def avg():
df = pd.read_csv('maoyan2.csv', error_bad_lines=False)
df.round(2)
# 根据日期统计各天的平均值，并保留两位小数
date_score_avg = df.groupby('date')['score'].mean().round(2)*2
# 根据评分数据生成柱状图
bar = Bar('评分走势图', '数据来源：不正经程序员-采集自猫眼',
title_pos='center', width=1500, height=600)
# line = Line()
# line.add('', attr, value)
bar.add('', date_score_avg.index, date_score_avg.values, is_visualmap=False, visual_range=[0, 3500], visual_text_color='#fff', is_more_utils=True,
is_label_show=True, xaxis_interval=0, xaxis_rotate=30, mark_line=["average"])
overlap = Overlap()
overlap.add(bar)
# overlap.add(line)
overlap.show_config()
overlap.render(
'picture\评分走势图.html')
# 根据日期统计各天的平均值，并保留两位小数
date_positive_prob_avg = df.groupby('date')['positive_prob'].mean().round(2)
bar = Bar('评论情感指数走势图', '数据来源：不正经程序员-采集自猫眼',
title_pos='center', width=1500, height=600)
# line = Line()
# line.add('', attr, value)
bar.add('', date_positive_prob_avg.index, date_positive_prob_avg.values, is_visualmap=False, visual_range=[0, 3500], visual_text_color='#fff', is_more_utils=True,
is_label_show=True, xaxis_interval=0, xaxis_rotate=30, mark_line=["average"])
overlap = Overlap()
overlap.add(bar)
# overlap.add(line)
overlap.show_config()
overlap.render(
'picture\评论情感指数走势图.html')
def test_overlap_1():
v1 = [10, 20, 30, 40, 50, 60]
v2 = [30, 30, 30, 30, 30, 30]
v3 = [50, 50, 50, 50, 50, 50]
v4 = [10, 10, 10, 10, 10, 10]
es = EffectScatter("Scatter - EffectScatter 示例")
es.add("es", v1, v2)
scatter = Scatter()
scatter.add("scatter", v1, v3)
es_1 = EffectScatter()
es_1.add("es_1", v1, v4, symbol='pin', effect_scale=5)
overlap = Overlap()
overlap.add(es)
overlap.add(scatter)
overlap.add(es_1)
overlap.show_config()
overlap.render()
def render():
# 获取评论中所有城市
with open(r'C:\Users\think\Desktop\情感分析\doc\maoyan.csv',
mode='r',
encoding='utf_8_sig') as f:
rows = f.readlines()
#print(rows)
#num = 0
for row in rows[1:]:
#print(row)
#num = num + 1
#print(num)
'''
if num == 10:
break
'''
#print(row)
#print(row.count(','))
if row.count(',') != 7:
continue
elements = row.split(',')
#print(elements)
score = elements[6]
city = elements[4]
gender = elements[3]
if score != '':
scores.append(float(score) * 2)
# if float(score) * 2 > 7:
# positive_text += comment
# elif float(score) * 2 < 4:
# negative_text += comment
if city != '': # 去掉城市名为空的值
cities.append(city)
if gender != '':
genders.append(gender)
# 按0-10进行排序
#print(scores)
score_data = Counter(scores).most_common()
score_data = sorted(score_data)
gender_data = Counter(genders).most_common()
print(gender_data)
#print(score_data)
# 定义样式
style = Style(title_color='#fff',
title_pos='center',
width=800,
height=600,
background_color='#404a59')
# 根据评分数据生成柱状图
bar = Bar('《海上钢琴师》各评分数量',
'数据来源：采集自猫眼',
title_pos='center',
width=900,
height=600)
attr, value = bar.cast(score_data)
#print(value)
# line = Line()
# line.add('', attr, value)
bar.add('',
attr,
value,
is_visualmap=True,
visual_range=[0, 3500],
visual_text_color='#fff',
is_more_utils=True,
is_label_show=True)
overlap = Overlap()
overlap.add(bar)
# overlap.add(line)
overlap.show_config()
overlap.render(r'C:\Users\think\Desktop\情感分析\picture\评分数量-柱状图.html')
# 对城市数据和坐标文件中的地名进行处理
handle(cities)
data = Counter(cities).most_common() # 使用Counter类统计出现的次数，并转换为元组列表
