Skip to content

Commit

Permalink
Merge pull request #19 from KasperSkytte/graph
Browse files Browse the repository at this point in the history 
Output predicted data too
  • Loading branch information
@KasperSkytte
KasperSkytte authored Feb 26, 2024
2 parents a56fea7 + 4b38479 commit f5aaa6d
Showing 1 changed file with 84 additions and 4 deletions.
88 changes: 84 additions & 4 deletions main.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -237,7 +237,7 @@ def find_best_graph(data, iterations, num_clusters, max_epochs, early_stopping,
best_performances.append(best_performance)
best_model.save_weights(f'{results_dir}/graph_{cluster_type}_weights/cluster_{c}')

prediction = make_prediction(data, best_model)
prediction, actual_prediction = make_prediction(data, best_model)
# reverse transform and overwrite.
# Better to implement it in data_handler,
# but this does the job
Expand Down Expand Up @@ -277,7 +277,44 @@ def find_best_graph(data, iterations, num_clusters, max_epochs, early_stopping,
max = data.transform_max[data.clusters_idec == c_id],
transform = data.transform_type
)


if cluster_type == "abund":
actual_prediction = rev_transform(
DF = actual_prediction,
mean = data.transform_mean[data.clusters_abund == c_id],
std = data.transform_std[data.clusters_abund == c_id],
min = data.transform_min[data.clusters_abund == c_id],
max = data.transform_max[data.clusters_abund == c_id],
transform = data.transform_type
)
elif cluster_type == "graph":
actual_prediction = rev_transform(
DF=actual_prediction,
mean=data.transform_mean[data.clusters_graph == c_id],
std=data.transform_std[data.clusters_graph == c_id],
min=data.transform_min[data.clusters_graph == c_id],
max=data.transform_max[data.clusters_graph == c_id],
transform=data.transform_type
)
elif cluster_type == "func":
actual_prediction = rev_transform(
DF = actual_prediction,
mean = data.transform_mean[data.clusters_func == c_id],
std = data.transform_std[data.clusters_func == c_id],
min = data.transform_min[data.clusters_func == c_id],
max = data.transform_max[data.clusters_func == c_id],
transform = data.transform_type
)
elif cluster_type == "idec":
actual_prediction = rev_transform(
DF = actual_prediction,
mean = data.transform_mean[data.clusters_idec == c_id],
std = data.transform_std[data.clusters_idec == c_id],
min = data.transform_min[data.clusters_idec == c_id],
max = data.transform_max[data.clusters_idec == c_id],
transform = data.transform_type
)

dates = data.get_metadata(data.all, 'Date').dt.date
dates_test = data.get_metadata(data.test, 'Date').dt.date
# Date of the first sample in the test set and
Expand All @@ -298,6 +335,7 @@ def find_best_graph(data, iterations, num_clusters, max_epochs, early_stopping,
if not path.exists(data_predicted_dir):
mkdir(data_predicted_dir)
prediction.to_csv(f'{data_predicted_dir}/graph_{cluster_type}_cluster_{c}_predicted.csv')
actual_prediction.to_csv(f'{data_predicted_dir}/graph_{cluster_type}_cluster_{c}_actual_prediction.csv')
data.all.to_csv(f'{data_predicted_dir}/graph_{cluster_type}_cluster_{c}_dataall.csv')
data.all_nontrans.to_csv(f'{data_predicted_dir}/graph_{cluster_type}_cluster_{c}_dataall_nontrans.csv')

Expand Down Expand Up @@ -380,12 +418,16 @@ def find_best_idec(data, iterations, num_clusters, tolerance):


def make_prediction(data, lstm_model):
actual_prediction = data.all[-data.window_width:].to_numpy().reshape([1, data.window_width, -1])
actual_prediction = lstm_model.predict(actual_prediction)
actual_prediction = actual_prediction.reshape([data.window_width, -1])

prediction = lstm_model.predict(data.all_batched)
prediction = prediction[:, 0]
index_pred = data.all.index[data.window_width:]

#needs to be reverse transformed for real values
return pd.DataFrame(data = prediction, index = index_pred, columns = data.all.columns)
return pd.DataFrame(data = prediction, index = index_pred, columns = data.all.columns), pd.DataFrame(data = actual_prediction, columns = data.all.columns)

def create_lstm_model(num_features, predict_timestamp=1):
"""Create a model without tuning hyperparameters.
Expand Down Expand Up @@ -449,7 +491,7 @@ def find_best_lstm(data, iterations, num_clusters, max_epochs, early_stopping, c
best_performances.append(best_performance)
best_model.save_weights(f'{results_dir}/lstm_{cluster_type}_weights/cluster_{c}')

prediction = make_prediction(data, best_model)
prediction, actual_prediction = make_prediction(data, best_model)
# reverse transform and overwrite.
# Better to implement it in data_handler,
# but this does the job
Expand Down Expand Up @@ -481,6 +523,43 @@ def find_best_lstm(data, iterations, num_clusters, max_epochs, early_stopping, c
transform = data.transform_type
)

if cluster_type == "abund":
actual_prediction = rev_transform(
DF = actual_prediction,
mean = data.transform_mean[data.clusters_abund == c_id],
std = data.transform_std[data.clusters_abund == c_id],
min = data.transform_min[data.clusters_abund == c_id],
max = data.transform_max[data.clusters_abund == c_id],
transform = data.transform_type
)
elif cluster_type == "graph":
actual_prediction = rev_transform(
DF=actual_prediction,
mean=data.transform_mean[data.clusters_graph == c_id],
std=data.transform_std[data.clusters_graph == c_id],
min=data.transform_min[data.clusters_graph == c_id],
max=data.transform_max[data.clusters_graph == c_id],
transform=data.transform_type
)
elif cluster_type == "func":
actual_prediction = rev_transform(
DF = actual_prediction,
mean = data.transform_mean[data.clusters_func == c_id],
std = data.transform_std[data.clusters_func == c_id],
min = data.transform_min[data.clusters_func == c_id],
max = data.transform_max[data.clusters_func == c_id],
transform = data.transform_type
)
elif cluster_type == "idec":
actual_prediction = rev_transform(
DF = actual_prediction,
mean = data.transform_mean[data.clusters_idec == c_id],
std = data.transform_std[data.clusters_idec == c_id],
min = data.transform_min[data.clusters_idec == c_id],
max = data.transform_max[data.clusters_idec == c_id],
transform = data.transform_type
)

dates = data.get_metadata(data.all, 'Date').dt.date
dates_test = data.get_metadata(data.test, 'Date').dt.date
# Date of the first sample in the test set and
Expand All @@ -501,6 +580,7 @@ def find_best_lstm(data, iterations, num_clusters, max_epochs, early_stopping, c
if not path.exists(data_predicted_dir):
mkdir(data_predicted_dir)
prediction.to_csv(f'{data_predicted_dir}/lstm_{cluster_type}_cluster_{c}_predicted.csv')
actual_prediction.to_csv(f'{data_predicted_dir}/graph_{cluster_type}_cluster_{c}_actual_prediction.csv')
data.all.to_csv(f'{data_predicted_dir}/lstm_{cluster_type}_cluster_{c}_dataall.csv')
data.all_nontrans.to_csv(f'{data_predicted_dir}/lstm_{cluster_type}_cluster_{c}_dataall_nontrans.csv')

Expand Down

0 comments on commit f5aaa6d

Please sign in to comment.