Spaces:
Running
Running
Reality123b
commited on
Update app.py
Browse files
app.py
CHANGED
@@ -9,42 +9,57 @@ from PIL import Image
|
|
9 |
|
10 |
@dataclass
|
11 |
class ChatMessage:
|
|
|
12 |
role: str
|
13 |
content: str
|
14 |
|
15 |
def to_dict(self):
|
|
|
16 |
return {"role": self.role, "content": self.content}
|
17 |
|
18 |
class XylariaChat:
|
19 |
def __init__(self):
|
|
|
20 |
self.hf_token = os.getenv("HF_TOKEN")
|
21 |
if not self.hf_token:
|
22 |
raise ValueError("HuggingFace token not found in environment variables")
|
23 |
|
|
|
24 |
self.client = InferenceClient(
|
25 |
-
model="Qwen/QwQ-32B-Preview",
|
26 |
api_key=self.hf_token
|
27 |
)
|
28 |
|
29 |
-
|
|
|
30 |
self.image_api_headers = {"Authorization": f"Bearer {self.hf_token}"}
|
31 |
|
|
|
32 |
self.conversation_history = []
|
33 |
self.persistent_memory = {}
|
34 |
|
|
|
35 |
self.system_prompt = """You are a helpful and harmless assistant. You are Xylaria developed by Sk Md Saad Amin . You should think step-by-step."""
|
36 |
|
37 |
def store_information(self, key, value):
|
|
|
38 |
self.persistent_memory[key] = value
|
39 |
return f"Stored: {key} = {value}"
|
40 |
|
41 |
def retrieve_information(self, key):
|
|
|
42 |
return self.persistent_memory.get(key, "No information found for this key.")
|
43 |
|
44 |
def reset_conversation(self):
|
|
|
|
|
|
|
|
|
|
|
45 |
self.conversation_history = []
|
46 |
self.persistent_memory.clear()
|
47 |
|
|
|
48 |
try:
|
49 |
self.client = InferenceClient(
|
50 |
model="Qwen/QwQ-32B-Preview",
|
@@ -53,26 +68,39 @@ class XylariaChat:
|
|
53 |
except Exception as e:
|
54 |
print(f"Error resetting API client: {e}")
|
55 |
|
56 |
-
return None
|
57 |
|
58 |
def caption_image(self, image):
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
59 |
try:
|
|
|
60 |
if isinstance(image, str) and os.path.isfile(image):
|
61 |
with open(image, "rb") as f:
|
62 |
data = f.read()
|
|
|
63 |
elif isinstance(image, str):
|
|
|
64 |
if image.startswith('data:image'):
|
65 |
image = image.split(',')[1]
|
66 |
data = base64.b64decode(image)
|
|
|
67 |
else:
|
68 |
data = image.read()
|
69 |
|
|
|
70 |
response = requests.post(
|
71 |
self.image_api_url,
|
72 |
headers=self.image_api_headers,
|
73 |
data=data
|
74 |
)
|
75 |
|
|
|
76 |
if response.status_code == 200:
|
77 |
caption = response.json()[0].get('generated_text', 'No caption generated')
|
78 |
return caption
|
@@ -83,22 +111,46 @@ class XylariaChat:
|
|
83 |
return f"Error processing image: {str(e)}"
|
84 |
|
85 |
def perform_math_ocr(self, image_path):
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
86 |
try:
|
|
|
87 |
img = Image.open(image_path)
|
|
|
|
|
88 |
text = pytesseract.image_to_string(img)
|
|
|
|
|
89 |
return text.strip()
|
|
|
90 |
except Exception as e:
|
91 |
return f"Error during Math OCR: {e}"
|
92 |
-
|
93 |
def get_response(self, user_input, image=None):
