bdext 0.1.69__py3-none-any.whl → 0.1.70__py3-none-any.whl

bdeissct_dl/dl_model.py

@@ -4,17 +4,9 @@ from tensorflow.python.keras.utils.generic_utils import register_keras_serializa
 from bdeissct_dl.bdeissct_model import F_S, UPSILON, REPRODUCTIVE_NUMBER, \
     INFECTION_DURATION, X_S, X_C, RHO, INCUBATION_PERIOD
 
-LEARNING_RATE = 0.001
+from collections import defaultdict
 
-LOSS_WEIGHTS = {
-    REPRODUCTIVE_NUMBER: 1,
-    INFECTION_DURATION: 1,
-    INCUBATION_PERIOD: 1,
-    F_S: 200, # as it is a value between 0 and 0.5, we multiply by 200 to scale it to [0, 100]
-    UPSILON: 100,
-    X_C: 1,
-    X_S: 1
-}
+LEARNING_RATE = 0.001
 
 @register_keras_serializable(package="bdeissct_dl", name="half_sigmoid")
 def half_sigmoid(x):
@@ -26,16 +18,17 @@ def relu_plus_one(x):
 
 
 
-LOSS_FUNCTIONS = {
-    REPRODUCTIVE_NUMBER: "mean_absolute_percentage_error",
-    INFECTION_DURATION: "mean_absolute_percentage_error",
-    INCUBATION_PERIOD: "mae",
-    UPSILON: 'mae',
-    RHO: 'mean_absolute_percentage_error',
-    X_C: "mean_absolute_percentage_error",
-    F_S: 'mae',
-    X_S: "mean_absolute_percentage_error",
-}
+LOSS_FUNCTIONS = defaultdict(lambda: "mean_squared_error")
+LOSS_FUNCTIONS.update({
+    REPRODUCTIVE_NUMBER: "mean_squared_error",
+    INFECTION_DURATION: "mean_squared_error",
+    INCUBATION_PERIOD: "mean_squared_error",
+    UPSILON: 'mean_squared_error',
+    RHO: 'mean_squared_error',
+    X_C: "mean_squared_error",
+    F_S: 'mean_squared_error',
+    X_S: "mean_squared_error",
+})
 
 
 def build_model(target_columns, n_x, optimizer=None, metrics=None):
@@ -63,11 +56,11 @@ def build_model(target_columns, n_x, optimizer=None, metrics=None):
     outputs = {}
 
     if REPRODUCTIVE_NUMBER in target_columns:
-        outputs[REPRODUCTIVE_NUMBER] = tf.keras.layers.Dense(1, activation="softplus", name=REPRODUCTIVE_NUMBER)(x) # positive values only
+        outputs[REPRODUCTIVE_NUMBER] = tf.keras.layers.Dense(1, activation="relu", name=REPRODUCTIVE_NUMBER)(x) # positive values only
     if INFECTION_DURATION in target_columns:
-        outputs[INFECTION_DURATION] = tf.keras.layers.Dense(1, activation="softplus", name=INFECTION_DURATION)(x) # positive values only
+        outputs[INFECTION_DURATION] = tf.keras.layers.Dense(1, activation="relu", name=INFECTION_DURATION)(x) # positive values only
     if INCUBATION_PERIOD in target_columns:
-        outputs[INCUBATION_PERIOD] = tf.keras.layers.Dense(1, activation="softplus", name=INCUBATION_PERIOD)(x) # positive values only
+        outputs[INCUBATION_PERIOD] = tf.keras.layers.Dense(1, activation="relu", name=INCUBATION_PERIOD)(x) # positive values only
     if F_S in target_columns:
         outputs[F_S] = tf.keras.layers.Dense(1, activation=half_sigmoid, name="FS_logits")(x)
     if X_S in target_columns:
@@ -84,6 +77,5 @@ def build_model(target_columns, n_x, optimizer=None, metrics=None):
 
     model.compile(optimizer=optimizer,
                   loss={col: LOSS_FUNCTIONS[col] for col in outputs.keys()},
-                  loss_weights={col: LOSS_WEIGHTS[col] for col in outputs.keys()},
                   metrics=metrics)
     return model

