bertrend 0.4.4__tar.gz → 0.4.6__tar.gz

{bertrend-0.4.4 → bertrend-0.4.6}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: bertrend
-Version: 0.4.4
+Version: 0.4.6
 Summary: BERTrend analyses topic evolution over time using state-of-the-art transformer models: topic modelling and LLM
 License: LICENSE.md
 License-File: AUTHORS.txt

{bertrend-0.4.4 → bertrend-0.4.6}/bertrend/BERTrend.py

@@ -782,6 +782,13 @@ class BERTrend:
         # Unserialize BERTrend object (using dill as an improvement of pickle for complex objects)
         with open(models_path / BERTREND_FILE, "rb") as f:
             bertrend = dill.load(f)
+
+        # Ensure compatibility with older HDBSCAN versions: cluster_selection_persistence attribute
+        if getattr(bertrend, "last_topic_model", None) is not None and not hasattr(
+            bertrend.last_topic_model.hdbscan_model, "cluster_selection_persistence"
+        ):
+            bertrend.last_topic_model.hdbscan_model.cluster_selection_persistence = 0.0
+
         return bertrend
 
     @classmethod

{bertrend-0.4.4 → bertrend-0.4.6}/bertrend/bertrend_apps/prospective_demo/feeds_config.py

@@ -342,6 +342,13 @@ def handle_delete(row_dict: dict):
             SCHEDULER_UTILS.remove_scheduled_training_for_user(
                 model_id=feed_id, user=st.session_state.username
             )
+            # Also remove report generation if it was scheduled
+            if SCHEDULER_UTILS.check_if_report_generation_active_for_user(
+                model_id=feed_id, user=st.session_state.username
+            ):
+                SCHEDULER_UTILS.remove_scheduled_report_generation_for_user(
+                    model_id=feed_id, user=st.session_state.username
+                )
             time.sleep(0.2)
             st.rerun()
     with col2:

{bertrend-0.4.4 → bertrend-0.4.6}/bertrend/bertrend_apps/services/queue_management/bertrend_worker.py

@@ -19,13 +19,17 @@ import aio_pika
 import pandas as pd
 from loguru import logger
 
+from bertrend import load_toml_config
 from bertrend.bertrend_apps.data_provider.data_provider_utils import (
     auto_scrape,
     generate_query_file,
     scrape,
     scrape_feed_from_config,
 )
-from bertrend.bertrend_apps.newsletters.newsletter_generation import process_newsletter
+from bertrend.bertrend_apps.newsletters.newsletter_generation import (
+    NEWSLETTER_SECTION,
+    process_newsletter,
+)
 from bertrend.bertrend_apps.prospective_demo.automated_report_generation import (
     generate_automated_report,
 )
@@ -51,6 +55,7 @@ from bertrend.bertrend_apps.services.queue_management.queue_manager import Queue
 from bertrend.bertrend_apps.services.queue_management.rabbitmq_config import (
     RabbitMQConfig,
 )
+from bertrend.bertrend_apps.services.utils.logging_utils import get_file_logger
 
 
 # Wrapper functions to match the core logic with the request models
@@ -81,7 +86,14 @@ async def handle_auto_scrape(req: AutoScrapeRequest):
 
 
 async def handle_scrape_feed(req: ScrapeFeedRequest):
-    return await asyncio.to_thread(scrape_feed_from_config, req.feed_cfg)
+    # Create a unique log file for this call
+    user = "" if not req.user else req.user
+    model_id = "" if not req.model_id else req.model_id
+    logger_id = get_file_logger(id="scrape_feed", user_name=user, model_id=model_id)
+    result = await asyncio.to_thread(scrape_feed_from_config, req.feed_cfg)
+    # Remove the logger to prevent writing to this file for other calls
+    logger.remove(logger_id)
+    return result
 
 
 async def handle_generate_query_file(req: GenerateQueryFileRequest):
@@ -96,34 +108,64 @@ async def handle_generate_query_file(req: GenerateQueryFileRequest):
 
 
 async def handle_train_new(req: TrainNewModelRequest):
-    return await asyncio.to_thread(
+    # Create a unique log file for this call
+    logger_id = get_file_logger(
+        id="train_new_model", user_name=req.user, model_id=req.model_id
+    )
+    result = await asyncio.to_thread(
         train_new_model, model_id=req.model_id, user_name=req.user
     )
+    # Remove the logger to prevent writing to this file for other calls
+    logger.remove(logger_id)
+    return result
 
