ddi-fw 0.0.61__py3-none-any.whl → 0.0.63__py3-none-any.whl

ddi_fw/experiments/evaluation_helper.py

@@ -85,10 +85,104 @@ def roc_aupr_score(y_true, y_score, average="macro"):
 
     return _average_binary_score(_binary_roc_aupr_score, y_true, y_score, average)
 
-# actual and pred are one-hot encoded
+
+def evaluate(actual, pred, info='', print=False):
+    # Precompute y_true and y_pred
+    y_true = np.argmax(actual, axis=1)
+    y_pred = np.argmax(pred, axis=1)
+    
+    # Generate classification report
+    c_report = classification_report(y_true, y_pred, output_dict=True)
+    
+    # Metrics initialization
+    metrics = Metrics(info)
+    
+    n_classes = actual.shape[1]
+    
+    precision = {}
+    recall = {}
+    f_score = {}
+    roc_aupr = {}
+    roc_auc = {
+        "weighted": 0,
+        "macro": 0,
+        "micro": 0
+    }
+
+    # Preallocate lists
+    precision_vals = [[] for _ in range(n_classes)]
+    recall_vals = [[] for _ in range(n_classes)]
+
+    # Compute metrics for each class
+    for i in range(n_classes):
+        precision_vals[i], recall_vals[i], _ = precision_recall_curve(
+            actual[:, i], pred[:, i])
+        roc_aupr[i] = auc(recall_vals[i], precision_vals[i])
+
+    # Calculate ROC AUC scores
+    roc_auc["weighted"] = roc_auc_score(actual, pred, multi_class='ovr', average='weighted')
+    roc_auc["macro"] = roc_auc_score(actual, pred, multi_class='ovr', average='macro')
+    roc_auc["micro"] = roc_auc_score(actual, pred, multi_class='ovr', average='micro')
+
+    # Micro-average Precision-Recall curve and ROC-AUPR
+    precision["micro_event"], recall["micro_event"], _ = precision_recall_curve(actual.ravel(), pred.ravel())
+    roc_aupr["micro"] = auc(recall["micro_event"], precision["micro_event"])
+
+    # Convert lists to numpy arrays for better performance
+    precision["micro_event"] = precision["micro_event"].tolist()
+    recall["micro_event"] = recall["micro_event"].tolist()
+
+    # Overall accuracy
+    acc = accuracy_score(y_true, y_pred)
+
+    # Aggregate precision, recall, and f_score
+    for avg_type in ['weighted', 'macro', 'micro']:
+        precision[avg_type] = precision_score(y_true, y_pred, average=avg_type)
+        recall[avg_type] = recall_score(y_true, y_pred, average=avg_type)
+        f_score[avg_type] = f1_score(y_true, y_pred, average=avg_type)
+
+    if print:
+        print(
+            f'''Accuracy: {acc}
+            , Precision:{precision['weighted']}
+            , Recall: {recall['weighted']}
+            , F1-score: {f_score['weighted']}
+            ''')
+
+    logs = {'accuracy': acc,
+            'weighted_precision': precision['weighted'],
+            'macro_precision': precision['macro'],
+            'micro_precision': precision['micro'],
+            'weighted_recall_score': recall['weighted'],
+            'macro_recall_score': recall['macro'],
+            'micro_recall_score': recall['micro'],
+            'weighted_f1_score': f_score['weighted'],
+            'macro_f1_score': f_score['macro'],
+            'micro_f1_score': f_score['micro'],
+            # 'weighted_roc_auc_score': weighted_roc_auc_score,
+            # 'macro_roc_auc_score': macro_roc_auc_score,
+            # 'micro_roc_auc_score': micro_roc_auc_score,
+            # 'macro_aupr_score': macro_aupr_score,
+            # 'micro_aupr_score': micro_aupr_score
+            "micro_roc_aupr": roc_aupr['micro'],
+            # "micro_precision_from_precision_recall_curve":precision["micro"],
+            # "micro_recall_from_precision_recall_curve":recall["micro"],
+            "weighted_roc_auc": roc_auc['weighted'],
+            "macro_roc_auc": roc_auc['macro'],
+            "micro_roc_auc": roc_auc['micro']
+            }
+    metrics.accuracy(acc)
+    metrics.precision(precision)
+    metrics.recall(recall)
+    metrics.f1_score(f_score)
+    metrics.roc_auc(roc_auc)
+    metrics.roc_aupr(roc_aupr)
+    metrics.classification_report(c_report)
+    return logs, metrics
 
