def training(path):
model_data = g.model_data_manager.models[path]
cached_data = model_data.get_metrics(50)
context = {
'precision_recall_curve':
plots.precision_recall_curve(cached_data),
'roc_curve':
plots.roc_curve(cached_data),
'score_distribution':
plots.score_distribution(cached_data[0]),
'absolute_score_distribution':
plots.absolute_score_distribution(cached_data[0]),
'marginal_precision_curve':
plots.marginal_precision_curve(cached_data[0]),
'threshold_graph':
plots.thresholds_graph(cached_data[0]),
'threshold_table':
plots.thresholds_table(cached_data[0]),
'auc':
auc(cached_data[0]['fprs'], cached_data[0]['recalls']),
'notes':
model_data.get_notes(),
'model_metadata':
model_data.get_metadata(),
'path':
path,
}
return render_template("results.html", **context)
def compare():
models = request.args.getlist('model[]')
cached_datas = []
for path in models:
model_data = g.model_data_manager.models[path]
cached_datas.append(model_data.get_metrics(10))
_, ax = plt.subplots(figsize=(12, 6))
for name, cached_data in zip(models, cached_datas):
prc = plots.precision_recall_curve(cached_data, ax=ax, label=name)
_, ax = plt.subplots(figsize=(12, 6))
for name, cached_data in zip(models, cached_datas):
roc = plots.roc_curve(cached_data, ax=ax, label=name)
context = {'precision_recall_curve': prc, 'roc_curve': roc}
return render_template("compare.html", **context)
def training(path):
model_data = g.model_data_manager.models[path]
cached_data = model_data.get_metrics(50)
context = {
'precision_recall_curve': plots.precision_recall_curve(cached_data),
'roc_curve': plots.roc_curve(cached_data),
'score_distribution': plots.score_distribution(cached_data[0]),
'marginal_precision_curve': plots.marginal_precision_curve(cached_data[0]),
'threshold_graph': plots.thresholds_graph(cached_data[0]),
'threshold_table': plots.thresholds_table(cached_data[0]),
'brier': plots.box_brier(cached_data),
'auc': auc(cached_data[0]['fprs'], cached_data[0]['recalls']),
'notes': model_data.get_notes(),
'path': path,
}
return render_template("results.html", **context)
def training(path):
model_data = ModelData(g.file_system, path)
cached_data = model_data.get_metrics(50)
# print cached_data
context = {
'precision_recall_curve': plots.precision_recall_curve(cached_data),
'roc_curve': plots.roc_curve(cached_data),
'score_distribution': plots.score_distribution(cached_data[0]),
'marginal_precision_curve':
plots.marginal_precision_curve(cached_data[0]),
'threshold_graph': plots.thresholds_graph(cached_data[0]),
'threshold_table': plots.thresholds_table(cached_data[0]),
'brier': plots.box_brier(cached_data),
'auc': auc(cached_data[0]['fprs'], cached_data[0]['recalls']),
'notes': ModelData(g.file_system, path).get_notes(),
'path': path,
}
return render_template("results.html", **context)
def training(path):
model_data = ModelData(g.file_system, path)
cached_data = model_data.get_metrics(50)
# print cached_data
context = {
# 'precision_recall_curve': plots.precision_recall_curve(cached_data),
'support_precision_curve': plots.support_precision_curve(cached_data),
'roc_curve': plots.roc_curve(cached_data),
'score_distribution': plots.score_distribution(cached_data[0]),
'marginal_precision_curve': plots.marginal_precision_curve(cached_data[0]),
'threshold_graph': plots.thresholds_graph(cached_data[0]),
'threshold_table': plots.thresholds_table(cached_data[0]),
'brier': plots.box_brier(cached_data),
'auc': auc(cached_data[0]['fprs'], cached_data[0]['recalls']),
'notes': ModelData(g.file_system, path).get_notes(),
'path': path,
}
return render_template("results.html", **context)
def compare():
models = request.args.getlist('model[]')
cached_datas = []
for path in models:
model_data = ModelData(g.file_system, path)
cached_datas.append(model_data.get_metrics(10))
_, ax = plt.subplots(figsize=(12, 6))
for name, cached_data in zip(models, cached_datas):
prc = plots.support_precision_curve(cached_data, ax=ax, label=name)
_, ax = plt.subplots(figsize=(12, 6))
for name, cached_data in zip(models, cached_datas):
roc = plots.roc_curve(cached_data, ax=ax, label=name)
context = {
'support_precision_curve': prc,
'roc_curve': roc}
return render_template("compare.html", **context)
def compare():
models = request.args.getlist('model[]')
cached_datas = []
for path in models:
model_data = g.model_data_manager.models[path]
cached_datas.append(model_data.get_metrics(10))
_, ax = plt.subplots(figsize=(12, 6))
for name, cached_data in zip(models, cached_datas):
prc = plots.precision_recall_curve(cached_data, ax=ax, label=name)
_, ax = plt.subplots(figsize=(12, 6))
for name, cached_data in zip(models, cached_datas):
roc = plots.roc_curve(cached_data, ax=ax, label=name)
context = {
'precision_recall_curve': prc,
'roc_curve': roc}
return render_template("compare.html", **context)