{bdext-0.1.69.dist-info → bdext-0.1.70.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: bdext
-Version: 0.1.69
+Version: 0.1.70
 Summary: Estimation of BDEISS-CT parameters from phylogenetic trees.
 Home-page: https://github.com/modpath/bdeissct
 Author: Anna Zhukova

{bdext-0.1.69.dist-info → bdext-0.1.70.dist-info}/RECORD

@@ -1,7 +1,7 @@
 README.md,sha256=Ngj8bt0Yu3LUsvwblmMtUqqjvGyqxv6ku2_cYCb5_DQ,6539
 bdeissct_dl/__init__.py,sha256=QPEiIP-xVqGQgydeqN_9AZgT26IYWeJC4-JlHnd8Rjo,296
 bdeissct_dl/bdeissct_model.py,sha256=sQclYN5V8utw6wEMDN0_Ua-0NeuyuWHG_e0_jQIUe8Q,1986
-bdeissct_dl/dl_model.py,sha256=eXyFHqgJtXovzPFYn9xW5-Th92f28Ci6iuRHEwSd8y0,3678
+bdeissct_dl/dl_model.py,sha256=wpwlUVy6kOhPIsT1zg-Us2_bdnntxdnCbNQB4UxYzTg,3433
 bdeissct_dl/estimator.py,sha256=QBWA8R0pBPZPd3JvItdJS2lN1J3VqvdJqBMzCi-NADs,3336
 bdeissct_dl/model_serializer.py,sha256=s1yBzQjhtr-w7eT8bTsNkG9_xnYRZrUc3HkeOzNZpQY,2464
 bdeissct_dl/scaler_fitting.py,sha256=9X0O7-Wc9xGTI-iF-Pfp1PPoW7j01wZUfJVZf8ky-IU,1752
@@ -9,9 +9,9 @@ bdeissct_dl/sumstat_checker.py,sha256=TQ0nb86-BXmusqgMnOJusLpR4ul3N3Hi886IWUovrM
 bdeissct_dl/training.py,sha256=H5wA3V72nhc9Km7kvKmzjCYw0N1itMGDbj9c-Uat5BU,8350
 bdeissct_dl/tree_encoder.py,sha256=V-7_Kis9x9JacI_mF7rWRGGKvxn7AWFCto7LkgRawBw,18286
 bdeissct_dl/tree_manager.py,sha256=UXxUVmEkxwUhKpJeACVgiXZ8Kp1o_hiv8Qb80b6qmVU,11814
-bdext-0.1.69.dist-info/LICENSE,sha256=OXLcl0T2SZ8Pmy2_dmlvKuetivmyPd5m1q-Gyd-zaYY,35149
-bdext-0.1.69.dist-info/METADATA,sha256=XR1qaonSH4YNjna_G_yMzDwYdCrvFHw75msvPJ2XMUo,7479
-bdext-0.1.69.dist-info/WHEEL,sha256=tZoeGjtWxWRfdplE7E3d45VPlLNQnvbKiYnx7gwAy8A,92
-bdext-0.1.69.dist-info/entry_points.txt,sha256=lcAwyk-Fc0G_w4Ex7KDivh7h1tzSA99PRMcy971b-nM,208
-bdext-0.1.69.dist-info/top_level.txt,sha256=z4dadFfcLghr4lwROy7QR3zEICpa-eCPT6mmcoHeEJY,12
-bdext-0.1.69.dist-info/RECORD,,
+bdext-0.1.70.dist-info/LICENSE,sha256=OXLcl0T2SZ8Pmy2_dmlvKuetivmyPd5m1q-Gyd-zaYY,35149
+bdext-0.1.70.dist-info/METADATA,sha256=USWAUX3zunofN9x2-6E63lFVrNAdRx2cZf2Sc8HKGe8,7479
+bdext-0.1.70.dist-info/WHEEL,sha256=tZoeGjtWxWRfdplE7E3d45VPlLNQnvbKiYnx7gwAy8A,92
+bdext-0.1.70.dist-info/entry_points.txt,sha256=lcAwyk-Fc0G_w4Ex7KDivh7h1tzSA99PRMcy971b-nM,208
+bdext-0.1.70.dist-info/top_level.txt,sha256=z4dadFfcLghr4lwROy7QR3zEICpa-eCPT6mmcoHeEJY,12
+bdext-0.1.70.dist-info/RECORD,,