#print(data)
# 根据城市数据生成地理坐标图
geo = Geo('观众地理分布', '数据来源：采集自猫眼', **style.init_style)
attr, value = geo.cast(data)
# print(attr)
# print(value)
geo.add('',
attr,
value,
visual_range=[0, 600],
maptype='china',
visual_text_color='#fff',
symbol_size=7,
is_visualmap=True,
is_piecewise=True,
visual_split_number=10)
geo.render(r'C:\Users\think\Desktop\情感分析\picture\观众地理分布-地理坐标图.html')
# 根据城市数据生成柱状图
data_top20 = Counter(cities).most_common(20) # 返回出现次数最多的20条
bar = Bar('观众来源排行TOP20',
'数据来源：采集自猫眼',
title_pos='center',
width=1200,
height=600)
attr, value = bar.cast(data_top20)
bar.add('',
attr,
value,
is_visualmap=True,
visual_range=[0, 3500],
visual_text_color='#fff',
is_more_utils=True,
is_label_show=True)
bar.render(r'C:\Users\think\Desktop\情感分析\picture\观众来源排行-柱状图.html')
#生成观众性别分布图
# 设置主标题与副标题，标题设置居中，设置宽度为900
pie = Pie("观众性别分布图", "数据来源：采集自猫眼", title_pos='center', width=900)
attr, value = geo.cast(gender_data)
print(value)
attr = ["其他", "男", "女"]
# 加入数据，设置坐标位置为【25，50】，上方的colums选项取消显示
'''
pie.add("", ["其他","男","女"], value ,visual_range=[0, 3500],
is_legend_show=False, is_label_show=True, is_more_utils=True)
'''
pie.add("", attr, value, is_label_show=True, is_more_utils=True)
bar.add('', attr, value, is_visualmap=False, visual_range=[0, 3500], visual_text_color='#fff', is_more_utils=True,
is_label_show=True)
bar.render('picture\观众来源排行-柱状图.html')
# 根据评分数据生成柱状图
bar = Bar('《海上钢琴师》各评分数量', '数据来源：不正经程序员-采集自猫眼',
title_pos='center', width=900, height=600)
attr, value = bar.cast(score_data)
# line = Line()
# line.add('', attr, value)
bar.add('', attr, value, is_visualmap=False, visual_range=[0, 3500], visual_text_color='#fff', is_more_utils=True,
is_label_show=True)
overlap = Overlap()
overlap.add(bar)
# overlap.add(line)
overlap.show_config()
overlap.render(
'picture\评分数量-柱状图.html')
# 根据评分数据生成柱状图
bar = Bar('评价人数走势图', '数据来源：不正经程序员-采集自猫眼',
title_pos='center', width=1200, height=600)
attr, value = bar.cast(date_data)
# line = Line()
# line.add('', attr, value)
bar.add('', attr, value, is_visualmap=False, visual_range=[0, 3500], visual_text_color='#fff', is_more_utils=True,
is_label_show=True)
overlap = Overlap()
overlap.add(bar)
# overlap.add(line)
overlap.show_config()
def render():
global positive_text
global negative_text
global luhan_positive_text
global luhan_negative_text
# 获取评论中所有城市
cities = []
with open('shanghai.csv', mode='r', encoding='utf-8') as f:
rows = f.readlines()
for row in rows[1:]:
if row.count(',') != 7:
continue
elements = row.split(',')
time = elements[0]
gender = elements[3]
city = elements[4]
comment = elements[7]
score = elements[6]
if not is_float(score):
continue
if '鹿晗' in comment:
scores_luhan.append(float(score) * 2)
if float(score) * 2 == 10:
luhan_positive_text += comment
elif float(score) * 2 == 1:
luhan_negative_text += comment
if city != '': # 去掉城市名为空的值
cities.append(city)
if score != '':
scores.append(float(score) * 2)
if float(score) * 2 == 10:
positive_text += comment
elif float(score) * 2 == 1:
negative_text += comment
if gender != '':
genders.append(gender)
if time != '':
dates.append(time)
# print(positive_text)
# print(negative_text)
with open("positive_text.txt", "w", encoding='utf-8') as f:
f.write(positive_text)
with open("negative_text.txt", "w", encoding='utf-8') as f:
