def test_verbose(self):
qs = TradeLog.objects.all()
df = read_frame(qs, verbose=True)
self.assertListEqual(
list(qs.values_list('trader__name', flat=True)),
df.trader.tolist()
)
df1 = read_frame(qs, verbose=False)
self.assertListEqual(
list(qs.values_list('trader__pk', flat=True)),
df1.trader.tolist()
)
# Testing verbose with annotated column:
if django.VERSION >= (1, 10):
from django.db.models import F, FloatField
from django.db.models.functions import Cast
qs1 = TradeLog.objects.all().annotate(
total_sum=Cast(F('price') * F('volume'), FloatField()),
)
df2 = read_frame(
qs1, fieldnames=['trader', 'total_sum'])
self.assertListEqual(
list(qs1.values_list('total_sum', flat=True)),
df2.total_sum.tolist()
)
self.assertListEqual(
list(qs1.values_list('trader__name', flat=True)),
df2.trader.tolist()
)
def test_basic(self):
qs = MyModel.objects.all()
df = read_frame(qs)
n, c = df.shape
self.assertEqual(n, qs.count())
fields = MyModel._meta.get_all_field_names()
self.assertEqual(c, len(fields))
df1 = read_frame(qs, ['col1', 'col2'])
self.assertEqual(df1.shape, (qs.count(), 2))
def test_verbose(self):
qs = TradeLog.objects.all()
df = read_frame(qs, verbose=True)
self.assertListEqual(
list(qs.values_list('trader__name', flat=True)),
df.trader.tolist()
)
df1 = read_frame(qs, verbose=False)
self.assertListEqual(
list(qs.values_list('trader__pk', flat=True)),
df1.trader.tolist()
)
def test_basic(self):
qs = MyModel.objects.all()
df = read_frame(qs)
n, c = df.shape
self.assertEqual(n, qs.count())
if django.VERSION < (1, 10):
fields = MyModel._meta.get_all_field_names()
else:
fields = [f.name for f in MyModel._meta.get_fields()]
self.assertEqual(c, len(fields))
df1 = read_frame(qs, ['col1', 'col2'])
self.assertEqual(df1.shape, (qs.count(), 2))
def test_index(self):
qs = MyModel.objects.all()
df = read_frame(qs, ['col1', 'col2', 'col3', 'col4'],
index_col='index_col')
self.assertEqual(df.shape, (qs.count(), 4))
self.assertEqual(set(df.index.tolist()),
set(qs.values_list('index_col', flat=True)))
def export_to_xls_users(self, request, queryset):
df = pd.DataFrame()
for el in queryset:
users = el.get_filtered_user_queryset()
users = users.annotate(userlogins=Count('usersession'))
users = users.annotate(usertime=Sum('usersession__duration'))
df = pd.concat([df, read_frame(users,
index_col='user__id',
verbose=True,
fieldnames=(
'user__id', 'user__username', 'user__first_name', 'user__last_name',
'patronimyc',
'user__email', 'user__last_login', 'userlogins', 'usertime',
'date_of_birth', 'age', 'gender', 'city__name', 'area__name',
'score', 'total_score',
'total_exchange',
'speciality__name',
'work__name', 'curing_form', 'position__name', 'category',
'bed_quantity',
'patient_quantity',
'main_phone', 'secondary_phone', 'work_phone',
'district', 'house', 'flat', 'index',))])
df = translate_column_names(df)
return export_to_xls(df, translit(el.name) + '.xls', engine='openpyxl')
def get_openings_df(from_date, to_date, freq='H'):
query = {}
if from_date is not None:
query['time__gte'] = from_date
if to_date is not None:
query['time__lt'] = to_date
# Grab openings in a pandas dataframe
df = read_frame(SpaceStatus.objects.filter(**query))
start = pd.to_datetime(from_date).tz_localize('Europe/Brussels')
end = pd.to_datetime(to_date).tz_localize('Europe/Brussels')
# No records found, just return a fake df, always closed
if len(df) == 0:
df = pd.DataFrame([
{'time': start, 'is_open': 0},
{'time': end, 'is_open': 0}
])
# Drop duplicate time index
df = df.groupby(df.time).last()
start = start.replace(minute=0, second=0, microsecond=0)
end = end.replace(minute=0, second=0, microsecond=0)
# Reindex on a monotonic time index
index = pd.date_range(start=start, end=end, freq=freq)
df = df.reindex(index, method='ffill')
nans = np.isnan(df.is_open.astype(np.float64))
df[nans] = not df[~nans].is_open[0]
return df
def test_basic(self):
qs = MyModel.objects.all()
df = read_frame(qs)
n, c = df.shape
self.assertEqual(n, qs.count())
from itertools import chain
if django.VERSION < (1, 10):
fields = MyModel._meta.get_all_field_names()
else:
fields = list(set(chain.from_iterable((field.name, field.attname) if hasattr(field, 'attname') else (field.name,)
for field in MyModel._meta.get_fields()
if not (field.many_to_one and field.related_model is None)
)))
self.assertEqual(c, len(fields))
df1 = read_frame(qs, ['col1', 'col2'])
self.assertEqual(df1.shape, (qs.count(), 2))
def test_verbose_duplicates_fieldnames(self):
qs = TradeLog.objects.all()
df = read_frame(qs, fieldnames=['trader', 'trader', 'price'])
self.assertListEqual(
list(qs.values_list('price', flat=True)),
df.price.tolist()
)
def export_to_xls_user_poll(self, request, queryset):
uf = queryset[0]
users = uf.get_filtered_user_queryset()
if not uf.polls:
uf.polls = Poll.objects.all()
polls = UsersPoll.objects.filter(user__pk__in=users.values_list('pk'),
poll__pk__in=uf.polls.values_list('pk'), )
if uf.date_from:
polls = polls.filter(date_passed__lte=uf.date_from)
if uf.date_to:
polls = polls.filter(date_passed__gte=uf.date_to)
polls = Poll.objects.filter(pk__in=polls.values_list('poll'))
answers = UserAnswer.objects.filter(user__pk__in=users.values_list('pk'),
question__poll__pk__in=polls.values_list('pk'))
answers = answers.order_by(uf.order, set(ORDERS - set(uf.order)).pop(), 'question__id', 'answer__id')
df = read_frame(answers,
verbose=True,
fieldnames=(
'user__user__id', 'user__user__username',
'question__poll__id', 'question__poll__name',
'question__id', 'question__question_text',
'answer__id', 'answer__answer_text',
'other_answer',
))
df = translate_column_names(df)
return export_to_xls(df, translit(uf.group.name) + '.xls', engine='openpyxl')
def poll_to_xls(request, queryset):
qs = None
for poll in queryset:
if qs:
qs |= Answer.objects.filter(question__poll=poll.id)
else:
qs = Answer.objects.filter(question__poll=poll.id)
qs = qs.annotate(useranswers=Count('useranswer'))
qs = qs.order_by('-question__poll__id', 'question__id', 'id')
df = read_frame(qs,
verbose=True,
fieldnames=('question__poll__id',
'question__poll__name',
'question__id',
'question__question_text',
'id',
'answer_text',
'useranswers'
))
for i, row in df.iterrows():
question = df.loc[i, 'question__id']
answers = df.loc[i, 'useranswers']
total_counts = sum(df[df['question__id'] == question]['useranswers'])
if total_counts:
percent = answers / total_counts * 100
else:
percent = 0
df.ix[i, 'percent'] = str(percent) + '%'
i = dt.datetime.now()
path = 'oprosy' + "_%s_%s_%s" % (i.day, i.month, i.year) + '.xls'
df = translate_column_names(df)
return export_to_xls(df, path)
def to_dataframe(self, fieldnames=(), verbose=True, index=None,
coerce_float=False):
"""
Returns a DataFrame from the queryset, overrides the base method to
enalbe lazy calculation
:param fieldnames: The model field names(columns) to utilise in
creating the DataFrame. You can span a relationships in
the usual Django ORM way by using the foreign key field
name separated by double underscores and refer to a field
in a related model.
