def get_best_line(path="data/data.csv"):
df = pd.read_csv(path)
X, y = [], []
country_name, _, country_type = get_country_info()
for i in range(len(df["Country"])):
if df.iloc[i][1].strip() in country_name:
name = df.iloc[i][1].strip()
X.append([df.iloc[i][2], df.iloc[i][3]])
y.append(country_type[country_name.index(name)])
LME, CME = {}, {}
for i in range(len(X)):
point = X[i]
# The conutry is CME:
if y[i] == 1:
if point[0] not in CME:
CME[point[0]] = [point[1]]
else:
CME[point[0]].append(point[1])
# The conutry is CME:
else:
if point[0] not in LME:
LME[point[0]] = [point[1]]
else:
LME[point[0]].append(point[1])
for keys in LME:
lst = LME[keys]
LME[keys] = sum(lst) / len(lst)
for keys in CME:
lst = CME[keys]
CME[keys] = sum(lst) / len(lst)
print(LME)
print(CME)
def init_output(df, yr_list, country_list):
# Init a Dataframe with all NaN.
for i in range(len(yr_list) * (len(country_list))):
df1 = pd.DataFrame([["NaN"] * len(df.columns)], columns=df.columns)
df = df1.append(df, ignore_index=True)
_, _, country_type = get_country_info()
# Write Country and Years into the column
for i in range(len(country_list)):
for j in range(len(yr_list)):
pos = i * len(yr_list) + j
df.iloc[pos][0] = country_list[i]
df.iloc[pos][1] = int(yr_list[j])
df.iloc[pos][2] = country_type[i]
return df
def get_target_country(
path="/Users/jinhanmei/Desktop/PATSTAT_Sample_2018Autumn_Global_CSV/tls801_country.csv"
):
country, _, country_type = get_country_info()
country = country[:country_type.index(0)]
df = pd.read_csv(path, sep=None)
titles = df.columns.ravel().tolist()
ctry_code, ctry_name = df[titles[0]].tolist(), df[titles[2]].tolist()
r_lst = []
for i in range(len(ctry_name)):
if ctry_name[i] in country:
r_lst.append(ctry_code[i])
# Add 'US' in the r_lst.
r_lst.append('US')
print(r_lst)
return r_lst
def check_countries_application(path):
_, _, all_target_country = get_country_info()
df = pd.read_csv(path, sep=None)
titles = df.columns.ravel().tolist()
country_dict = {}
target_ctry_code = get_target_country()
for i in range(len(df[titles[1]])):
# Check if the patent yr is in range:
if int(df[titles[5]][i]) >= 2000:
# Check if the patent is granted
if df[titles[20]][i] == "Y":
# Check if the patent country is wanted:
if df[titles[1]][i] in target_ctry_code:
if df[titles[1]][i] not in country_dict:
country_dict[df[titles[1]][i]] = 1
else:
country_dict[df[titles[1]][i]] += 1
print(country_dict)
count = 0
for keys in country_dict:
count += country_dict[keys]
def get_plot(path="data/data.csv"):
df = pd.read_csv(path)
X, y = [], []
country_name, _, country_type = get_country_info()
print(country_name)
for i in range(len(df["Country"])):
if df.iloc[i][1].strip() in country_name:
name = df.iloc[i][1].strip()
X.append([df.iloc[i][2], df.iloc[i][3]])
y.append(country_type[country_name.index(name)])
X = np.asarray(X)
plt.title("LME CME score through Change")
plt.scatter(X[:, 0],
X[:, 1],
color=[
"r" if y_point == -1 else "b" if y_point == 1 else "g"
for y_point in y
],
s=20,
alpha=0.5)
plt.show()