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
94 |
try:
|
|
|
95 |
messages = []
|
96 |
|
|
|
97 |
messages.append(ChatMessage(
|
98 |
role="system",
|
99 |
content=self.system_prompt
|
100 |
).to_dict())
|
101 |
|
|
|
102 |
if self.persistent_memory:
|
103 |
memory_context = "Remembered Information:\n" + "\n".join(
|
104 |
[f"{k}: {v}" for k, v in self.persistent_memory.items()]
|
@@ -108,23 +160,29 @@ class XylariaChat:
|
|
108 |
content=memory_context
|
109 |
).to_dict())
|
110 |
|
|
|
111 |
for msg in self.conversation_history:
|
112 |
messages.append(msg)
|
113 |
|
|
|
114 |
if image:
|
115 |
image_caption = self.caption_image(image)
|
116 |
user_input = f"description of an image: {image_caption}\n\nUser's message about it: {user_input}"
|
117 |
|
|
|
118 |
messages.append(ChatMessage(
|
119 |
role="user",
|
120 |
content=user_input
|
121 |
).to_dict())
|
122 |
|
|
|
123 |
input_tokens = sum(len(msg['content'].split()) for msg in messages)
|
124 |
-
max_new_tokens = 16384 - input_tokens - 50
|
125 |
|
|
|
126 |
max_new_tokens = min(max_new_tokens, 10020)
|
127 |
|
|
|
128 |
stream = self.client.chat_completion(
|
129 |
messages=messages,
|
130 |
model="Qwen/QwQ-32B-Preview",
|
@@ -133,14 +191,20 @@ class XylariaChat:
|
|
133 |
top_p=0.9,
|
134 |
stream=True
|
135 |
)
|
136 |
-
|
137 |
return stream
|
138 |
-
|
139 |
except Exception as e:
|
140 |
print(f"Detailed error in get_response: {e}")
|
141 |
return f"Error generating response: {str(e)}"
|
142 |
|
143 |
def messages_to_prompt(self, messages):
|
|
|
|
|
|
|
|
|
|
|
|
|
144 |
prompt = ""
|
145 |
for msg in messages:
|
146 |
if msg["role"] == "system":
|
@@ -149,59 +213,68 @@ class XylariaChat:
|
|
149 |
prompt += f"<|user|>\n{msg['content']}<|end|>\n"
|
150 |
elif msg["role"] == "assistant":
|
151 |
prompt += f"<|assistant|>\n{msg['content']}<|end|>\n"
|
152 |
-
prompt += "<|assistant|>\n"
|
153 |
return prompt
|
154 |
-
|
|
|
155 |
def create_interface(self):
|
156 |
def streaming_response(message, chat_history, image_filepath, math_ocr_image_path):
|
|
|
157 |
ocr_text = ""
|
158 |
if math_ocr_image_path:
|
159 |
ocr_text = self.perform_math_ocr(math_ocr_image_path)
|
160 |
if ocr_text.startswith("Error"):
|
161 |
-
|
162 |
-
|
163 |
-
return
|
164 |
-
elif len(ocr_text) > 500:
|
165 |
-
ocr_text = "OCR output is too large to be processed."
|
166 |
-
updated_history = chat_history + [[{"role": "user", "content": message}, {"role": "assistant", "content": ocr_text}]]
|
167 |
yield "", updated_history, None, None
|
168 |
return
|
169 |
else:
|
170 |
message = f"Math OCR Result: {ocr_text}\n\nUser's message: {message}"
|
171 |
|
|
|
172 |
if image_filepath:
|
173 |
response_stream = self.get_response(message, image_filepath)
|
174 |
else:
|
175 |
response_stream = self.get_response(message)
|
|
|
176 |
|
|
|
177 |
if isinstance(response_stream, str):
|
178 |
-
|
|
|
179 |
yield "", updated_history, None, None
|
