Upload 5 files
Browse files- app.py +0 -0
- middleware.py +4 -4
- milvus_manager.py +9 -6
- rag.py +179 -36
app.py
CHANGED
The diff for this file is too large to render.
See raw diff
|
|
middleware.py
CHANGED
@@ -46,16 +46,16 @@ class Middleware:
|
|
46 |
|
47 |
|
48 |
|
49 |
-
def search(self, search_queries: list[str]):
|
50 |
-
print(f"Searching for {len(search_queries)} queries")
|
51 |
|
52 |
final_res = []
|
53 |
|
54 |
for query in search_queries:
|
55 |
print(f"Searching for query: {query}")
|
56 |
query_vec = colpali_manager.process_text([query])[0]
|
57 |
-
search_res = self.milvus_manager.search(query_vec, topk=
|
58 |
-
print(f"Search result: {search_res} for query: {query}")
|
59 |
final_res.append(search_res)
|
60 |
|
61 |
return final_res
|
|
|
46 |
|
47 |
|
48 |
|
49 |
+
def search(self, search_queries: list[str], topk: int = 10):
|
50 |
+
print(f"Searching for {len(search_queries)} queries with topk={topk}")
|
51 |
|
52 |
final_res = []
|
53 |
|
54 |
for query in search_queries:
|
55 |
print(f"Searching for query: {query}")
|
56 |
query_vec = colpali_manager.process_text([query])[0]
|
57 |
+
search_res = self.milvus_manager.search(query_vec, topk=topk)
|
58 |
+
print(f"Search result: {len(search_res)} results for query: {query}")
|
59 |
final_res.append(search_res)
|
60 |
|
61 |
return final_res
|
milvus_manager.py
CHANGED
@@ -13,7 +13,7 @@ class MilvusManager:
|
|
13 |
dotenv_file = dotenv.find_dotenv()
|
14 |
dotenv.load_dotenv(dotenv_file)
|
15 |
|
16 |
-
self.client = MilvusClient(uri=
|
17 |
self.collection_name = collection_name
|
18 |
self.dim = dim
|
19 |
|
@@ -50,10 +50,13 @@ class MilvusManager:
|
|
50 |
|
51 |
index_params.add_index(
|
52 |
field_name="vector",
|
53 |
-
metric_type="COSINE",
|
54 |
-
index_type="IVF_FLAT",
|
55 |
index_name="vector_index",
|
56 |
-
|
|
|
|
|
|
|
|
|
|
|
57 |
)
|
58 |
|
59 |
self.client.create_index(
|
@@ -65,7 +68,7 @@ class MilvusManager:
|
|
65 |
collections = self.client.list_collections()
|
66 |
|
67 |
# Set search parameters (here, using Inner Product metric).
|
68 |
-
search_params = {"metric_type": "
|
69 |
|
70 |
# Set to store unique (doc_id, collection_name) pairs across all collections.
|
71 |
doc_collection_pairs = set()
|
@@ -121,7 +124,7 @@ class MilvusManager:
|
|
121 |
# Unload the collection after search to free memory.
|
122 |
self.client.release_collection(collection_name=collection)
|
123 |
|
124 |
-
return scores[:topk] if len(scores) >= topk else scores
|
125 |
"""
|
126 |
search_params = {"metric_type": "IP", "params": {}}
|
127 |
results = self.client.search(
|
|
|
13 |
dotenv_file = dotenv.find_dotenv()
|
14 |
dotenv.load_dotenv(dotenv_file)
|
15 |
|
16 |
+
self.client = MilvusClient(uri="http://localhost:19530", token="root:Milvus")
|
17 |
self.collection_name = collection_name
|
18 |
self.dim = dim
|
19 |
|
|
|
50 |
|
51 |
index_params.add_index(
|
52 |
field_name="vector",
|
|
|
|
|
53 |
index_name="vector_index",
|
54 |
+
index_type="HNSW", #use HNSW option if got more mem, if not use IVF for faster processing
|
55 |
+
metric_type=os.environ["metrictype"], #"IP"
|
56 |
+
params={
|
57 |
+
"M": int(os.environ["mnum"]), #M:16 for HNSW, capital M
|
58 |
+
"efConstruction": int(os.environ["efnum"]), #500 for HNSW
|
59 |
+
},
|
60 |
)
|
61 |
|
62 |
self.client.create_index(
|
|
|
68 |
collections = self.client.list_collections()
