running 5k

deepskies · Aug 2, 2024 · 4b4d1f7 · 4b4d1f7
1 parent 93c4b7c
commit 4b4d1f7
Show file tree

Hide file tree

Showing 5 changed files with 167 additions and 124 deletions.
diff --git a/notebooks/fig1.ipynb b/notebooks/fig1.ipynb
diff --git a/notebooks/model_validation.ipynb b/notebooks/model_validation.ipynb
diff --git a/notebooks/sigma_in_sigma_out.ipynb b/notebooks/sigma_in_sigma_out.ipynb
diff --git a/src/data/data.py b/src/data/data.py
@@ -199,7 +199,6 @@ def simulate_data(
             if vary_sigma:
                 print("YES WERE VARYING SIGMA")
                 new_sig = self.get_sigma_m(sigma, m)
-                print('new sig', new_sig)
                 ε = rs.normal(
                     loc=0, scale=new_sig, size=(len(x), thetas.shape[0])
                 )

diff --git a/src/scripts/DeepEnsemble.py b/src/scripts/DeepEnsemble.py
@@ -328,17 +328,13 @@ def beta_type(value):
     prescription = config.get_item("data", "data_prescription", "DE")
     injection = config.get_item("data", "data_injection", "DE")
     dim = config.get_item("data", "data_dimension", "DE")
-    sigma = DataPreparation.get_sigma(
-        noise, inject_type=injection, data_dimension=dim
-    )
-    print(f"inject type is {injection}, dim is {dim}, sigma is {sigma}")
     if config.get_item("data", "generatedata", "DE", raise_exception=False):
         # generate the df
         print("generating the data")
         data = DataPreparation()
         if dim == "0D":
             data.sample_params_from_prior(size_df)
-            print("injecting this noise", noise, sigma)
+            print("injecting this noise", noise)
             if injection == "feature":
                 vary_sigma = True
                 print('are we varying sigma', vary_sigma)
@@ -351,6 +347,10 @@ def beta_type(value):
                     vary_sigma=vary_sigma
                 )
             elif injection == "predictive":
+                sigma = DataPreparation.get_sigma(
+                    noise, inject_type=injection, data_dimension=dim
+                )
+                print(f"inject type is {injection}, dim is {dim}, sigma is {sigma}")
                 data.simulate_data(
                     data.params,
                     sigma,
@@ -371,6 +371,10 @@ def beta_type(value):
                     df[key] = torch.tensor(value)
         elif dim == "2D":
             print("2D data")
+            sigma = DataPreparation.get_sigma(
+                noise, inject_type=injection, data_dimension=dim
+            )
+            print(f"inject type is {injection}, dim is {dim}, sigma is {sigma}")
             data.sample_params_from_prior(
                 size_df,
                 low=[0, 1, -1.5],