180 |
return
|
181 |
|
|
|
182 |
full_response = ""
|
183 |
-
updated_history = chat_history + [[
|
184 |
|
|
|
185 |
try:
|
186 |
for chunk in response_stream:
|
187 |
if chunk.choices and chunk.choices[0].delta and chunk.choices[0].delta.content:
|
188 |
chunk_content = chunk.choices[0].delta.content
|
189 |
full_response += chunk_content
|
190 |
-
|
191 |
-
|
|
|
192 |
yield "", updated_history, None, None
|
193 |
except Exception as e:
|
194 |
print(f"Streaming error: {e}")
|
195 |
-
|
|
|
196 |
yield "", updated_history, None, None
|
197 |
return
|
198 |
|
|
|
199 |
self.conversation_history.append(ChatMessage(role="user", content=message).to_dict())
|
200 |
self.conversation_history.append(ChatMessage(role="assistant", content=full_response).to_dict())
|
201 |
|
|
|
202 |
if len(self.conversation_history) > 10:
|
203 |
self.conversation_history = self.conversation_history[-10:]
|
204 |
|
|
|
205 |
custom_css = """
|
206 |
@import url('https://fonts.googleapis.com/css2?family=Inter:wght@300;400;500;600;700&display=swap');
|
207 |
body, .gradio-container {
|
@@ -215,6 +288,7 @@ class XylariaChat:
|
|
215 |
.gradio-container button {
|
216 |
font-family: 'Inter', sans-serif !important;
|
217 |
}
|
|
|
218 |
.image-container {
|
219 |
display: flex;
|
220 |
gap: 10px;
|
@@ -231,9 +305,11 @@ class XylariaChat:
|
|
231 |
max-height: 200px;
|
232 |
border-radius: 8px;
|
233 |
}
|
|
|
234 |
.clear-button {
|
235 |
display: none;
|
236 |
}
|
|
|
237 |
.chatbot-container .message {
|
238 |
opacity: 0;
|
239 |
animation: fadeIn 0.5s ease-in-out forwards;
|
@@ -248,27 +324,20 @@ class XylariaChat:
|
|
248 |
transform: translateY(0);
|
249 |
}
|
250 |
}
|
251 |
-
.gradio-accordion {
|
252 |
-
overflow: hidden;
|
253 |
-
transition: max-height 0.3s ease-in-out;
|
254 |
-
max-height: 0;
|
255 |
-
}
|
256 |
-
.gradio-accordion.open {
|
257 |
-
max-height: 500px;
|
258 |
-
}
|
259 |
"""
|
260 |
|
261 |
with gr.Blocks(theme='soft', css=custom_css) as demo:
|
|
|
262 |
with gr.Column():
|
263 |
chatbot = gr.Chatbot(
|
264 |
label="Xylaria 1.5 Senoa (EXPERIMENTAL)",
|
265 |
height=500,
|
266 |
show_copy_button=True,
|
267 |
-
type='messages'
|
268 |
)
|
269 |
|
270 |
-
|
271 |
-
|
|
|
272 |
with gr.Column(elem_classes="image-upload"):
|
273 |
img = gr.Image(
|
274 |
sources=["upload", "webcam"],
|
@@ -283,7 +352,9 @@ class XylariaChat:
|
|
283 |
label="Upload Image for Math OCR",
|
284 |
elem_classes="image-preview"
|
285 |
)
|
|
|
286 |
|
|
|
287 |
with gr.Row():
|
288 |
with gr.Column(scale=4):
|
289 |
txt = gr.Textbox(
|
@@ -293,10 +364,12 @@ class XylariaChat:
|
|
293 |
)
|
294 |
btn = gr.Button("Send", scale=1)
|
295 |
|
|
|
296 |
with gr.Row():
|
297 |
clear = gr.Button("Clear Conversation")
|
298 |
clear_memory = gr.Button("Clear Memory")
|
299 |
|
|
|
300 |
btn.click(
|
301 |
fn=streaming_response,
|
302 |
inputs=[txt, chatbot, img, math_ocr_img],
|
