changeset 44:bb9fc9d46ea4 draft

planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/tools/sklearn commit 80417bf0158a9b596e485dd66408f738f405145a
author bgruening
date Mon, 02 Oct 2023 09:42:16 +0000
parents aeca97d9a214
children 838ca001438d
files keras_train_and_eval.py
diffstat 1 files changed, 18 insertions(+), 6 deletions(-) [+]
line wrap: on
line diff
--- a/keras_train_and_eval.py	Fri Sep 22 17:14:20 2023 +0000
+++ b/keras_train_and_eval.py	Mon Oct 02 09:42:16 2023 +0000
@@ -188,6 +188,7 @@
     infile1,
     infile2,
     outfile_result,
+    outfile_history=None,
     outfile_object=None,
     outfile_y_true=None,
     outfile_y_preds=None,
@@ -215,6 +216,9 @@
     outfile_result : str
         File path to save the results, either cv_results or test result.
 
+    outfile_history : str, optional
+        File path to save the training history.
+
     outfile_object : str, optional
         File path to save searchCV object.
 
@@ -253,9 +257,7 @@
     swapping = params["experiment_schemes"]["hyperparams_swapping"]
     swap_params = _eval_swap_params(swapping)
     estimator.set_params(**swap_params)
-
     estimator_params = estimator.get_params()
-
     # store read dataframe object
     loaded_df = {}
 
@@ -448,12 +450,20 @@
     # train and eval
     if hasattr(estimator, "config") and hasattr(estimator, "model_type"):
         if exp_scheme == "train_val_test":
-            estimator.fit(X_train, y_train, validation_data=(X_val, y_val))
+            history = estimator.fit(X_train, y_train, validation_data=(X_val, y_val))
         else:
-            estimator.fit(X_train, y_train, validation_data=(X_test, y_test))
+            history = estimator.fit(X_train, y_train, validation_data=(X_test, y_test))
     else:
-        estimator.fit(X_train, y_train)
-
+        history = estimator.fit(X_train, y_train)
+    if "callbacks" in estimator_params:
+        for cb in estimator_params["callbacks"]:
+            if cb["callback_selection"]["callback_type"] == "CSVLogger":
+                hist_df = pd.DataFrame(history.history)
+                hist_df["epoch"] = np.arange(1, estimator_params["epochs"] + 1)
+                epo_col = hist_df.pop('epoch')
+                hist_df.insert(0, 'epoch', epo_col)
+                hist_df.to_csv(path_or_buf=outfile_history, sep="\t", header=True, index=False)
+                break
     if isinstance(estimator, KerasGBatchClassifier):
         scores = {}
         steps = estimator.prediction_steps
@@ -526,6 +536,7 @@
     aparser.add_argument("-X", "--infile1", dest="infile1")
     aparser.add_argument("-y", "--infile2", dest="infile2")
     aparser.add_argument("-O", "--outfile_result", dest="outfile_result")
+    aparser.add_argument("-hi", "--outfile_history", dest="outfile_history")
     aparser.add_argument("-o", "--outfile_object", dest="outfile_object")
     aparser.add_argument("-l", "--outfile_y_true", dest="outfile_y_true")
     aparser.add_argument("-p", "--outfile_y_preds", dest="outfile_y_preds")
@@ -542,6 +553,7 @@
         args.infile1,
         args.infile2,
         args.outfile_result,
+        outfile_history=args.outfile_history,
         outfile_object=args.outfile_object,
         outfile_y_true=args.outfile_y_true,
         outfile_y_preds=args.outfile_y_preds,