Update app.py
Browse files
app.py
CHANGED
@@ -100,7 +100,7 @@ def calculate_statistics(embeddings):
|
|
100 |
return {"tokens": len(embeddings), "time_taken": time.time()}
|
101 |
|
102 |
import shutil
|
103 |
-
def upload_file(file, model_name, split_strategy, chunk_size
|
104 |
# Create the full destination file path
|
105 |
#file_path = os.path.join(FILES_DIR, file.name)
|
106 |
|
@@ -109,6 +109,12 @@ def upload_file(file, model_name, split_strategy, chunk_size=100, overlap_size=0
|
|
109 |
# f.write(file.read()) # Read and write the file content
|
110 |
|
111 |
# Process files and get embeddings
|
|
|
|
|
|
|
|
|
|
|
|
|
112 |
embeddings, chunks = process_files(model_name, split_strategy, chunk_size, overlap_size, max_tokens)
|
113 |
|
114 |
# Perform search
|
|
|
100 |
return {"tokens": len(embeddings), "time_taken": time.time()}
|
101 |
|
102 |
import shutil
|
103 |
+
def upload_file(file, model_name, split_strategy, chunk_size, overlap_size, max_tokens, query, top_k):
|
104 |
# Create the full destination file path
|
105 |
#file_path = os.path.join(FILES_DIR, file.name)
|
106 |
|
|
|
109 |
# f.write(file.read()) # Read and write the file content
|
110 |
|
111 |
# Process files and get embeddings
|
112 |
+
|
113 |
+
if overlap_size is None:
|
114 |
+
overlap_size = 0 # Set a default overlap size
|
115 |
+
if chunk_size is None:
|
116 |
+
chunk_size = 200 # Set a default chunk size
|
117 |
+
|
118 |
embeddings, chunks = process_files(model_name, split_strategy, chunk_size, overlap_size, max_tokens)
|
119 |
|
120 |
# Perform search
|