glam4cm 0.1.1__py3-none-any.whl → 1.0.0__py3-none-any.whl

glam4cm/utils.py

@@ -1,16 +1,23 @@
 from argparse import ArgumentParser
+from ast import Dict
 import random
 import numpy as np
 import torch
 import os
 import fnmatch
 import json
-from typing import List
 import xmltodict
 from torch_geometric.data import Data
 import hashlib
 import networkx as nx
 from collections import deque
+import struct
+from tensorboardX.proto import event_pb2
+from collections import deque
+from typing import Any, List, Tuple, Optional, Set
+import networkx as nx
+
+
 
 
 
@@ -191,4 +198,260 @@ def count_total_lines_of_code(directory):
 
 
 def snake_to_title(snake_str: str):
-    return snake_str.replace("_", " ").title()
+    return snake_str.replace("_", " ").title()
+
+
+def parse_event_file(filepath):
+    """
+    Generator that yields `Event` protocol buffer messages
+    from a single TensorBoard event file, using TFRecord-like
+    parsing without TensorFlow.
+    """
+    with open(filepath, 'rb') as f:
+        while True:
+            # 1) Read the length of the next record (8 bytes, little-endian).
+            header = f.read(8)
+            if len(header) < 8:
+                break  # no more data
+
+            record_length = struct.unpack('Q', header)[0]
+
+            # 2) Skip the 4-byte length CRC (unused here).
+            _ = f.read(4)
+
+            # 3) Read the actual record data.
+            record_data = f.read(record_length)
+
+            # 4) Skip the 4-byte data CRC.
+            _ = f.read(4)
+
+            if len(record_data) < record_length:
+                # Incomplete record at end of file
+                break
+
+            # Parse the record into an Event proto.
+            event = event_pb2.Event()
+            event.ParseFromString(record_data)
+            yield event
+
+
+def get_max_scalars_with_step_epoch(logdir, epoch_tag="eval/epoch"):
+    """
+    Scans all `events.out.tfevents.*` files in `logdir` and returns a dict:
+        {
+          scalar_tag: {
+            "max_value": float,
+            "step": int or None,
+            "epoch": int or float or None
+          },
+          ...
+        }
+    By default, it looks for an 'eval/epoch' scalar to determine the epoch.
+    If that scalar isn't found, epoch will be None.
+    """
+    max_scalars = {}
+
+    # Gather all event files in the directory
+    event_files = [
+        os.path.join(logdir, f)
+        for f in os.listdir(logdir)
+        if f.startswith("events.out.tfevents")
+    ]
+
+    for filepath in event_files:
+        for event in parse_event_file(filepath):
+            # In proto3, "step" is always present as an int64.
+            # If it's not explicitly set, it'll be 0.
+            step = event.step
+            if step == 0:
+                step = None  # treat zero as "no step logged"
+
+            # Try to find an epoch value in the same event (if you're logging it).
+            epoch_val = None
+            if event.summary and event.summary.value:
+                # First pass: see if there's a dedicated epoch tag in this event
+                for v in event.summary.value:
+                    if v.tag == epoch_tag and v.HasField("simple_value"):
+                        epoch_val = v.simple_value
+                        break
+
+                # Second pass: for each scalar, update the max if we see a bigger value
+                for v in event.summary.value:
+                    if v.HasField('simple_value'):
+                        tag = v.tag
+                        val = v.simple_value
+
+                        # Ignore the epoch tag itself; we only want other scalar tags
+                        if tag == epoch_tag:
+                            continue
+
+                        # Update if this tag is new or if we found a bigger value
+                        if (tag not in max_scalars) or (val > max_scalars[tag]["max_value"]):
+                            max_scalars[tag] = {
+                                "max_value": val,
+                                "step": step,
+                                "epoch": epoch_val
+                            }
+
+    return max_scalars
+
+
+def update_config_results(logs_dir='logs'):
+    
+    logs_dir = "logs"
+    
+    def is_tf_dir(dir_path):
+        return any([f.startswith("events.out.tfevents") for f in os.listdir(dir_path)])
