2])
print(sys.argv[3])
print(type(sys.argv[0]))
print(type(sys.argv[1]))
print(type(sys.argv[2]))
print(type(sys.argv[3]))
#y=read_excel(r'imu_data.xlsx', '04', 0)
y = read_excel(sys.argv[1], sys.argv[2], int(sys.argv[3]))
n = len(y) # length of the signal
k = np.arange(n)
T = n / Fs
t = np.arange(0, T, Ts) # time vector,这里Ts也是步长
frq = k / T # two sides frequency range
frq1 = frq[range(int(n / 2))] # one side frequency range
YY = np.fft.fft(y) # 未归一化
Y = np.fft.fft(y) / n # fft computing and normalization 归一化
Y1 = Y[range(int(n / 2))]
fig, ax = plt.subplots(4, 1)
ax[0].plot(t, y)
ax[0].set_xlabel('Time')
ax[0].set_ylabel('Amplitude')
import read_excel as re
file_name = sys.argv[1] #excel文件全名
JENKINS_QUDAOS = sys.argv[2] #["HW","WX","GDT",2,-1];["XM","WX","GDT",1,-1]
IS_DEBUG_MODE = sys.argv[3] #true or false
if __name__ == "__main__":
if not os.path.exists(constant.base_path):
constant.base_path = os.path.abspath('')
constant.set()
# 开始清理apps下所有生成过的工程
appAbsPath = os.path.join(constant.get_base_path(), constant.apps_path)
if not os.path.exists(appAbsPath):
os.makedirs(appAbsPath)
else:
print 'app path exist'
excel_json = re.read_excel(constant.excel_file)
re.temp_write_out_file(excel_json)
copy_file.cpDirs(constant.old_path, constant.new_path)
builder = am.AndroidProjectMaker.Builder(
).set_excel_json(excel_json).set_base_path(
'%s/%s' %
(constant.get_base_path(),
constant.apps_path)).set_qudaos(JENKINS_QUDAOS).build(IS_DEBUG_MODE)
def predict(result_filename, testing_data_filename, model_name, weight,
column):
path = "./result/" + result_filename
newdirs = []
for root, dirs, files in walk(testing_data_filename):
if files is not None:
for i in range(len(files)):
newdirs.append(root + '/' + files[i])
print(root + '/' + files[i])
model = load_model(path + '/' + weight)
print(model.summary())
for i in range(len(newdirs)):
name_tmp = newdirs[i].split('/')
print(name_tmp)
name = name_tmp[2] + '_' + name_tmp[3] + '_' + name_tmp[4]
x_newtest, y_newtest = read_excel(newdirs[i])
train_inputs, d1 = convert_data(x_newtest, y_newtest, column)
num_event = np.size(train_inputs, 0)
num_order = np.size(train_inputs, 1)
num_sensor = np.size(train_inputs, 2)
train_inputs = np.reshape(train_inputs,
[num_event, 1, num_order, num_sensor, 1])
Y_pred = model.predict(train_inputs)
MAE = mean_absolute_error(d1, Y_pred)
MAE = '%.4f' % (MAE)
print(MAE)
# plot a picture
output = []
for m in range(len(Y_pred)):
output.append(Y_pred[m][0])
# d1[m] = d1[m]/100
nlist = np.arange(0, len(Y_pred))
fig, ax = plt.subplots()
ax.plot(nlist, output, label='predict')
ax.plot(nlist, d1, label='true')
ax.set(xlabel='time step',
ylabel='SOC(%)',
title=name + "\nmean absolute error =" + str(MAE))
ax.grid()
ax.legend()
fig.savefig(path + "/" + model_name + '_' + name + ".png")
#plt.show()
temp = []
temp.append(['predict', 'real'])
for m in range(len(output)):
temp.append([output[m], d1[m]])
with open(path + "/" + model_name + '_' + name + ".csv",
'w',
newline='') as csvfile:
writer = csv.writer(csvfile)
writer.writerows(temp)
with warnings.catch_warnings():
warnings.simplefilter("ignore")
m = cobra.io.read_legacy_sbml(filepath) if model_id in legacy_SBML else cobra.io.read_sbml_model(filepath)
m.id = m.description = model_id.replace(".", "_")
if m.id in open_boundaries:
open_exchanges(m)
models.append(m)
# ### Some models are only available as Microsoft Excel files
# In[4]:
m = read_excel("xls/iJS747.xls", verbose=False, rxn_sheet_header=7)
models.append(m)
# In[5]:
m = read_excel("xls/iRM588.xls", verbose=False, rxn_sheet_header=5)
models.append(m)
# In[6]:
m = read_excel("xls/iSO783.xls", verbose=False, rxn_sheet_header=2)
models.append(m)