f.write(negative_text)
with open("luhan_positive_text.txt", "w", encoding='utf-8') as f:
f.write(luhan_positive_text)
with open("luhan_negative_text.txt", "w", encoding='utf-8') as f:
f.write(luhan_negative_text)
# 对城市数据和坐标文件中的地名进行处理
handle(cities)
data = Counter(cities).most_common() # 使用Counter类统计出现的次数，并转换为元组列表
score_data = Counter(scores).most_common() # 使用Counter类统计出现的次数，并转换为元组列表
# 按0-10进行排序
score_data = sorted(score_data)
score_data_luhan = Counter(
scores_luhan).most_common() # 使用Counter类统计出现的次数，并转换为元组列表
# 按0-10进行排序
score_data_luhan = sorted(score_data_luhan)
gender_data = Counter(genders).most_common() # 使用Counter类统计出现的次数，并转换为元组列表
print(gender_data)
date_data = Counter(dates).most_common()
# 按日期进行排序
date_data = sorted(date_data)
# print(data)
# 定义样式
style = Style(title_color='#fff',
title_pos='center',
width=800,
height=600,
background_color='#404a59')
# 根据城市数据生成地理坐标图
geo = Geo('观众地理分布', '数据来源：不正经程序员-采集自猫眼', **style.init_style)
attr, value = geo.cast(data)
geo.add('',
attr,
value,
visual_range=[0, 600],
visual_text_color='#fff',
symbol_size=7,
is_visualmap=True,
is_piecewise=True,
visual_split_number=10)
geo.render('picture\观众地理分布-地理坐标图.html')
# 根据城市数据生成柱状图
data_top20 = Counter(cities).most_common(20) # 返回出现次数最多的20条
bar = Bar('观众来源排行TOP20',
'数据来源：不正经程序员-采集自猫眼',
title_pos='center',
width=1200,
height=600)
attr, value = bar.cast(data_top20)
bar.add('',
attr,
value,
is_visualmap=False,
visual_range=[0, 3500],
visual_text_color='#fff',
is_more_utils=True,
is_label_show=True)
bar.render('picture\观众来源排行-柱状图.html')
# 根据评分数据生成柱状图
bar = Bar('各段评分数量',
'数据来源：不正经程序员-采集自猫眼',
title_pos='center',
width=900,
height=600)
attr, value = bar.cast(score_data)
# line = Line()
# line.add('', attr, value)
bar.add('',
attr,
value,
is_visualmap=True,
visual_range=[0, 3500],
visual_text_color='#fff',
is_more_utils=True,
is_label_show=True)
overlap = Overlap()
overlap.add(bar)
# overlap.add(line)
overlap.show_config()
overlap.render('picture\评分数量-柱状图.html')
# 根据评分数据生成柱状图
bar = Bar('评论带有鹿晗的各段评分数量',
'数据来源：不正经程序员-采集自猫眼',
title_pos='center',
width=900,
height=600)
attr, value = bar.cast(score_data_luhan)
# line = Line()
# line.add('', attr, value)
bar.add('',
attr,
value,
is_visualmap=True,
visual_range=[0, 3500],
visual_text_color='#fff',
is_more_utils=True,
is_label_show=True)
overlap = Overlap()
overlap.add(bar)
# overlap.add(line)
overlap.show_config()
overlap.render('picture\评论带有鹿晗的各段评分数量-柱状图.html')
# 根据评分数据生成柱状图
bar = Bar('评价人数走势图',
'数据来源：不正经程序员-采集自猫眼',
title_pos='center',
width=1200,
height=600)
attr, value = bar.cast(date_data)
# line = Line()
# line.add('', attr, value)
bar.add('',
attr,
value,
is_visualmap=True,
xaxis_rotate=30,
visual_range=[0, 3500],
visual_text_color='#fff',
is_more_utils=True,
is_label_show=True)
overlap = Overlap()
overlap.add(bar)
# overlap.add(line)
overlap.show_config()
overlap.render('picture\评价人数走势图.html')
from pyecharts import Pie
# 设置主标题与副标题，标题设置居中，设置宽度为900
pie = Pie("观众性别分布图", "数据来源：不正经程序员-采集自猫眼", title_pos='center', width=900)
attr, value = geo.cast(gender_data)
# 加入数据，设置坐标位置为【25，50】，上方的colums选项取消显示
pie.add("", ["其他", "男", "女"],
value,
visual_range=[0, 3500],
is_legend_show=False,
is_label_show=True)
# 保存图表
pie.render('picture\观众性别分布图.html')