+    
+    graph_data_dir = "datasets/graph_data"
+    dataset_config = dict()
+
+    for dataset_dir in os.listdir(graph_data_dir):
+        if dataset_dir not in ['ecore_555', 'eamodelset', 'modelset', 'ontouml']:
+            continue
+        with open(os.path.join(graph_data_dir, dataset_dir, 'configs.json')) as f:
+            dataset_config[dataset_dir] = json.load(f)
+    
+
+    for dataset_dir in os.listdir(logs_dir):
+        if dataset_dir not in ['ecore_555', 'eamodelset', 'modelset', 'ontouml']:
+            continue
+        for task in os.listdir(os.path.join(logs_dir, dataset_dir)):
+            task_dir = os.path.join(logs_dir, dataset_dir, task)
+            if not os.path.isdir(task_dir) or "_comp_" in task:
+                continue
+            
+            for root, dirs, _ in os.walk(task_dir):
+                for dir_name in dirs:
+                    dir_path = os.path.join(root, dir_name)
+                    if is_tf_dir(dir_path):
+                        
+                        config_id = dir_name.split(os.sep)[-1]
+                        
+                        if "_" in config_id:
+                            config_id = config_id.split("_")[0]
+                            
+                        
+                        assert config_id in dataset_config[dataset_dir], f"Config {config_id} not found in {dataset_dir}"
+                        config = dataset_config[dataset_dir][config_id]
+                        config_results = get_max_scalars_with_step_epoch(dir_path)
+                        
+                        if 'results' not in config:
+                            config['results'] = list()
+                        
+                        config['results'].append(config_results)
+                        config['task'] = task
+                        
+        
+    for dataset_dir in os.listdir(graph_data_dir):
+        if dataset_dir not in ['ecore_555', 'eamodelset', 'modelset', 'ontouml']:
+            continue
+        with open(os.path.join(graph_data_dir, dataset_dir, 'configs.json'), 'w') as f:
+             json.dump(dataset_config[dataset_dir], f)
+    
+
+
+def set_encoded_labels(train_ds, test_ds):
+    train_labels = train_ds.inputs['labels']
+    test_labels = test_ds.inputs['labels']
+    all_labels = torch.cat([train_labels, test_labels])
+    unique_labels = torch.unique(all_labels)
+    label_to_encoded = dict()
+    encoded_to_label = dict()
+    for i, label in enumerate(unique_labels):
+        label_to_encoded[label.item()] = i
+        encoded_to_label[i] = label.item()
+    
+    train_ds.inputs['labels'] = torch.tensor([label_to_encoded[label.item()] for label in train_labels])
+    test_ds.inputs['labels'] = torch.tensor([label_to_encoded[label.item()] for label in test_labels])
+    
+
+def set_torch_encoding_labels(dataset: list, cls_label, exclude_labels: List[str] = None):
+    print(f"Setting encoding labels for {cls_label}")
+    labels = [getattr(data, cls_label) for data in dataset]
+    all_labels = torch.cat(labels)
+    unique_labels = [label for label in torch.unique(all_labels) if label not in exclude_labels]
+    label_to_encoded = dict()
+    encoded_to_label = dict()
+    for i, label in enumerate(unique_labels):
+        label_to_encoded[label.item()] = i
+        encoded_to_label[i] = label.item()
+            
+    for i, data in enumerate(dataset):
+        setattr(
+            data, 
+            cls_label, 
+            torch.tensor([label_to_encoded.get(label.item(), -1) for label in getattr(data, cls_label)])
+        )
+    
+    print(f"Set encoding labels for {cls_label}")
+    
+
+def find_nodes_within_distance(
+    graph: nx.DiGraph,
+    start_node: Any,
+    distance: int,
+    exclude_edges: Optional[List[Tuple[Any, Any]]] = None
+) -> List[Tuple[Any, int]]:
+    """
+    Find all nodes reachable from start_node within a given distance,
+    optionally excluding specified edges.
+
+    Parameters
+    ----------
+    graph : nx.DiGraph
+        Directed graph to traverse.
+    start_node : Any
+        Node from which to start the search.
+    distance : int
+        Maximum graph distance (number of edges) to traverse.
+    exclude_edges : Optional[List[Tuple[Any, Any]]]
+        List of directed edges to exclude, each as a (u, v) tuple.