:param index: specify the field to use for the index. If the index
field is not in fieldnames it will be appended. This
is mandatory for timeseries.
:param verbose: If this is ``True`` then populate the DataFrame with
the human readable versions for foreign key fields else
use the actual values set in the model
"""
if self.count_invalid() < self.MAX_NUM_CALCULATION:
outdated_values = self.filter(_is_valid=False)
outdated_values.recalculation() #pylint: disable=E1101
read_calc_value = '_calc_value' in fieldnames
if 'value' in fieldnames:
fieldnames[fieldnames.index('value')] = '_calc_value'
frame = read_frame(self, fieldnames=fieldnames, verbose=verbose,
index_col=index, coerce_float=coerce_float)
if '_calc_value' in frame.columns and not read_calc_value:
new_labels = list(frame.columns)
new_labels[new_labels.index('_calc_value')] = 'value'
frame.columns = new_labels
return frame
def similar_products2(deep_f):
qs = Product.objects.all()
df=read_frame(qs)
df['idx'] = range(1, len(df) + 1)
feature_list=[]
asin_list=[]
for prod in qs:
feature_list.append(prod.get_features())
asin_list.append(prod.asin)
nparray = np.asarray(feature_list)
#print nparray
tree = BallTree(nparray)
dist, ind = tree.query(deep_f, k=5)
print ind
index = ind[0]
recom = index[0:]
recommended_asins =[];
for i in recom:
recommended_asins.append(asin_list[i])
recommended_prods = Product.objects.filter(asin__in = recommended_asins)
return recommended_prods
# image_train = graphlab.SFrame(data=df)
# cur_prod = image_train[18:19]
# print cur_prod
# print image_train
# knn_model = graphlab.nearest_neighbors.create(image_train, features = ['features'],label = 'asin',distance = 'levenshtein',method = 'ball_tree')
# knn_model.save('my_knn')
# #knn_model= graphlab.load_model('my_knn')
# #print knn_model.query(cur_prod)
# #knn_model = graphlab.nearest_neighbors.create(image_train, features = ['features'],label = 'keywords')
def similar_products(product): qs = Product.objects.all() df=read_frame(qs) df['idx'] = range(1, len(df) + 1) feature_list=[] asin_list=[] product_index = 0 inn=0 for prod in qs: feature_list.append(prod.get_features()) asin_list.append(prod.asin) if prod.asin == product.asin: product_index = inn inn+=1 nparray = np.asarray(feature_list) #print nparray tree = BallTree(nparray) dist, ind = tree.query(nparray[product_index], k=5) print ind index = ind[0] recom = index[1:] recommended_asins =[]; for i in recom: recommended_asins.append(asin_list[i]) recommended_prods = Product.objects.filter(asin__in = recommended_asins) return recommended_prods
def test_duplicate_annotation(self):
qs = MyModel.objects.all()
qs = qs.values('index_col')
qs = qs.annotate(col1=Sum('col1'))
qs = qs.values()
df = read_frame(qs)
self.assertEqual(list(df.columns),
['id', 'index_col', 'col1', 'col2', 'col3', 'col4'])
def dataset_detail(request, dataset_id):
dataset = get_object_or_404(DataSet, pk=dataset_id)
DataSetModel = getattr(shop.models, dataset.code)
qs = DataSetModel.objects.all()
df = read_frame(qs)
df.to_pickle("dataset/{}".format(dataset.code))
datahtml = df.to_html()
return render(request, 'shop/dataset/detail.html', {'dataset': dataset, 'datavalues': datahtml})
def test_related_cols(self):
qs = TradeLog.objects.all()
cols = ["log_datetime", "symbol", "symbol__isin", "trader__name", "price", "volume", "note__note"]
df = read_frame(qs, cols, verbose=False)
self.assertEqual(df.shape, (qs.count(), len(cols)))
self.assertListEqual(list(qs.values_list("symbol__isin", flat=True)), df.symbol__isin.tolist())
self.assertListEqual(list(qs.values_list("trader__name", flat=True)), df.trader__name.tolist())
def test_values(self):
qs = MyModel.objects.all()
qs = qs.extra(select={"ecol1": "col1+1"})
qs = qs.values("index_col", "ecol1", "col1")
qs = qs.annotate(scol1 = Sum("col1"))
df = read_frame(qs)
self.assertEqual(list(df.columns), ['index_col', 'col1', 'scol1', 'ecol1'])
self.assertEqual(list(df["col1"]), list(df["scol1"]))
def test_choices(self):
MyModelChoice.objects.create(col1=1, col2=9999.99)
MyModelChoice.objects.create(col1=2, col2=0.99)
MyModelChoice.objects.create(col1=3, col2=45.6)
MyModelChoice.objects.create(col1=2, col2=2.6)
qs = MyModelChoice.objects.all()
df = read_frame(qs, verbose=True)
self.assertEqual(df.col1[0], 'First')
self.assertEqual(df.col1[1], 'Second')
self.assertEqual(df.col1[2], 'Third')
self.assertEqual(df.col1[3], 'Second')
df = read_frame(qs, verbose=False)
self.assertEqual(df.col1[0], 1)
self.assertEqual(df.col1[1], 2)
self.assertEqual(df.col1[2], 3)
self.assertEqual(df.col1[3], 2)
def get_openings_df(freq='H', **filter_args):
# Grab openings in a pandas dataframe
df = read_frame(SpaceStatus.objects.filter(**filter_args))
# Drop duplicate time index
df = df[df.time.diff().dt.total_seconds() > 0]
df.index = df.time
# Reindex on a monotonic hourly time index
index = pd.date_range(start=df.time.min(), end=df.time.max(), freq=freq)
return df.reindex(index, method='pad')
def test_many_to_many(self):
qs = Portfolio.objects.all()
cols = ['name', 'securities__symbol', 'securities__tradelog__log_datetime']
df = read_frame(qs, cols, verbose=True)
denormalized = Portfolio.objects.all().values_list(*cols)
self.assertEqual(df.shape, (len(denormalized), len(cols)))
for idx, row in enumerate(denormalized):
self.assertListEqual(
df.iloc[idx].tolist(),
list(row)
)
def get_user_data(username):
"""Get user's personal data."""
