def score_to_hundred(self, project: scoring_base.ScoringProject) -> float:
"""Compute a score for the given ScoringProject."""
area_type = project.details.area_type
if not area_type:
raise scoring_base.NotEnoughDataException()
if not project.imt_proto().yearly_avg_offers_per_10_candidates:
raise scoring_base.NotEnoughDataException()
num_better_departements = project.local_diagnosis(
).num_less_stressful_departements
# User is already in one of the top 3 départements.
if num_better_departements < 3:
raise scoring_base.NotEnoughDataException()
# Give a score centered around 50 corresponding to the user's mobility.
# The worse the current département is, the more extreme we set the
# score.
score_range = min(num_better_departements * 1.5, 100) / 2
if area_type >= geo_pb2.COUNTRY:
return 50 + score_range
if area_type >= geo_pb2.REGION:
return 50
if area_type >= geo_pb2.DEPARTEMENT:
return 50 - score_range / 2
return 50 - score_range
def get_close_jobs(self, project: scoring_base.ScoringProject) \
-> reorient_to_close_pb2.ReorientCloseJobs:
"""Get the jobs close to a job group."""
recommended_jobs = reorient_to_close_pb2.ReorientCloseJobs()
local_diagnosis = project.local_diagnosis()
market_score = local_diagnosis.imt.yearly_avg_offers_per_10_candidates
if not market_score:
return recommended_jobs
reorientation_jobs = local_diagnosis.less_stressful_job_groups
close_jobs = [
job for job in reorientation_jobs
if job.mobility_type == job_pb2.CLOSE
]
evolution_jobs = [
job for job in reorientation_jobs
if job.mobility_type == job_pb2.EVOLUTION
]
database = project.database
recommended_jobs.close_jobs.extend(
self._convert_to_reorient_jobs(database, close_jobs, market_score,
project))
recommended_jobs.evolution_jobs.extend(
self._convert_to_reorient_jobs(database, evolution_jobs,
market_score, project))
return recommended_jobs
def score_to_hundred(self, project: scoring_base.ScoringProject) -> float:
local_diagnosis = project.local_diagnosis()
offers_change = local_diagnosis.job_offers_change
if local_diagnosis.num_job_offers_previous_year < 5\
and local_diagnosis.num_job_offers_last_year < 5:
raise scoring_base.NotEnoughDataException()
if offers_change < 0:
return 0
# rescale offers: >=0 -> 50, >=10 -> 100
return offers_change * 5 + 50
def score_to_hundred(self, project: scoring_base.ScoringProject) -> float:
"""Compute a percentage score for the given ScoringProject."""
local_diagnosis = project.local_diagnosis()
if not local_diagnosis.unemployment_duration.days:
raise scoring_base.NotEnoughDataException()
# Rescale unemployment_duration: 0 -> 100, >= 12 months -> 0
# As unemployment_duration has much more low values in DB, use ** .7 to have more variance
# in high scores (50% -> 136 ~= median).
return 100 * (1 -
(local_diagnosis.unemployment_duration.days / 365)**.7)
def score_and_explain(self, project: scoring_base.ScoringProject) \
-> scoring_base.ExplainedScore:
"""Compute a score for the given ScoringProject."""
reasons = []
if project.details.area_type < geo_pb2.COUNTRY:
return scoring_base.NULL_EXPLAINED_SCORE
reasons.append(
project.populate_template(
project.translate_static_string(
'vous nous avez dit être prêt%eFeminine à déménager')))
local_stats = project.local_diagnosis()
if local_stats.imt.yearly_avg_offers_per_10_candidates and \
local_stats.num_less_stressful_departements:
reasons.append(
project.translate_static_string(
"il y a beaucoup plus d'offres par habitants dans d'autres villes"
))
return scoring_base.ExplainedScore(2, reasons)
return scoring_base.NULL_EXPLAINED_SCORE
def score_to_hundred(self, project: scoring_base.ScoringProject) -> int:
bmo = project.local_diagnosis().bmo
if not bmo.percent_difficult:
raise scoring_base.NotEnoughDataException()
return bmo.percent_difficult