@@ -308,6 +381,7 @@ class XylariaChat:
|
|
308 |
outputs=[txt, chatbot, img, math_ocr_img]
|
309 |
)
|
310 |
|
|
|
311 |
clear.click(
|
312 |
fn=lambda: None,
|
313 |
inputs=None,
|
@@ -315,6 +389,7 @@ class XylariaChat:
|
|
315 |
queue=False
|
316 |
)
|
317 |
|
|
|
318 |
clear_memory.click(
|
319 |
fn=self.reset_conversation,
|
320 |
inputs=None,
|
@@ -322,30 +397,18 @@ class XylariaChat:
|
|
322 |
queue=False
|
323 |
)
|
324 |
|
325 |
-
|
326 |
-
() => {
|
327 |
-
const accordion = document.querySelector(".gradio-accordion");
|
328 |
-
|
329 |
-
if (accordion) {
|
330 |
-
const accordionHeader = accordion.querySelector(".label-wrap");
|
331 |
-
|
332 |
-
accordionHeader.addEventListener("click", () => {
|
333 |
-
accordion.classList.toggle("open");
|
334 |
-
});
|
335 |
-
}
|
336 |
-
}
|
337 |
-
""")
|
338 |
-
|
339 |
demo.load(self.reset_conversation, None, None)
|
340 |
|
341 |
return demo
|
342 |
|
|
|
343 |
def main():
|
344 |
chat = XylariaChat()
|
345 |
interface = chat.create_interface()
|
346 |
interface.launch(
|
347 |
-
share=
|
348 |
-
debug=True
|
349 |
)
|
350 |
|
351 |
if __name__ == "__main__":
|
|
|
9 |
|
10 |
@dataclass
|
11 |
class ChatMessage:
|
12 |
+
"""Custom ChatMessage class since huggingface_hub doesn't provide one"""
|
13 |
role: str
|
14 |
content: str
|
15 |
|
16 |
def to_dict(self):
|
17 |
+
"""Converts ChatMessage to a dictionary for JSON serialization."""
|
18 |
return {"role": self.role, "content": self.content}
|
19 |
|
20 |
class XylariaChat:
|
21 |
def __init__(self):
|
22 |
+
# Securely load HuggingFace token
|
23 |
self.hf_token = os.getenv("HF_TOKEN")
|
24 |
if not self.hf_token:
|
25 |
raise ValueError("HuggingFace token not found in environment variables")
|
26 |
|
27 |
+
# Initialize the inference client with the Qwen model
|
28 |
self.client = InferenceClient(
|
29 |
+
model="Qwen/QwQ-32B-Preview", # Using the specified model
|
30 |
api_key=self.hf_token
|
31 |
)
|
32 |
|
33 |
+
# Image captioning API setup
|
34 |
+
self.image_api_url = "https://api-inference.huggingface.co/models/microsoft/git-large-coco"
|
35 |
self.image_api_headers = {"Authorization": f"Bearer {self.hf_token}"}
|
36 |
|
37 |
+
# Initialize conversation history and persistent memory
|
38 |
self.conversation_history = []
|
39 |
self.persistent_memory = {}
|
40 |
|
41 |
+
# System prompt with more detailed instructions
|
42 |
self.system_prompt = """You are a helpful and harmless assistant. You are Xylaria developed by Sk Md Saad Amin . You should think step-by-step."""
|
43 |
|
44 |
def store_information(self, key, value):
|
45 |
+
"""Store important information in persistent memory"""
|
46 |
self.persistent_memory[key] = value
|
47 |
return f"Stored: {key} = {value}"
|
48 |
|
49 |
def retrieve_information(self, key):
|
50 |
+
"""Retrieve information from persistent memory"""
|
51 |
return self.persistent_memory.get(key, "No information found for this key.")