|
69 |
|
70 |
# Set search parameters (here, using Inner Product metric).
|
71 |
+
search_params = {"metric_type": os.environ["metrictype"], "params": {}} #default metric type is "IP"
|
72 |
|
73 |
# Set to store unique (doc_id, collection_name) pairs across all collections.
|
74 |
doc_collection_pairs = set()
|
|
|
124 |
# Unload the collection after search to free memory.
|
125 |
self.client.release_collection(collection_name=collection)
|
126 |
|
127 |
+
return scores[:topk] if len(scores) >= topk else scores #topk is the number of scores to return back
|
128 |
"""
|
129 |
search_params = {"metric_type": "IP", "params": {}}
|
130 |
results = self.client.search(
|
rag.py
CHANGED
@@ -1,27 +1,77 @@
|
|
1 |
import requests
|
2 |
import os
|
|
|
3 |
|
4 |
from typing import List
|
5 |
from utils import encode_image
|
6 |
from PIL import Image
|
|
|
7 |
import torch
|
8 |
import subprocess
|
9 |
import psutil
|
10 |
import torch
|
11 |
from transformers import AutoModel, AutoTokenizer
|
12 |
-
|
13 |
-
|
14 |
|
15 |
|
16 |
class Rag:
|
17 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
18 |
def get_answer_from_gemini(self, query, imagePaths):
|
|
|
19 |
|
20 |
print(f"Querying Gemini for query={query}, imagePaths={imagePaths}")
|
21 |
|
22 |
try:
|
23 |
-
genai.configure(api_key=
|
24 |
-
model = genai.GenerativeModel('gemini-2.
|
25 |
|
26 |
images = [Image.open(path) for path in imagePaths]
|
27 |
|
@@ -45,35 +95,10 @@ class Rag:
|
|
45 |
#import environ variables from .env
|
46 |
import dotenv
|
47 |
|
48 |
-
|
49 |
dotenv_file = dotenv.find_dotenv()
|
50 |
dotenv.load_dotenv(dotenv_file)
|
51 |
-
""" #scuffed local hf inference (transformers incompatible to colpali version req, use ollama, more reliable, easier to use plus web server ready)
|
52 |
-
print(f"Querying for query={query}, imagesPaths={imagesPaths}")
|
53 |
-
|
54 |
-
model = AutoModel.from_pretrained(
|
55 |
-
'openbmb/MiniCPM-o-2_6-int4',
|
56 |
-
trust_remote_code=True,
|
57 |
-
attn_implementation='flash_attention_2', # sdpa or flash_attention_2
|
58 |
-
torch_dtype=torch.bfloat16,
|
59 |
-
init_vision=True,
|
60 |
-
)
|
61 |
-
|
62 |
-
|
63 |
-
model = model.eval().cuda()
|
64 |
-
tokenizer = AutoTokenizer.from_pretrained('openbmb/MiniCPM-V-2_6-int4', trust_remote_code=True)
|
65 |
-
image = Image.open(imagesPaths[0]).convert('RGB')
|
66 |
|
67 |
-
msgs = [{'role': 'user', 'content': [image, query]}]
|
68 |
-
answer = model.chat(
|
69 |
-
image=None,
|
70 |
-
msgs=msgs,
|
71 |
-
tokenizer=tokenizer
|
72 |
-
)
|
73 |
-
print(answer)
|
74 |
-
return answer
|
75 |
-
"""
|
76 |
-
|
77 |
#ollama method below
|
78 |
|
79 |
torch.cuda.empty_cache() #release cuda so that ollama can use gpu!
