Skip to content

Updating intial weights based on local betas and fixing bug computing… #2

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Sep 4, 2024
Merged

Updating intial weights based on local betas and fixing bug computing… #2

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 2 additions & 0 deletions scripts/local.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -80,6 +80,7 @@ def local_1(args):
output_dict = {
"beta_vector_local": beta_vector,
"beta_vec_size": beta_vec_size,
"augmented_X_labels": list(augmented_X.columns),
"number_of_regressions": len(y_labels),
"computation_phase": "local_1"
}
Expand Down Expand Up @@ -275,3 +276,4 @@ def start(PARAM_DICT):
else:
raise ValueError("Error occurred at Local")


24 changes: 19 additions & 5 deletions scripts/remote.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -61,9 +61,13 @@ def remote_1(args):
#log(args, args['state'])
"""Need this function for performing multi-shot regression"""
input_list = args["input"]
first_user_id = list(input_list.keys())[0]
first_user_id = list(sorted(input_list.keys()))[0]
beta_vec_size = input_list[first_user_id]["beta_vec_size"]
number_of_regressions = input_list[first_user_id]["number_of_regressions"]
mean_local_betas = np.array(
[np.array(input_list[site]["beta_vector_local"]) for site in input_list]).mean(
axis=0)


# Initial setup
beta1 = 0.9
Expand All @@ -78,6 +82,15 @@ def remote_1(args):
np.zeros((number_of_regressions, beta_vec_size), dtype=float)
for _ in range(4)
]

#Update initial weights based on the local beta's
local_X_labels=['const']
local_X_labels.extend(args['cache']['X_labels'])
augmented_X_labels=input_list[first_user_id]['augmented_X_labels']
for label_idx, curr_label in enumerate(local_X_labels):
idx = augmented_X_labels.index(curr_label)
wp[:, idx] = mean_local_betas[:, label_idx]

prev_cost = [None] * number_of_regressions

iter_flag = 1
Expand Down Expand Up @@ -253,11 +266,11 @@ def remote_3(args):
site: input_list[site]["local_stats_list"] for site in sorted_site_ids
}]

mean_y_local = [input_list[site]["mean_y_local"] for site in input_list]
count_y_local = [np.array(input_list[site]["count_local"]) for site in input_list]
mean_y_global = np.array(mean_y_local) * np.array(count_y_local)
mean_y_local = np.array([input_list[site]["mean_y_local"] for site in input_list])
count_y_local = np.array([input_list[site]["count_local"] for site in input_list])
mean_y_global = mean_y_local * count_y_local
#mean_y_global = np.average(mean_y_global, axis=0)
mean_y_global = mean_y_global.sum(axis=0) / np.sum(count_y_local)
mean_y_global = mean_y_global.sum(axis=0) / count_y_local.sum(axis=0)

dof_global = sum(count_y_local) - avg_beta_vector.shape[1]

Expand Down Expand Up @@ -417,3 +430,4 @@ def start(PARAM_DICT):
else:
raise ValueError("Error occurred at Remote")