bdext 0.1.71__py3-none-any.whl → 0.1.72__py3-none-any.whl

bdeissct_dl/dl_model.py

@@ -6,7 +6,7 @@ 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, INCUBATION_FRACTION
 
-LEARNING_RATE = 0.001
+LEARNING_RATE = 0.01
 
 @register_keras_serializable(package="bdeissct_dl", name="half_sigmoid")
 def half_sigmoid(x):
@@ -36,12 +36,14 @@ def build_model(target_columns, n_x, optimizer=None, metrics=None):
     inputs = tf.keras.Input(shape=(n_x,))
 
     # (Your hidden layers go here)
-    x = tf.keras.layers.Dense(128, activation='elu', name=f'layer1_dense256_elu')(inputs)
-    x = tf.keras.layers.Dropout(0.5, name='dropout1_50')(x)
-    x = tf.keras.layers.Dense(64, activation='elu', name=f'layer2_dense128_elu')(x)
-    x = tf.keras.layers.Dropout(0.5, name='dropout2_50')(x)
-    x = tf.keras.layers.Dense(32, activation='elu', name=f'layer3_dense64elu')(x)
-    x = tf.keras.layers.Dense(16, activation='elu', name=f'layer4_dense32_elu')(x)
+    x = tf.keras.layers.Dense(128, activation='elu', name=f'layer1_dense128_elu')(inputs)
+    # x = tf.keras.layers.Dropout(0.5, name='dropout1_50')(x)
+    x = tf.keras.layers.Dense(64, activation='elu', name=f'layer2_dense64_elu')(x)
+    # x = tf.keras.layers.Dropout(0.5, name='dropout2_50')(x)
+    x = tf.keras.layers.Dense(32, activation='elu', name=f'layer3_dense32elu')(x)
+    x = tf.keras.layers.Dense(16, activation='elu', name=f'layer4_dense16_elu')(x)
+    x = tf.keras.layers.Dense(8, activation='elu', name=f'layer5_dense8_elu')(x)
+    x = tf.keras.layers.Dense(4, activation='elu', name=f'layer5_dense4_elu')(x)
 
     outputs = {}
 
@@ -52,13 +54,13 @@ def build_model(target_columns, n_x, optimizer=None, metrics=None):
     if INCUBATION_FRACTION in target_columns:
         outputs[INCUBATION_FRACTION] = tf.keras.layers.Dense(1, activation="sigmoid", name=INCUBATION_FRACTION)(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)
+        outputs[F_S] = tf.keras.layers.Dense(1, activation=half_sigmoid, name=F_S)(x)
     if X_S in target_columns:
-        outputs[X_S] = tf.keras.layers.Dense(1, activation=relu_plus_one, name="XS_logits")(x)
+        outputs[X_S] = tf.keras.layers.Dense(1, activation=relu_plus_one, name=X_S)(x)
     if UPSILON in target_columns:
-        outputs[UPSILON] = tf.keras.layers.Dense(1, activation="sigmoid", name="ups_logits")(x)
+        outputs[UPSILON] = tf.keras.layers.Dense(1, activation="sigmoid", name=UPSILON)(x)
     if X_C in target_columns:
-        outputs[X_C] = tf.keras.layers.Dense(1, activation=relu_plus_one, name="XC_logits")(x)
+        outputs[X_C] = tf.keras.layers.Dense(1, activation=relu_plus_one, name=X_C)(x)
 
     model = tf.keras.models.Model(inputs=inputs, outputs=outputs)
 

bdeissct_dl/scaler_fitting.py

@@ -27,7 +27,6 @@ def main():
     parser = \
         argparse.ArgumentParser(description="Fit a BD(EI)(SS)(CT) data scaler.")
     parser.add_argument('--train_data', type=str, nargs='+',
-                        # default=[f'/home/azhukova/projects/bdeissct_dl/simulations_bdeissct/training/500_1000/{model}/{i}/trees.csv.xz' for i in range(120) for model in [BD, BDCT, BDEI, BDEICT, BDSS, BDSSCT, BDEISS, BDEISSCT]],
                         help="path to the files where the encoded training data are stored")
     parser.add_argument('--model_path', default=MODEL_PATH, type=str,
                         help="path to the folder where the scaler should be stored.")