dfs = {}
df = read_frame(Video.objects.all(),
fieldnames=['id', 'video_id'] + VIDEO_FIELDS)
dfs['all_video_ratings'] = df
df = read_frame(
VideoRating.objects.filter(user__user__username=username).all())
dfs['my_video_scores'] = df
df = read_frame(
ExpertRating.objects.filter(user__user__username=username).all())
dfs['my_expert_ratings'] = df
df = read_frame(
VideoReports.objects.filter(user__user__username=username).all())
df.columns = [
VIDEO_REPORT_FIELDS[x] if x in VIDEO_REPORT_FIELDS else x
for x in df.columns
]
dfs['my_video_reports'] = df
df = read_frame(
VideoComment.objects.filter(user__user__username=username).all())
dfs['my_video_comments'] = df
df = read_frame(
VideoCommentMarker.objects.filter(user__user__username=username).all(),
fieldnames=['id', 'user', 'comment__id', 'user', 'which'])
dfs['my_video_comment_markers'] = df
return dfs
def read_ligand_batches(ligand_batch):
object_capitalized = ligand_batch.capitalize()
model = get_model_by_name(object_capitalized)
if ligand_batch in ["peptideBatch","antibodyBatch", "complexBatch"]:
df = read_frame(model.objects.all(), ["sid", "labeling", "concentration","buffer", "ph","purity", "produced_by__username",
"production_date", "comment",
"ligand__sid"])
df.replace([np.NaN], [None], inplace=True)
df['purity'] = list(map(str, df['purity'].values))
df['concentration'] = list(map(str, df['concentration'].values))
elif ligand_batch == "virusBatch":
df = read_frame(model.objects.all(),
["sid", "labeling", "concentration","buffer", "ph","purity", "produced_by__username",
"production_date", "comment",
"ligand__sid", "passage_history","active"])
df.replace([np.NaN], [None], inplace=True)
df['active'] = list(map(str, df['active'].values))
df['passage_history'] = list(map(str, df['passage_history'].values))
df['purity'] = list(map(str, df['purity'].values))
df['concentration'] = list(map(str, df['concentration'].values) )
elif ligand_batch == "bufferBatch":
df = read_frame(model.objects.all(),
["sid", "buffer", "ph", "produced_by__username",
"production_date", "comment"])
df.replace([np.NaN], [None], inplace=True)
df['ph']=list(map(str, df['ph'].values))
df['production_date']=list(map(str, df['production_date'].values))
df.replace(["None"],[None], inplace=True)
df = df.rename(columns={"ligand__sid":"ligand","produced_by__username":"******"})
return df
def anticodon_counts(request, clade_txid, isotype):
try:
tax_qs = models.Taxonomy.objects.filter(taxid=clade_txid).values(
'name', 'rank', 'taxid', 'domain')
if len(tax_qs) > 1: tax_qs = tax_qs.exclude(rank='assembly')
tax = tax_qs.get()
if tax['rank'] == 'class': tax['rank'] = 'taxclass'
trna_qs = models.tRNA.objects.filter(Q(**{tax['rank']: tax['taxid']}))
if isotype != 'All':
trna_qs = trna_qs.filter(isotype=isotype)
trna_qs = trna_qs.values('anticodon',
'isotype').annotate(clade=Count('anticodon'))
clade_counts = read_frame(trna_qs)
clade_counts = clade_counts.set_index(['isotype', 'anticodon'])
domain_name = models.Taxonomy.objects.filter(
taxid=tax['domain']).get().name
trna_qs = models.tRNA.objects.filter(domain=tax['domain'])
if isotype != 'All':
trna_qs = trna_qs.filter(isotype=isotype)
trna_qs = trna_qs.values('anticodon',
'isotype').annotate(domain=Count('anticodon'))
domain_counts = read_frame(trna_qs)
domain_counts = domain_counts.set_index(['isotype', 'anticodon'])
counts = clade_counts.join(domain_counts).sort_index().reset_index()
counts.columns = ['Isotype', 'Anticodon', tax['name'], domain_name]
counts = counts.set_index(['Isotype', 'Anticodon'])
counts.index.names = [None, None]
return HttpResponse(
counts.to_html(classes='table',
border=0,
justify='left',
bold_rows=False,
na_rep='0',
sparsify=True))
except:
return HttpResponse('Unknown server error')
def get_database_as_pd():
"""Get database tables as pandas DataFrames."""