+
+    Returns
+    -------
+    List[Tuple[Any, int]]
+        Sorted list of (node, distance) pairs.
+    """
+    # Normalize exclude_edges to a set for fast lookups
+    excluded: Set[Tuple[Any, Any]] = set(exclude_edges or [])
+
+    # BFS initialization
+    queue = deque([(start_node, 0)])
+    visited: Dict[Any, int] = {}
+
+    while queue:
+        node, dist = queue.popleft()
+        # Only process within the allowed distance
+        if dist > distance:
+            continue
+
+        # Record the shortest distance to this node
+        if node not in visited or dist < visited[node]:
+            visited[node] = dist
+
+        # Explore neighbors if we haven't reached max distance yet
+        if dist < distance:
+            for nbr in graph.neighbors(node):
+                # Skip self-loops
+                if nbr == node:
+                    continue
+                # Skip if edge is excluded
+                if (node, nbr) in excluded:
+                    continue
+                # Skip already-visited nodes at shorter or equal distance
+                if nbr in visited and visited[nbr] <= dist + 1:
+                    continue
+                queue.append((nbr, dist + 1))
+
+    # Return nodes sorted by distance
+    return sorted(visited.items(), key=lambda x: x[1])
+
+
+
+
+def get_node_neighbours(graph, start_node, distance, exclude_edges: List[str] = None):
+    neighbours = find_nodes_within_distance(graph, start_node, distance, exclude_edges)
+    max_distance = max(distance for _, distance in neighbours)
+    distance = min(distance, max_distance)
+    return [node for node, d in neighbours if d == distance]

{glam4cm-0.1.1.dist-info → glam4cm-1.0.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
-Metadata-Version: 2.2
+Metadata-Version: 2.4
 Name: glam4cm
-Version: 0.1.1
+Version: 1.0.0
 Summary: Graph Neural Networks and Language Models Trainer (Separate or combined) for conceptual models
 Author-email: Syed Juned Ali <syed.juned.ali@tuwien.ac.at>
 License: MIT License
@@ -47,6 +47,7 @@ Requires-Dist: fasttext
 Provides-Extra: dev
 Requires-Dist: pytest; extra == "dev"
 Requires-Dist: black; extra == "dev"
+Dynamic: license-file
 
 # glam4cm
 

glam4cm-1.0.0.dist-info/RECORD

@@ -0,0 +1,75 @@
+glam4cm/__init__.py,sha256=tiiFpZvXNLMFGNr3Abo3SuW0Gfj2Wc7wBsf37zeMVbQ,180
+glam4cm/run.py,sha256=vtOIbGWkUVzpZkeWzjW7N4ioCr1J_grjdNUZzQIhS0I,4259
+glam4cm/run_conf_v2.py,sha256=ixs7xzoly4MGugYbs-HNY_Mizs3_Ejq8W-CXC6PTtAc,14779
+glam4cm/run_configs.py,sha256=lGTrrk_3enri6geSRwMdwO7G7fIA2HNyD86Cl4BM92Y,2873
