Fix bugs

core/Dockerfile

@@ -3,4 +3,6 @@ WORKDIR /code
 COPY ./core /code/core
 COPY ./simulator /code/simulator
 RUN pip install --no-cache-dir -r core/requirements.txt
+ENV PYTHONUNBUFFERED=1
+ENV PYTHONIOENCODING=UTF-8
 CMD ["uvicorn", "core.main:app", "--host", "0.0.0.0", "--port", "8000", "--proxy-headers"]

core/functions/project/prescribe.py

@@ -26,15 +26,18 @@
 logger = logging.getLogger(__name__)
 
 
-def get_ongoing_dataset_result_key(file: BinaryIO, extension: str, seperator: str,
+def get_ongoing_dataset_result_key(file: BinaryIO | bytes, extension: str, seperator: str,
                                    db_project: project_model.Project) -> str:
     # Get the result key of the ongoing dataset
     result = ""
 
     try:
         temp_path = get_new_path(f"{path.TEMP_PATH}/", suffix=f".{extension}")
         with open(temp_path, "wb") as f:
-            f.write(file.read())
+            if isinstance(file, bytes):
+                f.write(file)
+            else:
+                f.write(file.read())
 
         # Get dataframe from file
         try:

core/services/event_log.py

@@ -48,9 +48,10 @@ def process_uploaded_event_log(file: UploadFile, separator: str, test: UploadFil
 
     # Save test file to memory
     if test and test.file:
+        content = test.file.read()
         memory.log_tests[db_event_log.id] = {
             "date": datetime.now(),
-            "file": deepcopy(test.file),
+            "file": content,
             "extension": get_extension(test.filename),
             "separator": separator
         }

core/services/project.py

@@ -251,6 +251,8 @@ def process_ongoing_dataset_result(project_id: int, result_key: str, background_
     logger.warning("Start to merge the result")
     for plugin_result in result["results"].values():
         for case_id, case_result in plugin_result.items():
+            if case_id not in result["cases"]:
+                continue
             result["cases"][case_id]["prescriptions"].append(case_result)
     logger.warning("Merge the result successfully")
 

plugins/causallift_resource_allocation/Dockerfile

@@ -3,4 +3,6 @@ WORKDIR /code
 COPY ./core /code/core
 COPY ./plugins /code/plugins
 RUN pip install --no-cache-dir -r plugins/causallift_resource_allocation/requirements.txt
+ENV PYTHONUNBUFFERED=1
+ENV PYTHONIOENCODING=UTF-8
 CMD ["python", "-m", "plugins.causallift_resource_allocation.main"]

plugins/causallift_treatment_effect/Dockerfile

@@ -3,4 +3,6 @@ WORKDIR /code
 COPY ./core /code/core
 COPY ./plugins /code/plugins
 RUN pip install --no-cache-dir -r plugins/causallift_treatment_effect/requirements.txt
+ENV PYTHONUNBUFFERED=1
+ENV PYTHONIOENCODING=UTF-8
 CMD ["python", "-m", "plugins.causallift_treatment_effect.main"]

plugins/causallift_treatment_effect/algorithm.py

@@ -113,15 +113,24 @@ def predict_df(self, df: DataFrame) -> dict:
 
         # Get the result for each length
         threads = []
-        for length, test_df in test_dfs.items():
-            training_df = self.get_data()["training_dfs"].get(length)
-            if training_df is None:
-                continue
-            t = Thread(target=self.get_result_thread, args=(self, result_dfs, length, training_df, test_df))
-            t.start()
-            threads.append(t)
-        for t in threads:
-            t.join()
+        count_of_length = len(test_dfs)
+        if count_of_length <= 50:
+            for length, test_df in test_dfs.items():
+                training_df = self.get_data()["training_dfs"].get(length)
+                if training_df is None:
+                    continue
+                t = Thread(target=self.get_result_thread, args=(self, result_dfs, length, training_df, test_df))
+                t.start()
+                threads.append(t)
+            for t in threads:
+                t.join()
+        else:
+            for length, test_df in test_dfs.items():
+                training_df = self.get_data()["training_dfs"].get(length)
+                if training_df is None:
+                    continue
+                result_df = self.get_result(training_df, test_df)
+                result_dfs[length] = result_df
 
         # Merge the result
         if len(result_dfs) <= 0:

plugins/knn_next_activity/Dockerfile

@@ -3,4 +3,6 @@ WORKDIR /code
 COPY ./core /code/core
 COPY ./plugins /code/plugins
 RUN pip install --no-cache-dir -r plugins/knn_next_activity/requirements.txt
+ENV PYTHONUNBUFFERED=1
+ENV PYTHONIOENCODING=UTF-8
 CMD ["python", "-m", "plugins.knn_next_activity.main"]

plugins/random_forest_alarm/Dockerfile

@@ -3,4 +3,6 @@ WORKDIR /code
 COPY ./core /code/core
 COPY ./plugins /code/plugins
 RUN pip install --no-cache-dir -r plugins/random_forest_alarm/requirements.txt
+ENV PYTHONUNBUFFERED=1
+ENV PYTHONIOENCODING=UTF-8
 CMD ["python", "-m", "plugins.random_forest_alarm.main"]

processor/Dockerfile

@@ -3,4 +3,6 @@ WORKDIR /code
 COPY ./core /code/core
 COPY ./processor /code/processor
 RUN pip install --no-cache-dir -r processor/requirements.txt
+ENV PYTHONUNBUFFERED=1
+ENV PYTHONIOENCODING=UTF-8
 CMD ["python", "-m", "processor.main"]