result_df = {}
df = read_frame(Video.objects.all(),
fieldnames=[
'id', 'video_id', 'name', 'duration', 'language',
'publication_date', 'views', 'uploader'
] + VIDEO_FIELDS)
result_df['videos'] = df
df = read_frame(VideoRating.objects.all())
result_df['video_ratings'] = df
df = read_frame(ExpertRating.objects.all())
result_df['expert_rating'] = df
df = read_frame(DjangoUser.objects.all(),
fieldnames=[
'id', 'last_login', 'is_superuser', 'username',
'first_name', 'last_name', 'is_staff', 'is_active',
'date_joined', 'email', 'userpreferences__id'
] + [f"userpreferences__{x}" for x in VIDEO_FIELDS])
result_df['users'] = df
df = read_frame(VideoComment.objects.all())
result_df['video_comment'] = df
df = read_frame(VideoCommentMarker.objects.all(),
fieldnames=['id', 'user', 'comment__id', 'user', 'which'])
result_df['video_comment_marker'] = df
return result_df
def scoreJson(request):
print (request.body)
qs = rankings.objects.all()
df = read_frame(qs)
df['weighted_points'] = df['cur_year_avg_weighted'] + df['two_year_ago_weighted'] + df['three_year_ago_weighted']
data = json.loads(request.body)
x = data['Home'].capitalize()
y = data['Away'].capitalize()
home = df.loc[df['country_full'] == x]
away = df.loc[df['country_full'] == y]
frames = [home, away]
result = pd.concat(frames)
if x in df['country_full'].values and y in df['country_full'].values:
X_test3 = ['average_rank', 'rank_difference', 'point_difference', 'is_stake']
row = pd.DataFrame(np.array([[np.nan, np.nan, np.nan, True]]), columns=X_test3)
home_rank = result['rank'].iloc[0]
home_points = result['weighted_points'].iloc[0]
opp_rank = result['rank'].iloc[1]
opp_points = result['weighted_points'].iloc[1]
row['average_rank'] = (home_rank + opp_rank) / 2
row['rank_difference'] = home_rank - opp_rank
row['point_difference'] = home_points - opp_points
home_win = model.predict_proba(row)
print(home_win)
#dataF = pd.DataFrame({'x':data})
#print(dataF)
#print(df)
if home_win[0][0]>=0.45 and home_win[0][0]<=0.55:
r="draw"
print(r)
return JsonResponse({'result':r})
if home_win[0][0]>0.55:
r = result['country_full'].iloc[1], "win with a probability of",home_win[0][0]
print(r)
return JsonResponse({'result':r})
if home_win[0][1]>0.55:
r = result['country_full'].iloc[0], "win with a probability of",home_win[0][1]
print(r)
return JsonResponse({'result':r})
else:
return JsonResponse({'result':'Error team not found!'})
def json(request):
qs = Stock.objects.all()
df = read_frame(qs)
df2 = df.groupby(['symbol'])
# d1 = df2['symbol']
d = ["hhh", "hhhh", "kkkk"]
d1 = df2.to_dict()
context = {
'latest_question_list': "error",
}
template = loader.get_template('stock/json.html')
return JsonResponse({'foo': d1})
def get_ratings_dataframe(self, run):
rating_qs = self.filter(run=run).values("id", "parent_network__pk", "log_gamma", "log_gamma_uncertainty", "log_gamma_game_count").all()
rating = read_frame(rating_qs)
rating = rating.set_index("id")
rating = rating.sort_index()
rating["parent_network__pk"] = rating["parent_network__pk"].fillna(-1)
rating["log_gamma"] = rating["log_gamma"].fillna(0)
rating["log_gamma_uncertainty"] = rating["log_gamma_uncertainty"].fillna(0)
rating["log_gamma_uncertainty"] = rating["log_gamma_uncertainty"].replace([np.inf, -np.inf], 0)
rating["log_gamma_game_count"] = rating["log_gamma_game_count"].fillna(0)
return rating
def MakeDf(self):
"""[summary] make pd.DataFrame(df) from candles, df.rows=['time', 'open', 'high', 'low', 'close', 'volume'], df.index = df.time, where df.time.dtype = datetime64[ns].len(df) = self.num_data
Returns:
[type]: [description] pd.DataFrame
"""
df = read_frame(
self.datas.objects.filter(
product_code=self.product_code).all().order_by("time"))
df = df[["time", 'open', 'high', 'low', 'close', 'volume']]
df = df.set_index("time")
df = df[-self.num_data:]
return df
def publication_tags(self):
if self._publication_tags is None:
self._publication_tags = read_frame(
PublicationTags.objects.filter(
publication__in=self.publication_queryset),
fieldnames=['publication__id', 'tag__id', 'tag__name'],
index_col='publication__id')
self._publication_tags.rename(columns={
'tag__name': 'name',
'tag__id': 'related__id'
},
inplace=True)
return self._publication_tags
def procesa_tabla(tabla):
df= read_frame(tabla)
for item in tabla:
print(item.temperatura)
for item in tabla:
print(item.color)
for item in tabla:
print(item.inflamacion)
print(df)
return "Mi proceso"
def test_related_cols(self):
qs = TradeLog.objects.all()
cols = [
'log_datetime', 'symbol', 'symbol__isin', 'trader__name', 'price',
'volume', 'note__note'
]
df = read_frame(qs, cols, verbose=False)
self.assertEqual(df.shape, (qs.count(), len(cols)))
self.assertListEqual(list(qs.values_list('symbol__isin', flat=True)),
df.symbol__isin.tolist())
self.assertListEqual(list(qs.values_list('trader__name', flat=True)),
df.trader__name.tolist())
def get_user_data(username):
"""Get user's personal data."""
dfs = {}
df = read_frame(
VideoRating.objects.filter(user__user__username=username).all())
dfs['my_video_scores'] = df
df = read_frame(
ExpertRating.objects.filter(user__user__username=username).all())
dfs['my_expert_ratings'] = df
df = read_frame(
VideoComment.objects.filter(user__user__username=username).all())
dfs['my_video_comments'] = df
df = read_frame(
VideoCommentMarker.objects.filter(user__user__username=username).all(),
fieldnames=['id', 'user', 'comment__id', 'user', 'which'])
dfs['my_video_comment_markers'] = df
return dfs
def index(request):
boyds = BlackOystercatcher.objects.all()
df = read_frame(boyds)
ids = df['bird_id']
data = df.head(0)
template = loader.get_template('data_playground/index.html')
context = {
'props': json.dumps(ids.tolist()),
'df': df,
'data': data,
'ids': ids
}
return HttpResponse(template.render(context, request))
def bird_lookup(bird_type):
"""Combine datasets for given bird and year"""
# Retrieve datasets
routes = read_frame(
Routes.objects.values('countrynum', 'statenum', 'route', 'routename',
'active', 'latitude', 'longitude'))
weather = read_frame(
Weather.objects.values('countrynum', 'statenum', 'route', 'year',
'routedataid', 'rpid'))
species = read_frame(Species.objects.values('aou', 'english_common_name'))
total_state_df = read_frame(
UsStates.objects.filter(aou=bird_type).values('countrynum', 'statenum',
'route', 'year',
'routedataid', 'rpid',
'aou', 'stoptotal',
'speciestotal'))
# Merge into one dataframe
all_df = pd.merge(weather,
routes,
how='left',
left_on=['countrynum', 'statenum', 'route'],
right_on=['countrynum', 'statenum', 'route'])
all_df = pd.merge(all_df,
total_state_df,
how='right',
left_on=[
'countrynum', 'statenum', 'route', 'routedataid',
'rpid', 'year'
],
right_on=[
'countrynum', 'statenum', 'route', 'routedataid',
'rpid', 'year'
])
all_df = pd.merge(all_df, species, how='left', on='aou')
# Get sizes of dots per number of sightings
all_df['species_size'] = all_df['speciestotal'].apply(species_sizer)
all_df['routename'] = [capwords(state) for state in all_df['routename']]
return (all_df)
def update_rules(request):
if request.user.is_authenticated:
my_tags = UserRule.objects.filter(user=request.user).all()
df = read_frame(my_tags)
df = df[[
'begins_with', 'ends_with', 'description', 'category', 'tag', 'id'
]]
df = df.values.tolist()
template = loader.get_template('core_app/rule_details.html')
context = {'results': df}
return HttpResponse(template.render(context, request))
else:
return redirect('home')
def sampleScenario(number_of_points=100,
buffer_size_km=10,
feature_names=[
"statenm", "statecd", "stateab", "subp", "n_inventor",
"elev", "invyr", "tree", "spcd", "family", "ht_m",
"ba_m2", "biomass_kg"
]):
points = sampleScenario1pre(number_of_points=number_of_points)
selections = selectFromBuffer(points, buffer_size_km)
dfsels = map(lambda q: read_frame(q, fieldnames=feature_names), selections)
## compent if prefer custom feature selection
dfsels = getSummaryStatistics(dfsels)
return dfsels
def home(request):
try:
a = Start.objects.all()
if request.method == "GET":
b = read_frame(a)
return render(request, 'myapp/home.html', {'b': b})
c = request.POST['resname']
m = request.POST['addSuggestion']
if (m == "0"):
q = Start.objects.values("coupon").filter(name=c)
qq = read_frame(q)
print(q)
# qqq = qq.coupon
# return HttpResponse(q)
return render(request, 'myapp/coupon.html', {'cou': qq})
else:
b = read_frame(a)
suggestion = AddSuggestion.objects.create(suggestion=c)
return render(request, 'myapp/home.html', {'b': b})
except MultiValueDictKeyError:
return render(request, 'myapp/home.html')
def handle(self, *args, **kwargs):
# Create a version
version_obj, created = Version.objects.get_or_create(
reference='200101_demo', version_date=datetime(2020, 1, 1))
if kwargs['clean'] and not created:
# Delete old data
self.stdout.write('Deleting existing forecast...')