+glam4cm/run_confs.py,sha256=EOI6JcJbqocpd8wkUpsvUJgPUpSrEv6C3txfprrFuW8,1327
+glam4cm/settings.py,sha256=6G28EiL3S7mVdZ8ptQX5V0Oov6Ee2slbDfQZlbJEPmo,1364
+glam4cm/utils.py,sha256=pciZSCsO0ae-73bLTXvIP0sfY_vOKzHZm-nus0NA7PA,15464
+glam4cm/data_loading/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+glam4cm/data_loading/data.py,sha256=vYvALQMozAljrANdDTPutjN2C_augk8yhYVilEOv9bQ,20101
+glam4cm/data_loading/encoding.py,sha256=wawEIy5e6iPPMad6t13fnE6xL7IfJE5rqEMJgGoOxRU,2798
+glam4cm/data_loading/graph_dataset.py,sha256=V1Xk98LyZW-4K2ddFIWE0TozGlxJ-9YPFHMdqek1O5c,40510
+glam4cm/data_loading/metadata.py,sha256=EihqDhu7pJjw8jwn7pZwTD5SVWw9Ete7XihsH8b2HIA,1793
+glam4cm/data_loading/models_dataset.py,sha256=HEsMdsgp-YI15ZvNQYXYbT-G2xGmVUb-_l3IgBFeT3g,13636
+glam4cm/data_loading/utils.py,sha256=GM8DRHxjeAXAZWX1ITq1QfWWVY7jwdXGhbPx8L8V_IQ,907
+glam4cm/downstream_tasks/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+glam4cm/downstream_tasks/bert_edge_classification.py,sha256=kv6eft9kY2cb0mnvZook8uyMWXEtfhJFmiWjQqQTNEU,4926
+glam4cm/downstream_tasks/bert_graph_classification.py,sha256=q7Lz0eAmlVg6fNme9kNxqaWnJb_CVrwFrzTKGjfaz_A,5019
+glam4cm/downstream_tasks/bert_graph_classification_comp.py,sha256=8pAM5OqHpEovh2kO6j6hGu3tScoWeUAXE9JGI_KmoW8,5956
+glam4cm/downstream_tasks/bert_link_prediction.py,sha256=G5YxR41DG6VqJEM_M18u58T33VJpb-QiWdXJXF9dgpk,4554
+glam4cm/downstream_tasks/bert_node_classification.py,sha256=x_MfirCakbXuhI5_KxzgOQtatL2xhb2W38FOR1G4EkQ,6214
+glam4cm/downstream_tasks/cm_gpt_edge_classification.py,sha256=9VGjcFlsetndHoYrm03I1TDcw9PJin41oOBTwma6E5Y,2258
+glam4cm/downstream_tasks/cm_gpt_node_classification.py,sha256=KuW92x5gVkfmYl6FnFzdO9-cKBdqm6YcHnyOf4l9oqQ,3780
+glam4cm/downstream_tasks/cm_gpt_pretraining.py,sha256=qtsrvrE8t7HG9cbyiKI8bKRzzoWn-j-7Iq1UxzIf--Q,1685
+glam4cm/downstream_tasks/common_args.py,sha256=3hzgzNim5cc2AdvTrcyXyQuZPYeX0h44D6y4EKa0ycY,6510
+glam4cm/downstream_tasks/create_dataset.py,sha256=9ykTvvqX7f6j2zlhkU96fMrDgfLKvHQ5R9mH9pHxZ4c,1640
+glam4cm/downstream_tasks/gnn_edge_classification.py,sha256=q_dwKv3KiT6NDlndNzc189hRZapH9J_46gILY6wgh58,3952
+glam4cm/downstream_tasks/gnn_graph_cls.py,sha256=2hC6w2TXP_IOCNmFmljC0oq4I9ukIkn3nGMZah6XkC0,3988
+glam4cm/downstream_tasks/gnn_link_prediction.py,sha256=E_MIhgJmwY0pUIhPNON25xMBQcMdMEZXJ6PNSLcrWGc,3686
+glam4cm/downstream_tasks/gnn_node_classification.py,sha256=egg8Ngs9_C_p7WaTDZsMsBitrr4OWSYiSLIdc84eJuo,3658
+glam4cm/downstream_tasks/tf_idf_text_classification.py,sha256=_GUYIw86YM4q417IAazMaxZDOwDI2CrZAbQp7k0MAMg,691
+glam4cm/downstream_tasks/utils.py,sha256=b5L7UoGbuGu6HdpS72xp-FipXa07yjuhFdbeNszoSTA,1444