 
-def evaluate(actual, pred, info = '' ,print=False):
+# actual and pred are one-hot encoded
+def evaluate_ex(actual, pred, info = '' ,print=False):
     
     y_pred = np.argmax(pred, axis=1)
     y_true = np.argmax(actual, axis=1)

ddi_fw/experiments/pipeline.py

@@ -33,7 +33,8 @@ class Experiment:
                  experiment_tags=None, 
                  tracking_uri=None, 
                  dataset_type:BaseDataset=None, 
-                 columns=None, 
+                 columns=None,
+                 embedding_dict = None, 
                  vector_db_persist_directory=None, 
                  vector_db_collection_name=None, 
                  embedding_pooling_strategy_type:PoolingStrategy=None, 
@@ -48,6 +49,7 @@ class Experiment:
         self.tracking_uri = tracking_uri
         self.dataset_type = dataset_type
         self.columns = columns
+        self.embedding_dict = embedding_dict
         self.vector_db_persist_directory = vector_db_persist_directory
         self.vector_db_collection_name = vector_db_collection_name
         self.embedding_pooling_strategy_type = embedding_pooling_strategy_type
@@ -61,21 +63,22 @@ class Experiment:
         kwargs = {"columns": self.columns}
         for k, v in self.ner_threshold.items():
             kwargs[k] = v
-        if self.vector_db_persist_directory:
-            self.vector_db = chromadb.PersistentClient(
-                path=self.vector_db_persist_directory)
-            self.collection = self.vector_db.get_collection(
-                self.vector_db_collection_name)
-            dictionary = self.collection.get(include=['embeddings', 'metadatas'])
+        if self.embedding_dict == None:
+            if self.vector_db_persist_directory:
+                self.vector_db = chromadb.PersistentClient(
+                    path=self.vector_db_persist_directory)
+                self.collection = self.vector_db.get_collection(
+                    self.vector_db_collection_name)
+                dictionary = self.collection.get(include=['embeddings', 'metadatas'])
 
-            embedding_dict = defaultdict(lambda: defaultdict(list))
+                embedding_dict = defaultdict(lambda: defaultdict(list))
 
-            for metadata, embedding in zip(dictionary['metadatas'], dictionary['embeddings']):
-                embedding_dict[metadata["type"]][metadata["id"]].append(embedding)
+                for metadata, embedding in zip(dictionary['metadatas'], dictionary['embeddings']):
+                    embedding_dict[metadata["type"]][metadata["id"]].append(embedding)
 
-            embedding_size = dictionary['embeddings'].shape[1]
+                embedding_size = dictionary['embeddings'].shape[1]
 
-            pooling_strategy = self.embedding_pooling_strategy_type()
+        pooling_strategy = self.embedding_pooling_strategy_type()
 
         self.ner_df = CTakesNER().load(filename=self.ner_data_file)  if self.ner_data_file else None
 

ddi_fw/utils/py7zr_helper.py

@@ -93,7 +93,7 @@ class Py7ZipHelper:
         with py7zr.SevenZipFile(archive_name, 'r') as archive:
             archive.extractall(path=extract_path)
 
-    def extract_multiparts(self, input_path, output_path, output_file):
+    def extract_multiparts(self, input_path, output_path, output_file, remove_file = True):
         parts = [input_path+'/' + p for p in os.listdir(input_path)]
         sorted_parts = sorted(parts, key = lambda x: int(x.split(".")[-1][4:]))
         create_folder_if_not_exists(output_path)
@@ -102,4 +102,5 @@ class Py7ZipHelper:
                 with open(part, 'rb') as infile:
                     outfile.write(infile.read())
         self.extract_archive(f"{output_path}/{output_file}", output_path)
-        os.remove(f"{output_path}/{output_file}")
+        if remove_file:
+            os.remove(f"{output_path}/{output_file}")

{ddi_fw-0.0.61.dist-info → ddi_fw-0.0.63.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: ddi_fw
-Version: 0.0.61
+Version: 0.0.63
 Summary: Do not use :)
 Author-email: Kıvanç Bayraktar <bayraktarkivanc@gmail.com>
 Maintainer-email: Kıvanç Bayraktar <bayraktarkivanc@gmail.com>