bdeissct_dl/training.py

@@ -141,14 +141,8 @@ def main():
     parser = \
         argparse.ArgumentParser(description="Train a BD(EI)(SS)(CT) model.")
     parser.add_argument('--train_data', type=str, nargs='+',
-                        # default=[f'/home/azhukova/projects/bdeissct_dl/simulations_bdeissct/train/2000_5000/BDEI/{i}/trees.csv.xz' for i in range(100)] \
-                        #         + [f'/home/azhukova/projects/bdeissct_dl/simulations_bdeissct/training/2000_5000/BD/{i}/trees.csv.xz' for i in range(10)]
-                        # ,
                         help="path to the files where the encoded training data are stored")
     parser.add_argument('--val_data', type=str, nargs='+',
-                        # default=[f'/home/azhukova/projects/bdeissct_dl/simulations_bdeissct/train/2000_5000/BDEI/{i}/trees.csv.xz' for i in range(100, 120)] \
-                        #     + [f'/home/azhukova/projects/bdeissct_dl/simulations_bdeissct/train/2000_5000/BD/{i}/trees.csv.xz' for i in range(10, 12)]
-                        # ,
                         help="path to the files where the encoded validation data are stored")
 
     parser.add_argument('--epochs', type=int, default=EPOCHS, help='number of epochs to train the model')
@@ -178,6 +172,13 @@ def main():
 
 
     for col, y_idx in y_col2index.items():
+        try:
+            if load_model_keras(path=params.model_path, model_name=f'{params.model_name}.{col}'):
+                print(f'Model {params.model_name}.{col} already exists at {params.model_path}. Skipping training for this target.')
+                continue
+        except:
+            pass
+
         print(f'Training to predict {col} with {params.model_name}...')
 
         if params.base_model_name is not None:
@@ -194,7 +195,7 @@ def main():
                                 scaler_x=scaler_x, batch_size=BATCH_SIZE, shuffle=True)
 
         #early stopping to avoid overfitting
-        early_stop = tf.keras.callbacks.EarlyStopping(monitor='val_loss', patience=25)
+        early_stop = tf.keras.callbacks.EarlyStopping(monitor='val_loss', patience=10)
 
         #Training of the Network, with an independent validation set
         model.fit(ds_train, verbose=1, epochs=params.epochs, validation_data=ds_val, callbacks=[early_stop])

{bdext-0.1.71.dist-info → bdext-0.1.72.dist-info}/METADATA

@@ -1,13 +1,11 @@
-Metadata-Version: 2.1
+Metadata-Version: 2.4
 Name: bdext
-Version: 0.1.71
+Version: 0.1.72
 Summary: Estimation of BDEISS-CT parameters from phylogenetic trees.
 Home-page: https://github.com/modpath/bdeissct
 Author: Anna Zhukova
 Author-email: anna.zhukova@pasteur.fr
-License: UNKNOWN
 Keywords: phylogenetics,birth-death model,incubation,super-spreading,contact tracing
-Platform: UNKNOWN
 Classifier: Development Status :: 4 - Beta
 Classifier: Environment :: Console
 Classifier: Intended Audience :: Developers
@@ -15,6 +13,7 @@ Classifier: Topic :: Scientific/Engineering :: Bio-Informatics
 Classifier: Topic :: Software Development
 Classifier: Topic :: Software Development :: Libraries :: Python Modules
 Description-Content-Type: text/markdown
+License-File: LICENSE
 Requires-Dist: tensorflow==2.19.0
 Requires-Dist: six
 Requires-Dist: ete3
@@ -24,6 +23,16 @@ Requires-Dist: biopython
 Requires-Dist: scikit-learn==1.5.2
 Requires-Dist: pandas==2.2.3
 Requires-Dist: treesumstats==0.7
+Dynamic: author
+Dynamic: author-email
+Dynamic: classifier
+Dynamic: description
+Dynamic: description-content-type
+Dynamic: home-page
+Dynamic: keywords
+Dynamic: license-file
+Dynamic: requires-dist
+Dynamic: summary
 
 # bdext
 
@@ -236,5 +245,3 @@ The other parameters are estimated from a time-scaled phylogenetic tree.
 