forecast_objs = Forecast.objects.filter(version=version_obj)
forecast_objs.delete()
self.stdout.write('Existing forecast deleted')
# Get past orders
order_qs = OrderDetail.objects.values('warehouse__id', 'product__id',
'category__id', 'circuit__id',
'customer__id',
'ordered_quantity',
'order__ordered_at')
order_df = read_frame(order_qs)
order_df['ordered_quantity'] = order_df['ordered_quantity'].astype(
'int32')
self.stdout.write('Version object created')
# Create forecast dataframe
forecast_df = order_df.rename(
columns={
'ordered_quantity': 'forecasted_quantity',
'order__ordered_at': 'forecast_date',
})
# Generate different values based on defined rate
forecast_df['forecasted_quantity'] = pd.to_numeric(
forecast_df['forecasted_quantity'] * 0.7)
forecast_df['edited_forecasted_quantity'] = pd.to_numeric(
forecast_df['forecasted_quantity'] * 1.3)
self.stdout.write('Creating forecasts...')
forecast_objs = [
Forecast(
warehouse_id=row['warehouse__id'],
product_id=row['product__id'],
category_id=row['category__id'],
circuit_id=row['circuit__id'],
customer_id=row['customer__id'],
version_id=version_obj.id,
forecast_date=row['forecast_date'],
forecasted_quantity=row['forecasted_quantity'],
edited_forecasted_quantity=row['edited_forecasted_quantity'],
) for i, row in forecast_df.iterrows()
]
Forecast.objects.bulk_create(forecast_objs)
self.stdout.write(self.style.SUCCESS('Forecasts created successfully'))
def prediction(request):
template = 'stock_api/prediction.html'
# filtering company
kun_company = request.GET.get('company_abbr')
qs = StockData.objects.filter(
companyAbbr=kun_company) # RUSSELL 2000 INDEX(company name)
df = read_frame(qs)
data = df['open']
stock_data = np.array(data.values.reshape((len(data), 1)))
stock_data.shape # (1258, 1)
# Split the data into training/testing sets
stock_X_train = stock_data[:350]
stock_X_test = stock_data[700:979, ]
# Split the targets into training/testing sets
stock_y_train = stock_data[350:700]
stock_y_test = stock_data[979:]
regr = linear_model.LinearRegression()
regr.fit(stock_X_train, stock_y_train)
stock_y_pred = regr.predict(stock_X_test)
# stock_y_pred stock_X_test
# # Statistical parameter
# #coefficients
# coefficients = regr.coef_
# # mean squared error
# MSE = mean_squared_error(mark_y_test, mark_y_pred)
# # Explained variance score: 1 is perfect prediction
# RSQ = r2_score(mark_y_test, mark_y_pred)
# mark_y_pred_pd = pd.DataFrame(mark_y_pred.reshape(mark_y_pred.shape[1], -1), index=exam, columns=subject)
# mark_X_test_pd = pd.DataFrame(mark_y_test.reshape(mark_y_test.shape[1], -1), index=exam, columns=subject)
stock_y_pred_html = stock_y_pred #.to_html()
stock_X_test_html = stock_X_test #.to_html()
# plt.plot(stock_y_pred)
# plt.plot(stock_X_test)
print(stock_data[-1])
last_data = regr.predict(stock_data[-1])
print(last_data)
context = {
# 'MSE':MSE,
# 'RSQ':RSQ,
# 'coefficients':coefficients,
'stock_y_pred_html': stock_y_pred_html,
'stock_X_test_html': stock_X_test_html,
}
# return render(request, template, context)
return JsonResponse({'tomorrow_predicted_value': last_data[0][0]})
def map2():
adr = mergedTables.objects.all()
qs_adr = adr
df_of_query_result_adr = read_frame(qs_adr)
newTable_adr = df_of_query_result_adr
m = folium.Map(location=[48.8566, 2.3522],
tiles="Stamen Toner",
zoom_start=2)
#cergy=['49.034955','2.069925']
#pau=['43.319568','-0.360571']
newTable_ent_adr = newTable_adr[['ENT_LAT', 'ENT_LON']]
newTable_stu_adr = newTable_adr[['ADR_LAT', 'ADR_LON']]
newTable_site_adr = newTable_adr[['SITE_LAT', 'SITE_LON']]
newTable_ent_adr = newTable_ent_adr.dropna(axis=0,
subset=['ENT_LAT', 'ENT_LON'])
newTable_stu_adr = newTable_stu_adr.dropna(axis=0,
subset=['ADR_LAT', 'ADR_LON'])
newTable_site_adr = newTable_site_adr.dropna(
axis=0, subset=['SITE_LAT', 'SITE_LON'])
#newTable_ent_adr= [[row['ENT_LAT'],row['ENT_LON']] for index,row in newTable_ent_adr.iterrows() ]
#newTable_stu_adr= [[row['ADR_LAT'],row['ADR_LON']] for index,row in newTable_stu_adr.iterrows() ]
#HeatMap(cergy).add_to(m)
HeatMap(newTable_stu_adr).add_to(m)
mc = MarkerCluster()
folium.CircleMarker(
[49.034955, 2.069925],
radius=20,
popup='Cergy Campus',
color='red',
).add_to(m)
folium.CircleMarker(
[43.319568, -0.360571],
radius=20,
popup='Pau Campus',
color='red',
).add_to(m)
for i in range(0, len(newTable_stu_adr)):
#mc.add_child(folium.Marker([newTable_ent_adr.iloc[i]['ENT_LAT'],newTable_ent_adr.iloc[i]['ENT_LON']],icon=folium.Icon(icon='cloud'))).add_to(m)
mc.add_child(
folium.Marker([
newTable_ent_adr.iloc[i]['ENT_LAT'],
newTable_ent_adr.iloc[i]['ENT_LON']
])).add_to(m)
#mc.add_child(folium.Marker([newTable_stu_adr.iloc[i]['ADR_LAT'],newTable_stu_adr.iloc[i]['ADR_LON']],icon=folium.Icon(color='red'))).add_to(m)
#folium.Marker([newTable_site_adr.iloc[i]['SITE_LAT'],newTable_site_adr.iloc[i]['SITE_LON']],icon=folium.Icon(icon='green')).add_to(m)
#m.add_children(plugins.HeatMap(newTable_adr_lo, radius=15))
m.save(os.path.join(BASE_DIR,
'DjangoWeb V1\Interface\\templates\map.html'))
return None
def get_meta_info(icustay_id):
"""
This function is supposed to return an df containing meta info like:
- ward, moves, outcomes, admission ID, patient ID, etc FOR A GIVEN ICUSTAY.