|
52 |
|
53 |
def reset_conversation(self):
|
54 |
+
"""
|
55 |
+
Completely reset the conversation history, persistent memory,
|
56 |
+
and clear API-side memory
|
57 |
+
"""
|
58 |
+
# Clear local memory
|
59 |
self.conversation_history = []
|
60 |
self.persistent_memory.clear()
|
61 |
|
62 |
+
# Reinitialize the client (not strictly necessary for the API, but can help with local state)
|
63 |
try:
|
64 |
self.client = InferenceClient(
|
65 |
model="Qwen/QwQ-32B-Preview",
|
|
|
68 |
except Exception as e:
|
69 |
print(f"Error resetting API client: {e}")
|
70 |
|
71 |
+
return None # To clear the chatbot interface
|
72 |
|
73 |
def caption_image(self, image):
|
74 |
+
"""
|
75 |
+
Caption an uploaded image using Hugging Face API
|
76 |
+
Args:
|
77 |
+
image (str): Base64 encoded image or file path
|
78 |
+
Returns:
|
79 |
+
str: Image caption or error message
|
80 |
+
"""
|
81 |
try:
|
82 |
+
# If image is a file path, read and encode
|
83 |
if isinstance(image, str) and os.path.isfile(image):
|
84 |
with open(image, "rb") as f:
|
85 |
data = f.read()
|
86 |
+
# If image is already base64 encoded
|
87 |
elif isinstance(image, str):
|
88 |
+
# Remove data URI prefix if present
|
89 |
if image.startswith('data:image'):
|
90 |
image = image.split(',')[1]
|
91 |
data = base64.b64decode(image)
|
92 |
+
# If image is a file-like object (unlikely with Gradio, but good to have)
|
93 |
else:
|
94 |
data = image.read()
|
95 |
|
96 |
+
# Send request to Hugging Face API
|
97 |
response = requests.post(
|
98 |
self.image_api_url,
|
99 |
headers=self.image_api_headers,
|
100 |
data=data
|
101 |
)
|
102 |
|
103 |
+
# Check response
|
104 |
if response.status_code == 200:
|
105 |
caption = response.json()[0].get('generated_text', 'No caption generated')
|
106 |
return caption
|
|
|
111 |
return f"Error processing image: {str(e)}"
|
112 |
|
113 |
def perform_math_ocr(self, image_path):
|
114 |
+
"""
|
115 |
+
Perform OCR on an image and return the extracted text.
|
116 |
+
Args:
|
117 |
+
image_path (str): Path to the image file.
|
118 |
+
Returns:
|
119 |
+
str: Extracted text from the image, or an error message.
|
120 |
+
"""
|
121 |
try:
|
122 |
+
# Open the image using Pillow library
|
123 |
img = Image.open(image_path)
|
124 |
+
|
125 |
+
# Use Tesseract to do OCR on the image
|
126 |
text = pytesseract.image_to_string(img)
|
127 |
+
|
128 |
+
# Remove leading/trailing whitespace and return
|
129 |
return text.strip()
|
130 |
+
|
131 |
except Exception as e:
|
132 |
return f"Error during Math OCR: {e}"
|
133 |
+
|
134 |
def get_response(self, user_input, image=None):
|
135 |
+
"""
|
136 |
+
Generate a response using chat completions with improved error handling
|
137 |
+
Args:
|
138 |
+
user_input (str): User's message
|
139 |
+
image (optional): Uploaded image
|
140 |
+
Returns:
|
141 |
+
Stream of chat completions or error message
|
142 |
+
"""
|
143 |
try:
|
144 |
+
# Prepare messages with conversation context and persistent memory
|
145 |
messages = []
|
146 |
|
147 |
+
# Add system prompt as first message
|
148 |
messages.append(ChatMessage(
|
149 |
role="system",
|
150 |
content=self.system_prompt
|
151 |
).to_dict())
|
152 |
|
153 |
+
# Add persistent memory context if available
|
154 |
if self.persistent_memory:
|
155 |
memory_context = "Remembered Information:\n" + "\n".join(
|
156 |
[f"{k}: {v}" for k, v in self.persistent_memory.items()]
|
|
|
160 |
content=memory_context
|
161 |
).to_dict())
|
162 |
|
163 |
+
# Convert existing conversation history to ChatMessage objects and then to dictionaries
|
164 |
for msg in self.conversation_history:
|
165 |
messages.append(msg)
|
166 |
|
167 |
+
# Process image if uploaded
|
168 |
if image:
|
169 |
image_caption = self.caption_image(image)
|
170 |
user_input = f"description of an image: {image_caption}\n\nUser's message about it: {user_input}"
|
171 |
|
172 |
+
# Add user input
|
173 |
messages.append(ChatMessage(
|
174 |
role="user",
|
175 |
content=user_input
|
176 |
).to_dict())
|
177 |
|
178 |
+
# Calculate available tokens
|
179 |
input_tokens = sum(len(msg['content'].split()) for msg in messages)
|
180 |
+
max_new_tokens = 16384 - input_tokens - 50 # Reserve some tokens for safety
|
181 |
|
182 |
+
# Limit max_new_tokens to prevent exceeding the total limit
|
183 |
max_new_tokens = min(max_new_tokens, 10020)
|
184 |
|
185 |
+
# Generate response with streaming
|
186 |
stream = self.client.chat_completion(
|
187 |
messages=messages,
|
188 |
model="Qwen/QwQ-32B-Preview",
|
|
|
191 |
top_p=0.9,
|
192 |
stream=True
|
193 |
)
|
194 |
+
|
195 |
return stream
|
196 |
+
|
197 |
except Exception as e:
|
198 |
print(f"Detailed error in get_response: {e}")
|
199 |
return f"Error generating response: {str(e)}"
|
200 |
|
201 |
def messages_to_prompt(self, messages):
|
202 |
+
"""
|
203 |
+
Convert a list of ChatMessage dictionaries to a single prompt string.