 
 async def handle_regenerate(req: RegenerateRequest):
-    return await asyncio.to_thread(
+    # Create a unique log file for this call
+    logger_id = get_file_logger(
+        id="regenerate", user_name=req.user, model_id=req.model_id
+    )
+    result = await asyncio.to_thread(
         regenerate_models,
         model_id=req.model_id,
         user=req.user,
         with_analysis=req.with_analysis,
         since=pd.Timestamp(req.since) if req.since else None,
     )
+    # Remove the logger to prevent writing to this file for other calls
+    logger.remove(logger_id)
+    return result
 
 
 async def handle_generate_report(req: GenerateReportRequest):
-    return await asyncio.to_thread(
+    # Create a unique log file for this call
+    logger_id = get_file_logger(
+        id="generate_report", user_name=req.user, model_id=req.model_id
+    )
+    result = await asyncio.to_thread(
         generate_automated_report,
         user=req.user,
         model_id=req.model_id,
         reference_date=req.reference_date,
     )
+    # Remove the logger to prevent writing to this file for other calls
+    logger.remove(logger_id)
+    return result
 
 
 async def handle_generate_newsletters(req: NewsletterRequest):
-    return await asyncio.to_thread(
+    config = load_toml_config(req.newsletter_toml_path)
+    model_id = config.get(NEWSLETTER_SECTION).get("id")
+    # Create a unique log file for this call
+    logger_id = get_file_logger(
+        id="generate_newsletters", user_name="", model_id=model_id
+    )
+    result = await asyncio.to_thread(
         process_newsletter, req.newsletter_toml_path, req.data_feed_toml_path
     )
+    # Remove the logger to prevent writing to this file for other calls
+    logger.remove(logger_id)
+    return result
 
 
 # Mapping of endpoints to their handler functions and request models

{bertrend-0.4.4 → bertrend-0.4.6}/bertrend/config/bertopic_default_config.toml

@@ -27,6 +27,7 @@ min_samples = 5
 metric = "euclidean"
 cluster_selection_method = "eom"
 prediction_data = true
+cluster_selection_persistence = 0.0
 
 # CountVectorizer: https://scikit-learn.org/1.5/modules/generated/sklearn.feature_extraction.text.CountVectorizer.html
 [vectorizer_model]

{bertrend-0.4.4 → bertrend-0.4.6}/bertrend/services/scheduling/dashboard_scheduling.py

@@ -103,7 +103,6 @@ async def main_page():
     ui.colors(primary="#0052a1")
     dashboard = JobDashboard()
 
-    # --- LOGIC ---
     def update_display():
         f_users = user_filter.value
         f_models = model_filter.value
@@ -139,7 +138,7 @@ async def main_page():
         # 1. Update Chart Dashboard
         with chart_container:
             chart_options = {
-                "chart": {"type": "timeline", "height": 350, "zoomType": "x"},
+                "chart": {"type": "timeline", "height": 400, "zoomType": "x"},
                 "title": {
                     "text": "Job Schedule Timeline",
                     "style": {"fontSize": "14px"},
@@ -155,6 +154,11 @@ async def main_page():
                                 "name": f"/{j['endpoint']}",
                                 "label": f"{j['user']} | {j['model_id']}",
                                 "color": j["color"],
+                                "dataLabels": {
+                                    "style": {
+                                        "fontSize": "10px",
+                                    }
+                                },
                             }
                             for j in dashboard.filtered_jobs
                             if j["time"]
@@ -189,7 +193,7 @@ async def main_page():
                 for job in dashboard.filtered_jobs:
                     custom_color_name = f"c_{job['color'].lstrip('#')}"
                     ui.timeline_entry(
-                        f"Model: {job['model_id']}",
+                        f"Model: {job['model_id']}  |  (Job ID: {job['job_id']})",
                         title=f"/{job['endpoint']} ({job['user']})",
                         subtitle=job["time"],
                         color=custom_color_name,

{bertrend-0.4.4 → bertrend-0.4.6}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "bertrend"
-version = "0.4.4"
+version = "0.4.6"
 description = "BERTrend analyses topic evolution over time using state-of-the-art transformer models: topic modelling and LLM"
 license = "LICENSE.md"
 readme = "README.md"