- this is what will become the labels later!
:param icustay_id:
:return:
"""
stay = ICUSTAY.objects.filter(icustayID=icustay_id)
stay = read_frame(stay)
stay['intime'] = pd.to_datetime(stay['intime'])
stay['outtime'] = pd.to_datetime(stay['outtime'])
# admission
admission_obj_str = stay['admission'].values[0] # this is a string: `Admission object (199395)`
admission_id = re.search(re.compile('(\d+)'), admission_obj_str).group()
admission = Admission.objects.filter(admID=admission_id)
admission = read_frame(admission)
stay['admission'] = admission_id
# patient
subject_obj_str = stay['subject'].values[0] # this is a string: `Patient object (199395)`
subject_id = re.search(re.compile('(\d+)'), subject_obj_str).group()
patient = Patient.objects.filter(subjectID=subject_id)
patient = read_frame(patient)
stay['subject'] = subject_id
# gather info:
stay.drop(['first_ward_id', 'last_ward_id', 'first_cu', 'last_cu'], axis=1, inplace=True)
for c in ['adm_time', 'disch_time', 'inpmor', 'pdismor', 'read']:
stay[c] = admission[c]
for c in ['gender', 'age', 'date_of_death_hosp']:
stay[c] = patient[c]
return stay
def protocol_summary(request):
'''
Relies on django_highcharts for serializing data. Makes multiple chart
rendering on a page more simple, however it doesn't appear to allow much
in they way of customization.
define a range, avg, diplay max and min
'''
athlete = get_user(request)
data = BoulderingRoutineMetrics.objects.filter(session__athlete=athlete)
test = BoulderingRoutineMetrics.objects.values(
'min', 'max').filter(session__athlete=athlete)
df = read_frame(data, verbose=False)
df['avg'] = df['total_points'] / df['total_climbs']
df['range'] = df['max'] - df['min']
df = df.drop(['id', 'total_climbs', 'total_points'],
axis=1).set_index('session')
dataset = BoulderingRoutineMetrics.objects \
.values('session__sessionDate', 'min','max', 'total_points', 'total_climbs') \
.filter(session__athlete = athlete)
df2 = read_frame(dataset, verbose=False).rename(
columns={
'session__sessionDate': 'session date',
'min': 'min value',
'max': 'max value'
})
df3 = read_frame(dataset, fieldnames=['min','max'])\
.rename(columns={'session__sessionDate':'Session Date'})
# .drop(['total_points', 'total_climbs'])
df4 = df2.merge(df3, how='left', left_index=True, right_index=True)
df4['avg'] = df4['total_points'] / df4['total_climbs']
df4['range'] = df4['max value'] - df4['min value']
chart = serialize(df4,
render_to='my-chart',
kind="bar",
output_type='json')
return render(request, 'schedule/protocol_summary.html', {'chart': chart})
def get_stay_lab_timeseries(icustay_id):
"""
For a given admission ID -> get the ts with all the labevents.
then get in and out time from an icustay and cut the bits out that fall between
"""
# get ICUstay:
stay = ICUstay.objects.filter(icustayID=icustay_id)
stay = read_frame(stay)
stay['intime'] = pd.to_datetime(stay['intime'])
stay['outtime'] = pd.to_datetime(stay['outtime'])
admission_obj_str = stay['admission'].values[0] # this is a string: `Admission object (199395)`
admission_id = re.search(re.compile('(\d+)'), admission_obj_str).group()
stay_period = pd.Interval(stay['intime'].values[0],
stay['outtime'].values[0])
# filter the lab_events:
events = LabEventValue.objects.filter(admission_id=admission_id)
events = read_frame(events)
events['chart_time'] = pd.to_datetime(events['chart_time'])
events.set_index('chart_time', inplace=True)
idx = [t in stay_period for t in events.index.values]
events = events.loc[idx].reset_index()
events['icustayID'] = stay['icustayID'].unique()[0]
# convert to ts:
events_ts = _events_to_ts(events, variable_col='itemID', value='value').reset_index()
if events_ts.empty:
return events_ts
intime = pd.Timestamp(stay['intime'].values[0])
events_ts = add_hours_elpased_to_events(events_ts, dt=intime)
return events_ts
def results(request):
movies = Movies.objects.all()
movies_df = read_frame(movies)
ratings = Ratings.objects.all()
ratings_df = read_frame(ratings)
data = pd.merge(movies_df, ratings_df, on="movieid")
# creating dataframe with 'rating' count values
ratings = pd.DataFrame(data.groupby('title')['rating'].mean())
ratings['num of ratings'] = pd.DataFrame(
data.groupby('title')['rating'].count())
moviemat = data.pivot_table(index='userid',
columns='title',
values='rating')
ratings.sort_values('num of ratings', ascending=False).head(10)
currently_rated_movies = request.session.get('movies')
results = {}
for item in currently_rated_movies:
# analysing correlation with similar movies
_user_ratings = moviemat[item["title"]]
similar_to_ = moviemat.corrwith(_user_ratings)
corr_ = pd.DataFrame(similar_to_, columns=['Correlation'])
corr_.dropna(inplace=True)
# Similar movies like current title
corr_.sort_values('Correlation', ascending=False).head(10)
corr_ = corr_.join(ratings['num of ratings'])
strongest_corr = corr_[corr_['num of ratings'] > 100].sort_values(
'Correlation', ascending=False).head(1)
strongest_corr = strongest_corr.rename(
columns={"num of ratings": "no_of_ratings"})
if not strongest_corr.empty:
results[strongest_corr.index.values[0]] = Ratings.objects.filter(
userid=request.user.id).filter(
movieid=item["movieid"]).values_list(
'rating', flat=True)[0] * strongest_corr.Correlation[
0] * strongest_corr.no_of_ratings[0]
print(results)
return render(request, 'nxtwatch/results.html', {'results': results})
def election(request):
station_ids = request.GET.getlist('station_id')
if len(station_ids) < 2:
_response400("station_id is required at least 2")
try:
station_ids = [int(station_id) for station_id in station_ids]
except ValueError:
_response400(f"station_id is invalid: {station_ids}")
df_stations = read_frame(Station.objects.all())
results = Election(df_stations, station_ids).run()
response = dict(station_ids=station_ids, results=results)
return JsonResponse(response,
safe=False,
json_dumps_params={'ensure_ascii': False})
def update_tags(request):