|
204 |
+
|
205 |
+
This is a simple implementation and you might need to adjust it
|
206 |
+
based on the specific requirements of the model you are using.
|
207 |
+
"""
|
208 |
prompt = ""
|
209 |
for msg in messages:
|
210 |
if msg["role"] == "system":
|
|
|
213 |
prompt += f"<|user|>\n{msg['content']}<|end|>\n"
|
214 |
elif msg["role"] == "assistant":
|
215 |
prompt += f"<|assistant|>\n{msg['content']}<|end|>\n"
|
216 |
+
prompt += "<|assistant|>\n" # Start of assistant's turn
|
217 |
return prompt
|
218 |
+
|
219 |
+
|
220 |
def create_interface(self):
|
221 |
def streaming_response(message, chat_history, image_filepath, math_ocr_image_path):
|
222 |
+
|
223 |
ocr_text = ""
|
224 |
if math_ocr_image_path:
|
225 |
ocr_text = self.perform_math_ocr(math_ocr_image_path)
|
226 |
if ocr_text.startswith("Error"):
|
227 |
+
# Handle OCR error
|
228 |
+
updated_history = chat_history + [[message, ocr_text]]
|
|
|
|
|
|
|
|
|
229 |
yield "", updated_history, None, None
|
230 |
return
|
231 |
else:
|
232 |
message = f"Math OCR Result: {ocr_text}\n\nUser's message: {message}"
|
233 |
|
234 |
+
# Check if an image was actually uploaded
|
235 |
if image_filepath:
|
236 |
response_stream = self.get_response(message, image_filepath)
|
237 |
else:
|
238 |
response_stream = self.get_response(message)
|
239 |
+
|
240 |
|
241 |
+
# Handle errors in get_response
|
242 |
if isinstance(response_stream, str):
|
243 |
+
# Return immediately with the error message
|
244 |
+
updated_history = chat_history + [[message, response_stream]]
|
245 |
yield "", updated_history, None, None
|
246 |
return
|
247 |
|
248 |
+
# Prepare for streaming response
|
249 |
full_response = ""
|
250 |
+
updated_history = chat_history + [[message, ""]]
|
251 |
|
252 |
+
# Streaming output
|
253 |
try:
|
254 |
for chunk in response_stream:
|
255 |
if chunk.choices and chunk.choices[0].delta and chunk.choices[0].delta.content:
|
256 |
chunk_content = chunk.choices[0].delta.content
|
257 |
full_response += chunk_content
|
258 |
+
|
259 |
+
# Update the last message in chat history with partial response
|
260 |
+
updated_history[-1][1] = full_response
|
261 |
yield "", updated_history, None, None
|
262 |
except Exception as e:
|
263 |
print(f"Streaming error: {e}")
|
264 |
+
# Display error in the chat interface
|
265 |
+
updated_history[-1][1] = f"Error during response: {e}"
|
266 |
yield "", updated_history, None, None
|
267 |
return
|
268 |
|
269 |
+
# Update conversation history
|
270 |
self.conversation_history.append(ChatMessage(role="user", content=message).to_dict())
|
271 |
self.conversation_history.append(ChatMessage(role="assistant", content=full_response).to_dict())
|
272 |
|
273 |
+
# Limit conversation history
|
274 |
if len(self.conversation_history) > 10:
|
275 |
self.conversation_history = self.conversation_history[-10:]
|
276 |
|
277 |
+
# Custom CSS for Inter font and improved styling
|