+glam4cm/downstream_tasks/word2vec_text_classification.py,sha256=tQMaSo9PVJlpqEnRBG7OEikUtOHckzA8FilnGk1N0zY,3531
+glam4cm/embeddings/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+glam4cm/embeddings/bert.py,sha256=4injUWPZO6DH_OSoSblSGO9q4FyEOoKF-KMPVvbdRck,2788
+glam4cm/embeddings/common.py,sha256=C-4aW1ynZUHOxvkAZFjV9klKkbbHao2hSFvNqr_lTvo,1226
+glam4cm/embeddings/fasttext.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+glam4cm/embeddings/tfidf.py,sha256=-R10-053tm9mjgXSNgIzRGIDeL0AJKgVx7UtoHoaGgw,737
+glam4cm/embeddings/w2v.py,sha256=acQX9wvtNRF8ghkniR2xikQj_KLBOYRYqQgvg3qD8l4,1384
+glam4cm/encoding/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+glam4cm/encoding/common.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+glam4cm/encoding/encoders.py,sha256=wBAhlgcXpIBhdYPH_WNDe7pD3kaaODGOSMIhpUmkRiE,2743
+glam4cm/graph2str/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+glam4cm/graph2str/common.py,sha256=vYNkLUM3rfmj0c-BsYfBQnUJGXyIC7B68HlYZrz8auY,1121
+glam4cm/graph2str/constants.py,sha256=LwWXWTwlS_Q4Blbi9vHgqNh8OjWM5X_z-prEkPtnOJI,239
+glam4cm/graph2str/ontouml.py,sha256=ZGDFTxSMTf2p4q0QbVXmY8jMSzZSpkvREt9VBuN--eg,5499
+glam4cm/graph2str/uml.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+glam4cm/lang2graph/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+glam4cm/lang2graph/archimate.py,sha256=jRA8zXOMb3gO5WvfNvP_R8-CYiQsdo9FmDyf8fDIcn4,678
+glam4cm/lang2graph/bpmn.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+glam4cm/lang2graph/common.py,sha256=2vE9dUmdnQ79YlFQXQV91rCWmF254yRQYnNsiyMuRTo,15232
+glam4cm/lang2graph/ecore.py,sha256=Mcy3muODODCRtM5Q-8F2f4-CZcGE2a33qzlwMjWq3Ws,8071
+glam4cm/lang2graph/ontouml.py,sha256=TiLHyFHj6aVsYJsItAD_n5IPWw7FM5X2ZtcZAbBq0Y4,7517
+glam4cm/lang2graph/utils.py,sha256=d0b6k4MNwnA9GWewaIwr9uS7YzgRuhSAWaXu-GE-JMg,2089
+glam4cm/models/cmgpt.py,sha256=2vnsYO73XZCKwHm0XTKgK8bQiVssF5tRIFr67E2NrCE,13038
+glam4cm/models/gnn_layers.py,sha256=3-ZLqoWSaTabmSIOKCqhQvyrpgPuv3TsHQRrKMnQ_kY,8664
+glam4cm/models/hf.py,sha256=1r0y9wYvIj0XOkNnwAD6XLgWQ90SICLyTAwAZaDDTMo,570
+glam4cm/tokenization/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+glam4cm/tokenization/special_tokens.py,sha256=63FJAKOw2nTmFu_VypFOiylZ1U2cItsdgXxpPIFrp-Q,445
+glam4cm/tokenization/utils.py,sha256=NTNPlJT7wHYdn3hFIFqqcs23N8bhzEhOpugRMR1mya0,1732
+glam4cm/trainers/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+glam4cm/trainers/bert_classifier.py,sha256=4yGMX1bkYsvBaTwyP3CV7U05uIYqQdDeUa84N8_uc3I,3599
+glam4cm/trainers/cm_gpt_trainer.py,sha256=6OpKyO7nTnUlTKnWR62JL-LOebdSuBpe8QwZ8SKaw80,5586
