Update app.py
Browse files
app.py
CHANGED
@@ -101,25 +101,20 @@ def calculate_statistics(embeddings):
|
|
101 |
|
102 |
import shutil
|
103 |
def upload_file(file, model_name, split_strategy, chunk_size, overlap_size, max_tokens, query, top_k):
|
104 |
-
#
|
105 |
-
|
106 |
-
|
107 |
-
|
108 |
-
|
109 |
-
# f.write(file.read()) # Read and write the file content
|
110 |
|
111 |
-
#
|
112 |
try:
|
113 |
chunk_size = int(chunk_size) # Convert chunk_size to int
|
114 |
overlap_size = int(overlap_size) # Convert overlap_size to int
|
115 |
except ValueError:
|
116 |
return {"error": "Chunk size and overlap size must be valid integers."}
|
117 |
|
118 |
-
|
119 |
-
overlap_size = 0 # Set a default overlap size
|
120 |
-
if chunk_size is None:
|
121 |
-
chunk_size = 100 # Set a default chunk size
|
122 |
-
|
123 |
embeddings, chunks = process_files(model_name, split_strategy, chunk_size, overlap_size, max_tokens)
|
124 |
|
125 |
# Perform search
|
@@ -127,7 +122,7 @@ def upload_file(file, model_name, split_strategy, chunk_size, overlap_size, max_
|
|
127 |
|
128 |
# Calculate statistics
|
129 |
stats = calculate_statistics(embeddings)
|
130 |
-
|
131 |
return {"results": results, "stats": stats}
|
132 |
|
133 |
|
|
|
101 |
|
102 |
import shutil
|
103 |
def upload_file(file, model_name, split_strategy, chunk_size, overlap_size, max_tokens, query, top_k):
|
104 |
+
# Ensure default values are set if None is passed
|
105 |
+
if chunk_size is None:
|
106 |
+
chunk_size = 100 # Set a default chunk size
|
107 |
+
if overlap_size is None:
|
108 |
+
overlap_size = 0 # Set a default overlap size
|
|
|
109 |
|
110 |
+
# Convert chunk_size and overlap_size to integers after checking for None
|
111 |
try:
|
112 |
chunk_size = int(chunk_size) # Convert chunk_size to int
|
113 |
overlap_size = int(overlap_size) # Convert overlap_size to int
|
114 |
except ValueError:
|
115 |
return {"error": "Chunk size and overlap size must be valid integers."}
|
116 |
|
117 |
+
# Process files and get embeddings
|
|
|
|
|
|
|
|
|
118 |
embeddings, chunks = process_files(model_name, split_strategy, chunk_size, overlap_size, max_tokens)
|
119 |
|
120 |
# Perform search
|
|
|
122 |
|
123 |
# Calculate statistics
|
124 |
stats = calculate_statistics(embeddings)
|
125 |
+
|
126 |
return {"results": results, "stats": stats}
|
127 |
|
128 |
|