|
@@ -82,31 +107,149 @@ class Rag:
|
|
82 |
os.environ['OLLAMA_FLASH_ATTENTION'] = os.environ['flashattn'] #int "1"
|
83 |
if os.environ['ollama'] == "minicpm-v":
|
84 |
os.environ['ollama'] = "minicpm-v:8b-2.6-q8_0" #set to quantized version
|
|
|
|
|
|
|
|
|
|
|
85 |
|
86 |
|
87 |
# Close model thread (colpali)
|
88 |
print(f"Querying OpenAI for query={query}, imagesPaths={imagesPaths}")
|
89 |
-
from ollama import chat
|
90 |
|
91 |
try:
|
92 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
93 |
response = chat(
|
94 |
-
model=
|
95 |
messages=[
|
96 |
{
|
97 |
'role': 'user',
|
98 |
-
'content':
|
99 |
'images': imagesPaths,
|
100 |
"temperature":float(os.environ['temperature']), #test if temp makes a diff
|
101 |
}
|
102 |
],
|
|
|
103 |
)
|
104 |
|
105 |
answer = response.message.content
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
106 |
|
107 |
-
print(answer)
|
|
|
108 |
|
109 |
-
return
|
110 |
|
111 |
except Exception as e:
|
112 |
print(f"An error occurred while querying OpenAI: {e}")
|
@@ -153,4 +296,4 @@ class Rag:
|
|
153 |
# query = "Based on attached images, how many new cases were reported during second wave peak"
|
154 |
# imagesPaths = ["covid_slides_page_8.png", "covid_slides_page_8.png"]
|
155 |
|
156 |
-
# rag.get_answer_from_gemini(query, imagesPaths)
|
|
|
1 |
import requests
|
2 |
import os
|
3 |
+
import re
|
4 |
|
5 |
from typing import List
|
6 |
from utils import encode_image
|
7 |
from PIL import Image
|
8 |
+
from ollama import chat
|
9 |
import torch
|
10 |
import subprocess
|
11 |
import psutil
|
12 |
import torch
|
13 |
from transformers import AutoModel, AutoTokenizer
|
14 |
+
from google import genai
|
|
|
15 |
|
16 |
|
17 |
class Rag:
|
18 |
|
19 |
+
def _clean_raw_token_response(self, response_text):
|
20 |
+
"""
|
21 |
+
Clean raw token responses that contain undecoded token IDs
|
22 |
+
This handles cases where models return raw tokens instead of decoded text
|
23 |
+
"""
|
24 |
+
if not response_text:
|
25 |
+
return response_text
|
26 |
+
|
27 |
+
# Check if response contains raw token patterns
|
28 |
+
token_patterns = [
|
29 |
+
r'<unused\d+>', # unused tokens
|
30 |
+
r'<bos>', # beginning of sequence
|
31 |
+
r'<eos>', # end of sequence
|
32 |
+
r'<unk>', # unknown tokens
|
33 |
+
r'<mask>', # mask tokens
|
34 |
+
r'<pad>', # padding tokens
|
35 |
+
r'\[multimodal\]', # multimodal tokens
|
36 |
+
]
|
37 |
+
|
38 |
+
# If response contains raw tokens, try to clean them
|
39 |
+
has_raw_tokens = any(re.search(pattern, response_text) for pattern in token_patterns)
|
40 |
+
|
41 |
+
if has_raw_tokens:
|
42 |
+
print("⚠️ Detected raw token response, attempting to clean...")
|
43 |
+
|
44 |
+
# Remove common raw token patterns
|
45 |
+
cleaned_text = response_text
|
46 |
+
|
47 |
+
# Remove unused tokens
|
48 |
+
cleaned_text = re.sub(r'<unused\d+>', '', cleaned_text)
|
49 |
+
|
50 |
+
# Remove special tokens
|
51 |
+
cleaned_text = re.sub(r'<(bos|eos|unk|mask|pad)>', '', cleaned_text)
|
52 |
+
|
53 |
+
# Remove multimodal tokens
|
54 |
+
cleaned_text = re.sub(r'\[multimodal\]', '', cleaned_text)
|
55 |
+
|
56 |
+
# Clean up extra whitespace
|
57 |
+
cleaned_text = re.sub(r'\s+', ' ', cleaned_text).strip()
|
58 |
+
|
59 |
+
# If we still have mostly tokens, return an error message
|
60 |
+
if len(cleaned_text.strip()) < 10:
|
61 |
+
return "❌ **Model Response Error**: The model returned raw token IDs instead of decoded text. This may be due to model configuration issues. Please try:\n\n1. Restarting the Ollama server\n2. Using a different model\n3. Checking model compatibility with multimodal inputs"
|
62 |
+
|
63 |
+
return cleaned_text
|
64 |
+
|
65 |
+
return response_text
|
66 |
+
|
67 |
def get_answer_from_gemini(self, query, imagePaths):
|
68 |
+
|
69 |
|
70 |
print(f"Querying Gemini for query={query}, imagePaths={imagePaths}")
|
71 |
|
72 |
try:
|
73 |
+
genai.configure(api_key='AIzaSyCwRr9054tCuh2S8yGpwKFvOAxYMT4WNIs')
|
74 |
+
model = genai.GenerativeModel('gemini-2.0-flash')
|
75 |
|
76 |
images = [Image.open(path) for path in imagePaths]
|
77 |
|
|
|
95 |
#import environ variables from .env
|
96 |
import dotenv
|
97 |
|
98 |
+
# Load the .env file
|
99 |
dotenv_file = dotenv.find_dotenv()
|
100 |
dotenv.load_dotenv(dotenv_file)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
101 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
102 |
#ollama method below
|
103 |
|
104 |
torch.cuda.empty_cache() #release cuda so that ollama can use gpu!