 [//]: # ()
 [//]: # ()
-
-

bdext-0.1.72.dist-info/RECORD

@@ -0,0 +1,17 @@
+README.md,sha256=Ngj8bt0Yu3LUsvwblmMtUqqjvGyqxv6ku2_cYCb5_DQ,6539
+bdeissct_dl/__init__.py,sha256=QPEiIP-xVqGQgydeqN_9AZgT26IYWeJC4-JlHnd8Rjo,296
+bdeissct_dl/bdeissct_model.py,sha256=um1nEQf4uym_jkrkuUjpvIbVc7VRfmAY3gnW9xgXv6I,2016
+bdeissct_dl/dl_model.py,sha256=gl6uBK6rwEJxWgzInQfyn-1UWbePQJPDWwc7Lwq5F0U,3250
+bdeissct_dl/estimator.py,sha256=QBWA8R0pBPZPd3JvItdJS2lN1J3VqvdJqBMzCi-NADs,3336
+bdeissct_dl/model_serializer.py,sha256=s1yBzQjhtr-w7eT8bTsNkG9_xnYRZrUc3HkeOzNZpQY,2464
+bdeissct_dl/scaler_fitting.py,sha256=wvHLtLmg5QP58NKSUnYBOQ4TzAtTAi_AfLVaxKXfJzM,1522
+bdeissct_dl/sumstat_checker.py,sha256=TQ0nb86-BXmusqgMnOJusLpR4ul3N3Hi886IWUovrMI,1846
+bdeissct_dl/training.py,sha256=EvD1n3uiaUb8gubxwhP1kt4xUW2hokHB7ywoUScCtmI,7979
+bdeissct_dl/tree_encoder.py,sha256=WAwn3e1lPiksZNCnwTt9wsoEX3rgF8O0b2vOx7g0gUY,20286
+bdeissct_dl/tree_manager.py,sha256=UXxUVmEkxwUhKpJeACVgiXZ8Kp1o_hiv8Qb80b6qmVU,11814
+bdext-0.1.72.dist-info/licenses/LICENSE,sha256=OXLcl0T2SZ8Pmy2_dmlvKuetivmyPd5m1q-Gyd-zaYY,35149
+bdext-0.1.72.dist-info/METADATA,sha256=5ix9OE4DIpC4K4xoW61JNPePZro3J2K5tt1AcGR6puc,7676
+bdext-0.1.72.dist-info/WHEEL,sha256=wUyA8OaulRlbfwMtmQsvNngGrxQHAvkKcvRmdizlJi0,92
+bdext-0.1.72.dist-info/entry_points.txt,sha256=DP-XVnUjSLJt-PHOJUurpkEUkkicdtGoEuGVeVb0gGg,207
+bdext-0.1.72.dist-info/top_level.txt,sha256=z4dadFfcLghr4lwROy7QR3zEICpa-eCPT6mmcoHeEJY,12
+bdext-0.1.72.dist-info/RECORD,,

{bdext-0.1.71.dist-info → bdext-0.1.72.dist-info}/WHEEL

@@ -1,5 +1,5 @@
 Wheel-Version: 1.0
-Generator: bdist_wheel (0.45.1)
+Generator: setuptools (80.10.2)
 Root-Is-Purelib: true
 Tag: py3-none-any
 

{bdext-0.1.71.dist-info → bdext-0.1.72.dist-info}/entry_points.txt

@@ -3,4 +3,3 @@ bdeissct_encode = bdeissct_dl.tree_encoder:main
 bdeissct_fit_scaler = bdeissct_dl.scaler_fitting:main
 bdeissct_infer = bdeissct_dl.estimator:main
 bdeissct_train = bdeissct_dl.training:main
-

bdext-0.1.71.dist-info/RECORD

@@ -1,17 +0,0 @@
-README.md,sha256=Ngj8bt0Yu3LUsvwblmMtUqqjvGyqxv6ku2_cYCb5_DQ,6539
-bdeissct_dl/__init__.py,sha256=QPEiIP-xVqGQgydeqN_9AZgT26IYWeJC4-JlHnd8Rjo,296
-bdeissct_dl/bdeissct_model.py,sha256=um1nEQf4uym_jkrkuUjpvIbVc7VRfmAY3gnW9xgXv6I,2016
-bdeissct_dl/dl_model.py,sha256=xtAov5s_l5XUd569t1BUrgnmsT14FFRT-gissXZZkwo,3115
-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
-bdeissct_dl/sumstat_checker.py,sha256=TQ0nb86-BXmusqgMnOJusLpR4ul3N3Hi886IWUovrMI,1846
-bdeissct_dl/training.py,sha256=ahkQoVq88sVyX8Q5bEG5DBfeamu1TH5pXQ1daf5Z-Cw,8359
-bdeissct_dl/tree_encoder.py,sha256=WAwn3e1lPiksZNCnwTt9wsoEX3rgF8O0b2vOx7g0gUY,20286
-bdeissct_dl/tree_manager.py,sha256=UXxUVmEkxwUhKpJeACVgiXZ8Kp1o_hiv8Qb80b6qmVU,11814
-bdext-0.1.71.dist-info/LICENSE,sha256=OXLcl0T2SZ8Pmy2_dmlvKuetivmyPd5m1q-Gyd-zaYY,35149
-bdext-0.1.71.dist-info/METADATA,sha256=U2hCvUQLoGiZQruc8Y5MyipYQSNzoUT-A8YXzMue6Ag,7479
-bdext-0.1.71.dist-info/WHEEL,sha256=tZoeGjtWxWRfdplE7E3d45VPlLNQnvbKiYnx7gwAy8A,92
-bdext-0.1.71.dist-info/entry_points.txt,sha256=lcAwyk-Fc0G_w4Ex7KDivh7h1tzSA99PRMcy971b-nM,208
-bdext-0.1.71.dist-info/top_level.txt,sha256=z4dadFfcLghr4lwROy7QR3zEICpa-eCPT6mmcoHeEJY,12
-bdext-0.1.71.dist-info/RECORD,,