+glam4cm/trainers/gnn_edge_classifier.py,sha256=s7P_8CpuqOwU9ZbeC7P7wGQ_4uoc7XubsuiiDDz-oA8,4754
+glam4cm/trainers/gnn_graph_classifier.py,sha256=5Xb8GPkJSZInNMZs3tLFJyY-dcr_h3L8x89zOJnlw-c,4030
+glam4cm/trainers/gnn_link_predictor.py,sha256=1HguE_JZGyxOKu3_jrkyYkDqEqxMDBvgsBtr4sJZBSU,5821
+glam4cm/trainers/gnn_link_predictor_v2.py,sha256=9NZKVq66FxiMSSdfFOySRnHFYU3TqOywPZG9ViFTIYE,4930
+glam4cm/trainers/gnn_node_classifier.py,sha256=I4Rrx7m8oxevtBgelXuJbWUx33TBp1k12Y1to7oq3G8,4508
+glam4cm/trainers/gnn_trainer.py,sha256=Xdmk2yniIP2HYF59fjEDiy8hvJrvs1mqHM_R-cy8cZg,4259
+glam4cm/trainers/metrics.py,sha256=eSQdaJB-K-C4ipNsK18Jgf-9WTQmQBlSQL4c0sk8S0M,1859
+glam4cm-1.0.0.dist-info/licenses/LICENSE,sha256=NzIgfG9Z6TaC4HlY6ownebjdGY0DKUXSgF5sM7bmxZI,1071
+glam4cm-1.0.0.dist-info/METADATA,sha256=pxKwXMLuouZE_0J_LEHtvsKKHJJjEE-v--q2i_06gi0,3282
+glam4cm-1.0.0.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+glam4cm-1.0.0.dist-info/entry_points.txt,sha256=sZ-zOIJOyDP-vpTyVTTbdDyNando8uRVQmaFuAo_nuM,45
+glam4cm-1.0.0.dist-info/top_level.txt,sha256=6V4mFMBo1sE2bowD9n2sxYv_ao8IsS8rR1ArAhfpQ4w,8
+glam4cm-1.0.0.dist-info/RECORD,,

{glam4cm-0.1.1.dist-info → glam4cm-1.0.0.dist-info}/WHEEL

@@ -1,5 +1,5 @@
 Wheel-Version: 1.0
-Generator: setuptools (75.8.0)
+Generator: setuptools (80.9.0)
 Root-Is-Purelib: true
 Tag: py3-none-any
 

glam4cm-0.1.1.dist-info/RECORD

@@ -1,72 +0,0 @@
-glam4cm/__init__.py,sha256=7D0ayfMYZjOum3A9g2zP2FGabJTO4JUPQifMg9ALcWY,178
-glam4cm/run.py,sha256=fSd5oBpHxWLwskMfT6mohiROX9BdIGwmUT8BGxGV2fc,4188
-glam4cm/run_configs.py,sha256=100hGJm7tx2enEhSckNE-nxBdDtm8kZyu498WEnqhAU,10122
-glam4cm/settings.py,sha256=9PG4ZAqxCNg2Rt3JZ3ghl1_mzAw6vIVswC4JGY2Z2YE,1115
-glam4cm/utils.py,sha256=1qUTS2r0QB38y4UdcScAEMYEpyUohDqpgwoHJvCGbZY,5918
-glam4cm/data_loading/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-glam4cm/data_loading/data.py,sha256=rsH5vewA47km3I_pDS-MZoChK5Y7RSD7zQgtP8QhKhY,22088
-glam4cm/data_loading/encoding.py,sha256=3duFeOShibiG9WL2JGLMKTkBS9nvrAQV9dhk0GWbdaE,2396
-glam4cm/data_loading/graph_dataset.py,sha256=dSCCG8QXexpsVfpseLGxMEbs5qcYFYfVCsjnPTsUlYA,35061
-glam4cm/data_loading/metadata.py,sha256=LpWp7JU7NEIUbwHJ_-OIjJN8il3MvDT66gjJN7EQJXY,1799
-glam4cm/data_loading/models_dataset.py,sha256=mDfE15gDcEACD-rDIMM2Eltw5-xLXD55nvQ4Gws-9pM,12477
-glam4cm/data_loading/utils.py,sha256=GM8DRHxjeAXAZWX1ITq1QfWWVY7jwdXGhbPx8L8V_IQ,907
-glam4cm/downstream_tasks/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-glam4cm/downstream_tasks/bert_edge_classification.py,sha256=xV0lgUcdyVVe82MMkESH8iYATwcvktK3uH7TNJZOY70,4280