|
|
|
107 |
os.environ['OLLAMA_FLASH_ATTENTION'] = os.environ['flashattn'] #int "1"
|
108 |
if os.environ['ollama'] == "minicpm-v":
|
109 |
os.environ['ollama'] = "minicpm-v:8b-2.6-q8_0" #set to quantized version
|
110 |
+
elif os.environ['ollama'] == "gemma3":
|
111 |
+
os.environ['ollama'] = "gemma3:12b" #set to upscaled version
|
112 |
+
# Add specific environment variables for Gemma3 to prevent raw token issues
|
113 |
+
os.environ['OLLAMA_KEEP_ALIVE'] = "5m"
|
114 |
+
os.environ['OLLAMA_ORIGINS'] = "*"
|
115 |
|
116 |
|
117 |
# Close model thread (colpali)
|
118 |
print(f"Querying OpenAI for query={query}, imagesPaths={imagesPaths}")
|
|
|
119 |
|
120 |
try:
|
121 |
|
122 |
+
# Enhanced prompt for more detailed responses with explicit page usage
|
123 |
+
enhanced_query = f"""
|
124 |
+
Please provide a comprehensive and detailed answer to the following query.
|
125 |
+
Use ALL available information from the provided document images to give a thorough response.
|
126 |
+
|
127 |
+
Query: {query}
|
128 |
+
|
129 |
+
CRITICAL INSTRUCTIONS:
|
130 |
+
- You have been provided with {len(imagesPaths)} document page(s)
|
131 |
+
- You MUST reference information from ALL {len(imagesPaths)} page(s) in your response
|
132 |
+
- Do not skip any pages - each page contains relevant information
|
133 |
+
- If you mention one page, you must also mention the others
|
134 |
+
- Ensure your response reflects the complete information from all pages
|
135 |
+
|
136 |
+
Instructions for detailed response:
|
137 |
+
1. Provide extensive background information and context
|
138 |
+
2. Include specific details, examples, and data points from ALL documents
|
139 |
+
3. Explain concepts thoroughly with step-by-step breakdowns
|
140 |
+
4. Provide comprehensive analysis rather than simple answers when requested
|
141 |
+
5. Explicitly reference each page and what information it contributes
|
142 |
+
6. Cross-reference information between pages when relevant
|
143 |
+
7. Ensure no page is left unmentioned in your analysis
|
144 |
+
|
145 |
+
SPECIAL INSTRUCTIONS FOR TABULAR DATA:
|
146 |
+
- If the query requests a table, list, or structured data, organize your response in a clear, structured format
|
147 |
+
- Use numbered lists, bullet points, or clear categories when appropriate
|
148 |
+
- Include specific data points or comparisons when available
|
149 |
+
- Structure information in a way that can be easily converted to a table format
|
150 |
+
|
151 |
+
IMPORTANT: Respond with natural, human-readable text only. Do not include any special tokens, codes, or technical identifiers in your response.
|
152 |
+
|
153 |
+
Make sure to acknowledge and use information from all {len(imagesPaths)} provided pages.