if request.user.is_authenticated:
my_tags = Tags.objects.filter(
user=request.user).order_by('category').all()
df = read_frame(my_tags)
df = df[[
'category', 'tag', 'drill_down', 'id', 'budget', 'fixed_cost'
]]
df = df.values.tolist()
template = loader.get_template('core_app/tag_details.html')
context = {'results': df}
return HttpResponse(template.render(context, request))
else:
return redirect('home')
def get_list(request):
custid = request.POST.get('custid')
#history = purchases.objects.all
cust = customer.objects.all
customers = read_frame(customer.objects.all())
customers = customers['Customer_ID']
custid = int(custid)
if custid in customers.unique():
#funtion
df = read_frame(purchases.objects.all())
df = df.loc[df['Customer_ID'] == custid]
array = df['Products'].astype(int)
cat = read_frame(catalogue.objects.all())
cat = cat.head(5)
pd.to_numeric(cat['objectID'])
display = cat.loc[cat['objectID'].isin(array)]
#function
list1 = {"dataframe": cat, "customers": cust}
return render(request, "get_list.html", list1)
else:
messages.error(request, 'Please enter valid username')
return render(request, "index.html")
def players_ctx(pos,
ctx,
form,
x_axis='points_per_game',
y_axis='now_cost',
size='selected_by_percent',
limit=50):
position = Position.objects.get(id=pos)
players = Player.objects.filter(position=position).filter(
minutes__gt=0).order_by('-' + x_axis)
gw = Fixtures.objects.filter(
finished=True).order_by('event_id').last().event_id
## Create DataFrame for Goalkeepers
gkp = read_frame(
players,
fieldnames=[
'first_name', 'second_name', 'team', 'points_per_game',
'total_points', 'now_cost', 'form', 'minutes', 'assists',
'goals_scored', 'own_goals', 'influence', 'influence_rank',
'creativity', 'creativity_rank', 'threat', 'threat_rank',
'clean_sheets', 'goals_conceded', 'saves', 'penalties_saved',
'yellow_cards', 'red_cards', 'dreamteam_count',
'selected_by_percent', 'transfers_in', 'transfers_out'
])
# Merge first_name and second_name
gkp['name'] = gkp['first_name'] + ' ' + gkp['second_name']
gkp.drop(['first_name', 'second_name'], axis=1, inplace=True)
# Change number of minutes to percent of time
max_minutes = 90 * gw
gkp.minutes = gkp.minutes.apply(
lambda x: round(float(x) * 100 / max_minutes, 2))
# Filter
min_filter = gkp.minutes >= limit
gkp = gkp[min_filter]
# Add data to ctx dictionary
ctx['title'] = f'Chart - {position.name}s'
ctx['x'] = list(gkp[x_axis])
ctx['y'] = list(gkp[y_axis])
ctx['max_size'] = max(gkp[size]) if len(gkp[size]) > 0 else 0
ctx['size'] = list(((gkp[size] / ctx['max_size']) + 1) * 10)
ctx['players'] = gkp['name'].tolist()
ctx['form'] = form
x_regr, pred_regr = linear_regr(gkp[x_axis], gkp[y_axis])
ctx['x_regr'] = x_regr
ctx['pred_regr'] = pred_regr
x_regr_poly, pred_regr_poly = poly_regr(gkp[x_axis], gkp[y_axis])
ctx['x_regr_poly'] = x_regr_poly
ctx['pred_regr_poly'] = pred_regr_poly
return ctx
def get(self, *args, **kwargs):
user_taste = generics.get_object_or_404(UserTaste,
user=self.request.user)
reviews_from_user = Review.objects.all().filter(created_by=user_taste)
reviews_from_user_df = read_frame(reviews_from_user,
fieldnames=['game_id__id', 'rating'])
reviews_from_user_df = reviews_from_user_df.rename(columns={
'game_id__id': 'game_key',
'rating': 'rating'
})
return Response(selfmade_KnnWithMeans_approach(reviews_from_user_df))
def export_to_xls_search_history(self, request, queryset):
df = read_frame(queryset,
verbose=True,
fieldnames=(
'user__user__id', 'user__user__username',
'date',
'name',
))
df = translate_column_names(df)
return export_to_xls(df,
'search_history' +
request.GET.get('drf__day__gte ', '') +
request.GET.get('drf__day__lte', '') + '.xls',
engine='openpyxl')
def get_stocks_by_date_range(min: Optional[datetime] = None,
max: Optional[datetime] = None) -> pd.DataFrame:
"""
Gets stock data from the database based on date range.
"""
from stocks.stocks.models import Stock
if min is None:
return get_stocks()
if max is None:
max = datetime.now()
return read_frame(Stock.objects.filter(price_date__range=(min, max)))
def __init__(self):
self.df = read_frame(
get_user_model().objects.filter(
date_joined__gte=(datetime.now() - timedelta(days=15))
)
)
self.df['date_joined'] = pd.to_datetime(self.df['date_joined'])
# TODO: Optimise next queries
self.active = self.df[self.df.is_active==True][
['date_joined', 'id']
].groupby('date_joined').count().resample('D').count()
self.nonactive = self.df[self.df.is_active==False][
['date_joined', 'id']
].groupby('date_joined').count().resample('D').count()
def test_related_cols(self):
qs = TradeLog.objects.all()
cols = ['log_datetime', 'symbol', 'symbol__isin', 'trader__name',
'price', 'volume', 'note__note']
df = read_frame(qs, cols, verbose=False)
self.assertEqual(df.shape, (qs.count(), len(cols)))
self.assertListEqual(
list(qs.values_list('symbol__isin', flat=True)),
df.symbol__isin.tolist()
)
self.assertListEqual(
list(qs.values_list('trader__name', flat=True)),
df.trader__name.tolist()
)
def export_to_xls_exchange(self, request, queryset, ):
df = read_frame(queryset,
index_col='date',
verbose=True,
fieldnames=(
'date',
'user__user__id', 'user__user__username',
'user__user__email',
'exchange',
'user__score', 'user__total_score',
'user__total_exchange',
))
df = translate_column_names(df)
return export_to_xls(df,
'exchange' + request.GET.get('drf__day__gte ', '') + '_' +
request.GET.get('drf__day__lte', '') + '.xls',
engine='openpyxl')
def monthly_patients(self):
dataframe = read_frame(PatientVaccination.objects.all())
dataframe['month'] = [date.strftime('%B') for date in dataframe['dateofvaccinereceiption']]
groups = dataframe.groupby('month')['id'].count()
return groups
def runRedo(self):
'''
将更新票项的过程中失败的数据重做直到成功
'''
# 查出需要进行重做的数据
self.writeSystemLog(u'重做过程开始执行')
# 尝试进程登录
loginResult = self.login(self.username, self.password)
if loginResult:
self.writeSystemLog(u'登录成功!')
else:
self.writeSystemLog(u'登录失败!')
raise Exception(u'登录失败!')