-glam4cm/downstream_tasks/bert_graph_classification.py,sha256=FfWwRQAXKDPkKsMNaJUcvcpO4D4IjuXE-3byjycwLRc,4177
-glam4cm/downstream_tasks/bert_graph_classification_comp.py,sha256=JUyi9zjbyo6we7mFKLg1DVXKW-W95JPwaONcNOjZ-aA,4613
-glam4cm/downstream_tasks/bert_link_prediction.py,sha256=QogzTERQFNjhYJaB1JT2_0817fWyYJUiodvszXFOGPI,3976
-glam4cm/downstream_tasks/bert_node_classification.py,sha256=706fRltkgn7ucu7Xft1HSksGZY5vQuMqmXWthRLE8tc,4754
-glam4cm/downstream_tasks/cm_gpt_edge_classification.py,sha256=9VGjcFlsetndHoYrm03I1TDcw9PJin41oOBTwma6E5Y,2258
-glam4cm/downstream_tasks/cm_gpt_node_classification.py,sha256=JS0ntdro_WsIAxpL_nR71W8_IAG4enLeFH2Ak7kKQkA,2358
-glam4cm/downstream_tasks/cm_gpt_pretraining.py,sha256=qtsrvrE8t7HG9cbyiKI8bKRzzoWn-j-7Iq1UxzIf--Q,1685
-glam4cm/downstream_tasks/common_args.py,sha256=zj-rxPprn_V3R7nxqNFTw-vIiIqD5FNKixNdkhj-Y2s,5667
-glam4cm/downstream_tasks/create_dataset.py,sha256=9ykTvvqX7f6j2zlhkU96fMrDgfLKvHQ5R9mH9pHxZ4c,1640
-glam4cm/downstream_tasks/gnn_edge_classification.py,sha256=hnncLte0fGIWyyxCvWMEPKguVikqseyIGfd-4eQMYCc,3280
-glam4cm/downstream_tasks/gnn_graph_cls.py,sha256=Fv5tyt0sK_FaHDaB_l5D6FBXdl5v5pbL2fuHB9yX7ns,3179
-glam4cm/downstream_tasks/gnn_link_prediction.py,sha256=zbflVSQSck41XAbUkLRDANFvRTrbUE9E3Kl47mATTGY,3196
-glam4cm/downstream_tasks/gnn_node_classification.py,sha256=qsz-ed26GY5MHbscu6UVnRBX-KeQcGmU76sMUfOf2Ps,3191
-glam4cm/downstream_tasks/tf_idf_text_classification.py,sha256=_GUYIw86YM4q417IAazMaxZDOwDI2CrZAbQp7k0MAMg,691
-glam4cm/downstream_tasks/utils.py,sha256=Hk-M_REZwwc4s0S2agj81hIFXOVqwEKBVcSkUM8ZLbw,989
-glam4cm/downstream_tasks/word2vec_text_classification.py,sha256=tQMaSo9PVJlpqEnRBG7OEikUtOHckzA8FilnGk1N0zY,3531
-glam4cm/embeddings/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-glam4cm/embeddings/bert.py,sha256=hKitE9zwuUJdpC_VjyI1Ub3SWzV1TI24MQOkhaFlyAM,2789
-glam4cm/embeddings/common.py,sha256=7NVe8Jlyg3jaBrM3pKxHUh8IBlLaXKl0SLebsHTUMY8,1126
-glam4cm/embeddings/fasttext.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-glam4cm/embeddings/tfidf.py,sha256=-R10-053tm9mjgXSNgIzRGIDeL0AJKgVx7UtoHoaGgw,737
-glam4cm/embeddings/w2v.py,sha256=acQX9wvtNRF8ghkniR2xikQj_KLBOYRYqQgvg3qD8l4,1384
-glam4cm/encoding/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-glam4cm/encoding/common.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-glam4cm/encoding/encoders.py,sha256=-qIMGiyljB8hukyVJf3D89gpkKkQNgHlluSeESmzzog,2735
-glam4cm/graph2str/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-glam4cm/graph2str/common.py,sha256=vYNkLUM3rfmj0c-BsYfBQnUJGXyIC7B68HlYZrz8auY,1121
