From a374891afc56d2ce26baa6564572105807c895ee Mon Sep 17 00:00:00 2001
From: Jan Van Looy <jan.vanlooy@ml6.eu>
Date: Thu, 18 Jan 2024 11:14:41 +0100
Subject: [PATCH] Tiny fixes

---
 src/parameter_search.ipynb |  8 ++++----
 src/utils.py               | 10 ++++------
 2 files changed, 8 insertions(+), 10 deletions(-)

diff --git a/src/parameter_search.ipynb b/src/parameter_search.ipynb
index 3469361..6cdb36b 100644
--- a/src/parameter_search.ipynb
+++ b/src/parameter_search.ipynb
@@ -320,11 +320,11 @@
    "outputs": [],
    "source": [
     "searchable_index_params = {\n",
-    "    'chunk_size' : [128, 192, 256, 320],\n",
-    "    'chunk_overlap' : [32, 64, 96, 128, 160, 192],\n",
+    "    'chunk_size' : [128, 256, 384],\n",
+    "    'chunk_overlap' : [16, 64, 128],\n",
     "}\n",
     "searchable_shared_params = {\n",
-    "    'embed_model' : [(\"huggingface\",\"all-MiniLM-L6-v2\"),(\"huggingface\", \"BAAI/bge-base-en-v1.5\")]\n",
+    "    'embed_model' : [(\"huggingface\",\"all-MiniLM-L6-v2\")] # add more as tuples: ,(\"huggingface\", \"BAAI/bge-base-en-v1.5\")\n",
     "}\n",
     "searchable_eval_params = {\n",
     "    'retrieval_top_k' : [2, 4, 8]\n",
@@ -419,7 +419,7 @@
     "    search_method = search_method,\n",
     "    target_metric = target_metric,\n",
     "    evaluation_set_path=evaluation_set_path,\n",
-    "    debug=True,\n",
+    "    debug=False,\n",
     ")\n",
     "\n",
     "results = mysearch.run(weaviate_client)"
diff --git a/src/utils.py b/src/utils.py
index ca4d179..a621d55 100644
--- a/src/utils.py
+++ b/src/utils.py
@@ -187,23 +187,21 @@ def run(self, weaviate_client: weaviate.Client):
             )
 
             # run indexing pipeline
+            logger.info(
+                f"Starting indexing pipeline of RUN {run_count}/{n_runs} with {indexing_config}")
             self.runner.run(indexing_pipeline)
             
             check_weaviate_class_exists(
                 weaviate_client,
                 indexing_config["weaviate_class"],
             )
-            
-            logger.info(
-                f"Starting indexing pipeline of RUN {run_count}/{n_runs} with {indexing_config}")
 
             # run evaluation pipeline
+            logger.info(
+                f"Starting evaluation pipeline of run #{run_count} / {n_runs} with {evaluation_config}")
             self.runner.run(input=evaluation_pipeline, 
                             extra_volumes=self.extra_volumes,
             )
-            
-            logger.info(
-                f"Starting evaluation pipeline of run #{run_count} / {n_runs} with {evaluation_config}")
 
             # read metrics from pipeline output
             metrics = {}