# 执行时间变量的初始化
runStartTime = datetime.now()
currentTime = datetime.now()
timeSpend = currentTime - runStartTime
# 获取总出错项
qs = RefreshRedo.objects.filter(state=u'redoing')
redos = read_frame(qs)
errorCount = len(redos)
self.writeSystemLog(u'共有%d次错误需要重做' % errorCount)
# 不停处理每一个信息项,直到没有错误需要重做,或者执行时间到了
while errorCount > 0:
try:
qs = RefreshRedo.objects.filter(state=u'redoing')
redos = read_frame(qs)
errorCount = len(redos)
if errorCount == 0:
self.writeSystemLog(u'已无项目需要处理,程序将退出')
else:
self.writeSystemLog(u'开始一次重做，共%d需要处理...' % errorCount)
# 根据出错列表调用重新回订过程
successCount = 0
# 按唯一的日期获取数据，提高访问效率
for day in redos.beginDay.drop_duplicates():
self.writeSystemLog(u'检查%s的余票信息' % day)
# 获取有余票的项
ticketInfo = pitcher.getTicketInfo(day)
c1 = ticketInfo[u'余票'].apply(lambda x: int(x)) > 0
remain = ticketInfo[c1]
remain[u'开航日期'] = day
# 余票信息和重做记录关联，得出可以重做的票项
leftColumns = ['beginDay', 'beginTime', 'departure', 'arrival']
rightColumns = [u'开航日期', u'开航时间', u'出发码头', u'抵达码头']
todos = redos.merge(remain, left_on=leftColumns, right_on=rightColumns).to_dict(outtype='records')
if todos:
self.writeSystemLog(u'有%d个重做项出现余票' % len(todos))
else:
self.writeSystemLog(u'无重做项有余票')
# 避免过分频繁访问服务器
time.sleep(self.normalWaitingSecond)
# 循环对有余票的项目进程操作
for todo in todos:
# 初始化相关信息
departure = todo['departure']
arrival = todo['arrival']
beginDay = todo['beginDay']
beginTime = todo['beginTime']
cnt = todo['cnt']
remainCnt = int(todo[u'余票'])
redoId = todo['id']
redo = RefreshRedo.objects.get(id=redoId)
# 如果当前时间和开航日期0点的时间小于1天，则不再刷新这个票项
# 因为正常刷新提前5天前就刷新好了，所以不会有影响
# 这样做主要是防止重做票项无限增长，且去订之前日期的票
beginDateTime = parser.parse(beginDay)
if (beginDateTime - currentTime).days < 1:
redo.state = u'failed'
redo.save()
self.writeSystemLog(
u'出发:%s,抵达:%s,开航时间:%s %s,人数:%s' % (departure, arrival, beginDay, beginTime, cnt))
self.writeSystemLog(u'该项已经超过可以重做的时间，已标记为失败!!!')
continue
# 尝试重调订票过程
dailyFlightId = pitcher.getDailyFlightId(beginDay, beginTime, departure, arrival)
if dailyFlightId:
self.writeSystemLog(
u'出发:%s,抵达:%s,开航时间:%s %s,人数:%s' % (departure, arrival, beginDay, beginTime, cnt))
if remainCnt >= cnt:
# 如果票数量够，按需票量抢
result = pitcher.orderTicket(dailyFlightId, cnt)
if result:
# 重做成功将记录标识为完成
redo.state = u'finished'
redo.save()
self.writeSystemLog(u'该项已经重做成功!')
successCount += 1
else:
self.writeSystemLog(u'虽有余票项但抢订失败!')
else:
# 如果票数不够，按实际票数抢
result = pitcher.orderTicket(dailyFlightId, remainCnt)
if result:
redo.cnt -= remainCnt
redo.save()
self.writeSystemLog(u'重做项余票数不够,但已抢回%d张' % remainCnt)
else:
self.writeSystemLog(u'虽有余票项但抢订失败!')
# 避免过分频繁访问服务器(for todo1 in todos:)
time.sleep(self.normalWaitingSecond)
# 避免过分频繁访问服务器(for day in redos.beginDay.drop_duplicates())
time.sleep(self.normalWaitingSecond)
# 完成一次重做,打印执行信息
errorCount -= successCount
self.writeSystemLog(u'完成一次重做：%d项成功，%d项失败.' % (successCount, errorCount))
# 更新执行时间
currentTime = datetime.now()
timeSpend = currentTime - runStartTime
except Exception as e:
self.writeSystemLog(u'发生异常:%s，程序将继续执行' % unicode(e))
# 检查是否是登录状态，如果登录状态丢失则重新登录
if not self.isLogin():
self.writeSystemLog(u'连接丢失，将重新登录...')
loginErrorCount = 0
while True:
loginResult = self.login(self.username, self.password)
if loginResult:
# 如果登录成功清空原来登录失败的记录
self.writeSystemLog(u'登录成功!')
break
else:
loginErrorCount += 1
if loginErrorCount > self.maxLoginError:
self.writeSystemLog(u'登录失败超过%d次，程序退出!' % self.maxLoginError)
return
else:
self.writeSystemLog(u'登录失败!')
self.writeSystemLog(u'等待%s秒... ...' % self.errorWaitingSecond)
time.sleep(self.errorWaitingSecond)
self.writeSystemLog(u'重做过程执行结束')
def monthly_vaccine(self):
dataframe = read_frame(PatientVaccination.objects.all())
dataframe['month'] = [date.strftime('%B') for date in dataframe['dateofvaccinereceiption']]
groups = dataframe.groupby(['month', 'patient_vaccine'])['id'].count().reset_index(name='count')
return groups
def monthly_transactions(self):
dataframe = read_frame(Transaction.objects.all())
dataframe['month'] = [date.strftime('%B') for date in dataframe['date']]
dataframe['year'] = [date.strftime('%Y') for date in dataframe['date']]
groups = dataframe.groupby(['year', 'month', 'type'])['amount'].sum().reset_index(name='amount')
return groups
def top_recipients(self):
dataframe = read_frame(Transaction.objects.all())
recipient_dataframe = dataframe[dataframe.type == 'Sent Transactions']
return recipient_dataframe.groupby(['recipient', 'type'])['amount'].sum().reset_index(name='amount').sort('amount', ascending=False)