{ddi_fw-0.0.61.dist-info → ddi_fw-0.0.63.dist-info}/RECORD

@@ -58,8 +58,8 @@ ddi_fw/drugbank/drugbank_processor_org.py,sha256=eO5Yset50P91qkic79RUXPoEuxRxQKF
 ddi_fw/drugbank/event_extractor.py,sha256=6odoZohhK7OdLF-LF0l-5BFq0_NMG_5jrFJbHrBXsI8,4600
 ddi_fw/experiments/__init__.py,sha256=5L2xSolpFycNnflqOMdvJSiqRB16ExA5bbVGORKFX04,195
 ddi_fw/experiments/custom_torch_model.py,sha256=iQ_R_EApzD2JCcASN8cie6D21oh7VCxaOQ45_dkiGwc,2576
-ddi_fw/experiments/evaluation_helper.py,sha256=pY69cezV3WzrXw1bduIwRJfah1w3wXJ2YyTNim1J7ko,9349
-ddi_fw/experiments/pipeline.py,sha256=wttkvdzGP9d3jC9nx2iZul4hbogXkRho6eDns0yfLiE,5380
+ddi_fw/experiments/evaluation_helper.py,sha256=o4-w5Xa3t4olLW4ymx_8L-Buhe5wfQEmT2bh4Zz544c,13066
+ddi_fw/experiments/pipeline.py,sha256=VsKPgYsGTY2bYIajRBAewBgP9-izmrL0Qtbn48qV5tw,5544
 ddi_fw/experiments/pipeline_builder_pattern.py,sha256=q1PNEQFoO5U3UidEoGB8rgLA7KXr4FsJTXEug5c5UJg,5466
 ddi_fw/experiments/pipeline_ner.py,sha256=unxEJCYrG6wEZjLmqvGdLRTMOBwELbGKkdygSpAR3b8,5043
 ddi_fw/experiments/tensorflow_helper.py,sha256=xUnbntWyc2Wm4TvmVFAnpwLHg-o13oM26GUHom6d5m0,11776
@@ -80,10 +80,10 @@ ddi_fw/test/torch_cuda_test.py,sha256=R-4VGVErl_Ufk54DoZbgL_YXWoCYFyanIVWd6P39IE
 ddi_fw/test/type_guarding_test.py,sha256=KxjyBxohDu7lwpejalCj-REjtJ-k1S1wQbOB6TGY0O8,766
 ddi_fw/utils/__init__.py,sha256=x1ypYJRKJlbF9x4psHYGXj-YbDD8T_c28gXZkr03cdE,273
 ddi_fw/utils/enums.py,sha256=19eJ3fX5eRK_xPvkYcukmug144jXPH4X9zQqtsFBj5A,671
-ddi_fw/utils/py7zr_helper.py,sha256=dgfHqXDBWys1hmd1JlHhYyZGxrzYWi6siYiUq3bnLuI,4698
+ddi_fw/utils/py7zr_helper.py,sha256=gOqaFIyJvTjUM-btO2x9AQ69jZOS8PoKN0wetYIckJw,4747
 ddi_fw/utils/utils.py,sha256=szwnxMTDRrZoeNRyDuf3aCbtzriwtaRk4mHSH3asLdA,4301
 ddi_fw/utils/zip_helper.py,sha256=YRZA4tKZVBJwGQM0_WK6L-y5MoqkKoC-nXuuHK6CU9I,5567
-ddi_fw-0.0.61.dist-info/METADATA,sha256=7JkmFI1nIibnlPkq8_Hsnl8napLtHRrH8Y9dDiqdHVY,1565
-ddi_fw-0.0.61.dist-info/WHEEL,sha256=GV9aMThwP_4oNCtvEC2ec3qUYutgWeAzklro_0m4WJQ,91
-ddi_fw-0.0.61.dist-info/top_level.txt,sha256=PMwHICFZTZtcpzQNPV4UQnfNXYIeLR_Ste-Wfc1h810,7
-ddi_fw-0.0.61.dist-info/RECORD,,
+ddi_fw-0.0.63.dist-info/METADATA,sha256=5fuV5oU6k1S0RAqMYGxR3nGuyV0lXpcayGsd9ydsEmI,1565
+ddi_fw-0.0.63.dist-info/WHEEL,sha256=OVMc5UfuAQiSplgO0_WdW7vXVGAt9Hdd6qtN4HotdyA,91
+ddi_fw-0.0.63.dist-info/top_level.txt,sha256=PMwHICFZTZtcpzQNPV4UQnfNXYIeLR_Ste-Wfc1h810,7
+ddi_fw-0.0.63.dist-info/RECORD,,

{ddi_fw-0.0.61.dist-info → ddi_fw-0.0.63.dist-info}/WHEEL

@@ -1,5 +1,5 @@
 Wheel-Version: 1.0
-Generator: setuptools (75.1.0)
+Generator: setuptools (75.2.0)
 Root-Is-Purelib: true
 Tag: py3-none-any