-glam4cm/graph2str/constants.py,sha256=LwWXWTwlS_Q4Blbi9vHgqNh8OjWM5X_z-prEkPtnOJI,239
-glam4cm/graph2str/ontouml.py,sha256=ZGDFTxSMTf2p4q0QbVXmY8jMSzZSpkvREt9VBuN--eg,5499
-glam4cm/graph2str/uml.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-glam4cm/lang2graph/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-glam4cm/lang2graph/archimate.py,sha256=mg63U9OQB8LgDobW4ChRti8ya4UzfBVUbm-d8ljBhMw,839
-glam4cm/lang2graph/bpmn.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-glam4cm/lang2graph/common.py,sha256=snea2JOIt5xvy1wvn6rSCOMCefNTmjO0Ni3hf4-2d50,13148
-glam4cm/lang2graph/ecore.py,sha256=xzms1Bi3Wj6qsz960w_IK0X2dsNhRDBMPkDsaKEnT1M,8145
-glam4cm/lang2graph/ontouml.py,sha256=DAODrTZnEWTWWR_CB2j_KfRk4G0Sw6lovpAd1jInbRk,7485
-glam4cm/lang2graph/utils.py,sha256=d0b6k4MNwnA9GWewaIwr9uS7YzgRuhSAWaXu-GE-JMg,2089
-glam4cm/models/cmgpt.py,sha256=2vnsYO73XZCKwHm0XTKgK8bQiVssF5tRIFr67E2NrCE,13038
-glam4cm/models/gnn_layers.py,sha256=4hKiJTTartoXjS29hZEQxRWTFZ96FtDijUKT6dwe9lU,8127
-glam4cm/models/hf.py,sha256=BE5cnCdSnpUzq_3ww43AqkZCjG67r1ckPDPdF8yfOoQ,508
-glam4cm/tokenization/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-glam4cm/tokenization/special_tokens.py,sha256=tM2WJDSheURKXm7-5QDczMdHuomPEx6HLTW8BFO0EWs,107
-glam4cm/tokenization/utils.py,sha256=rrM2Owd2IQZAmErOHL5vTVDyVQoPZ-j8ztGt5VXK1fE,1206
-glam4cm/trainers/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-glam4cm/trainers/bert_classifier.py,sha256=4yGMX1bkYsvBaTwyP3CV7U05uIYqQdDeUa84N8_uc3I,3599
-glam4cm/trainers/cm_gpt_trainer.py,sha256=au3hXa9F8uEe-QBA139-Aw7IJ6GQEcjcfNcRgPF2TE8,5547
-glam4cm/trainers/gnn_edge_classifier.py,sha256=81SE7dzKeJG60fuOQMn7A4ydpDCVXwdEp-qgfpE40Ok,4249
-glam4cm/trainers/gnn_graph_classifier.py,sha256=vDfGzaklYrM-pXrj-9QrAJtuZWu3_xcKd0aQlwz7qRs,3873
-glam4cm/trainers/gnn_link_predictor.py,sha256=WMomOdmk7H2aBwDBugPPN5yLcjzgdlade0OKlNvYxEo,5093
-glam4cm/trainers/gnn_node_classifier.py,sha256=I4Rrx7m8oxevtBgelXuJbWUx33TBp1k12Y1to7oq3G8,4508
-glam4cm/trainers/gnn_trainer.py,sha256=r06rR1bsmuiIZhzVcuAx3X_R5u0lZKt1B3shgSdLbxA,3871
-glam4cm/trainers/metrics.py,sha256=LPFyRSAT50bPhQtI8yLK3VHuubxiIw6fTXkLUGTL3Ns,1823
-glam4cm-0.1.1.dist-info/LICENSE,sha256=NzIgfG9Z6TaC4HlY6ownebjdGY0DKUXSgF5sM7bmxZI,1071
-glam4cm-0.1.1.dist-info/METADATA,sha256=v9BrBomTVyTejdhYXvijkLpDjew2YgaHsTFf3tWdeZE,3260
-glam4cm-0.1.1.dist-info/WHEEL,sha256=In9FTNxeP60KnTkGw7wk6mJPYd_dQSjEZmXdBdMCI-8,91
-glam4cm-0.1.1.dist-info/entry_points.txt,sha256=sZ-zOIJOyDP-vpTyVTTbdDyNando8uRVQmaFuAo_nuM,45
-glam4cm-0.1.1.dist-info/top_level.txt,sha256=6V4mFMBo1sE2bowD9n2sxYv_ao8IsS8rR1ArAhfpQ4w,8
-glam4cm-0.1.1.dist-info/RECORD,,