|
154 |
+
"""
|
155 |
+
|
156 |
+
# Try with current model first
|
157 |
+
current_model = os.environ['ollama']
|
158 |
+
|
159 |
+
# Set different options based on the model
|
160 |
+
if "gemma3" in current_model.lower():
|
161 |
+
# Specific options for Gemma3 to prevent raw token issues
|
162 |
+
model_options = {
|
163 |
+
"num_predict": 1024, # Shorter responses for Gemma3
|
164 |
+
"stop": ["<eos>", "<|endoftext|>", "</s>", "<|im_end|>"], # More stop tokens
|
165 |
+
"top_k": 20, # Lower top_k for more focused generation
|
166 |
+
"top_p": 0.8, # Lower top_p for more deterministic output
|
167 |
+
"repeat_penalty": 1.2, # Higher repeat penalty
|
168 |
+
"seed": 42, # Consistent results
|
169 |
+
"temperature": 0.7, # Lower temperature for more focused responses
|
170 |
+
}
|
171 |
+
else:
|
172 |
+
# Default options for other models
|
173 |
+
model_options = {
|
174 |
+
"num_predict": 2048, # Limit response length
|
175 |
+
"stop": ["<eos>", "<|endoftext|>", "</s>"], # Stop at end tokens
|
176 |
+
"top_k": 40, # Reduce randomness
|
177 |
+
"top_p": 0.9, # Nucleus sampling
|
178 |
+
"repeat_penalty": 1.1, # Prevent repetition
|
179 |
+
"seed": 42, # Consistent results
|
180 |
+
}
|
181 |
+
|
182 |
response = chat(
|
183 |
+
model=current_model,
|
184 |
messages=[
|
185 |
{
|
186 |
'role': 'user',
|
187 |
+
'content': enhanced_query,
|
188 |
'images': imagesPaths,
|
189 |
"temperature":float(os.environ['temperature']), #test if temp makes a diff
|
190 |
}
|
191 |
],
|
192 |
+
options=model_options
|
193 |
)
|
194 |
|
195 |
answer = response.message.content
|
196 |
+
|
197 |
+
# Clean the response to handle raw token issues
|
198 |
+
cleaned_answer = self._clean_raw_token_response(answer)
|
199 |
+
|
200 |
+
# If the cleaned answer is still problematic, try fallback models
|
201 |
+
if cleaned_answer and "❌ **Model Response Error**" in cleaned_answer:
|
202 |
+
print(f"⚠️ Primary model {current_model} failed, trying fallback models...")
|
203 |
+
|
204 |
+
# List of fallback models to try
|
205 |
+
fallback_models = [
|
206 |
+
"llama3.2-vision:latest",
|
207 |
+
"llava:latest",
|
208 |
+
"bakllava:latest",
|
209 |
+
"llama3.2:latest"
|
210 |
+
]
|
211 |
+
|
212 |
+
for fallback_model in fallback_models:
|
213 |
+
try:
|
214 |
+
print(f"🔄 Trying fallback model: {fallback_model}")
|
215 |
+
response = chat(
|
216 |
+
model=fallback_model,
|
217 |
+
messages=[
|
218 |
+
{
|
219 |
+
'role': 'user',
|
220 |
+
'content': enhanced_query,
|
221 |
+
'images': imagesPaths,
|
222 |
+
"temperature":float(os.environ['temperature']),
|
223 |
+
}
|
224 |
+
],
|
225 |
+
options={
|
226 |
+
"num_predict": 2048,
|
227 |
+
"stop": ["<eos>", "<|endoftext|>", "</s>"],
|
228 |
+
"top_k": 40,
|
229 |
+
"top_p": 0.9,
|
230 |
+
"repeat_penalty": 1.1,
|
231 |
+
"seed": 42,
|
232 |
+
}
|
233 |
+
)
|
234 |
+
|
235 |
+
fallback_answer = response.message.content
|
236 |
+
cleaned_fallback = self._clean_raw_token_response(fallback_answer)
|
237 |
+
|
238 |
+
if cleaned_fallback and "❌ **Model Response Error**" not in cleaned_fallback:
|
239 |
+
print(f"✅ Fallback model {fallback_model} succeeded")
|
240 |
+
return cleaned_fallback
|
241 |
+
|
242 |
+
except Exception as fallback_error:
|
243 |
+
print(f"❌ Fallback model {fallback_model} failed: {fallback_error}")
|
244 |
+
continue
|
245 |
+
|
246 |
+
# If all fallbacks fail, return the original error
|
247 |
+
return cleaned_answer
|
248 |
|
249 |
+
print(f"Original response: {answer}")
|
250 |
+
print(f"Cleaned response: {cleaned_answer}")
|
251 |
|
252 |
+
return cleaned_answer
|
253 |
|
254 |
except Exception as e:
|
255 |
print(f"An error occurred while querying OpenAI: {e}")
|
|
|
296 |
# query = "Based on attached images, how many new cases were reported during second wave peak"
|
297 |
# imagesPaths = ["covid_slides_page_8.png", "covid_slides_page_8.png"]
|
298 |
|
299 |
+
# rag.get_answer_from_gemini(query, imagesPaths)
|