278 |
custom_css = """
|
279 |
@import url('https://fonts.googleapis.com/css2?family=Inter:wght@300;400;500;600;700&display=swap');
|
280 |
body, .gradio-container {
|
|
|
288 |
.gradio-container button {
|
289 |
font-family: 'Inter', sans-serif !important;
|
290 |
}
|
291 |
+
/* Image Upload Styling */
|
292 |
.image-container {
|
293 |
display: flex;
|
294 |
gap: 10px;
|
|
|
305 |
max-height: 200px;
|
306 |
border-radius: 8px;
|
307 |
}
|
308 |
+
/* Remove clear image buttons */
|
309 |
.clear-button {
|
310 |
display: none;
|
311 |
}
|
312 |
+
/* Animate chatbot messages */
|
313 |
.chatbot-container .message {
|
314 |
opacity: 0;
|
315 |
animation: fadeIn 0.5s ease-in-out forwards;
|
|
|
324 |
transform: translateY(0);
|
325 |
}
|
326 |
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
327 |
"""
|
328 |
|
329 |
with gr.Blocks(theme='soft', css=custom_css) as demo:
|
330 |
+
# Chat interface with improved styling
|
331 |
with gr.Column():
|
332 |
chatbot = gr.Chatbot(
|
333 |
label="Xylaria 1.5 Senoa (EXPERIMENTAL)",
|
334 |
height=500,
|
335 |
show_copy_button=True,
|
|
|
336 |
)
|
337 |
|
338 |
+
# Enhanced Image Upload Section
|
339 |
+
with gr.Accordion("Image Input", open=False):
|
340 |
+
with gr.Row(elem_classes="image-container"): # Use a Row for side-by-side layout
|
341 |
with gr.Column(elem_classes="image-upload"):
|
342 |
img = gr.Image(
|
343 |
sources=["upload", "webcam"],
|
|
|
352 |
label="Upload Image for Math OCR",
|
353 |
elem_classes="image-preview"
|
354 |
)
|
355 |
+
# Removed clear buttons as per requirement
|
356 |
|
357 |
+
# Input row with improved layout
|
358 |
with gr.Row():
|
359 |
with gr.Column(scale=4):
|
360 |
txt = gr.Textbox(
|
|
|
364 |
)
|
365 |
btn = gr.Button("Send", scale=1)
|
366 |
|
367 |
+
# Clear history and memory buttons
|
368 |
with gr.Row():
|
369 |
clear = gr.Button("Clear Conversation")
|
370 |
clear_memory = gr.Button("Clear Memory")
|
371 |
|
372 |
+
# Submit functionality with streaming and image support
|
373 |
btn.click(
|
374 |
fn=streaming_response,
|
375 |
inputs=[txt, chatbot, img, math_ocr_img],
|
|
|
381 |
outputs=[txt, chatbot, img, math_ocr_img]
|
382 |
)
|
383 |
|
384 |
+
# Clear conversation history
|
385 |
clear.click(
|
386 |
fn=lambda: None,
|
387 |
inputs=None,
|
|
|
389 |
queue=False
|
390 |
)
|
391 |
|
392 |
+
# Clear persistent memory and reset conversation
|
393 |
clear_memory.click(
|
394 |
fn=self.reset_conversation,
|
395 |
inputs=None,
|
|
|
397 |
queue=False
|
398 |
)
|
399 |
|
400 |
+
# Ensure memory is cleared when the interface is closed
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
401 |
demo.load(self.reset_conversation, None, None)
|
402 |
|
403 |
return demo
|
404 |
|
405 |
+
# Launch the interface
|
406 |
def main():
|
407 |
chat = XylariaChat()
|
408 |
interface = chat.create_interface()
|
409 |
interface.launch(
|
410 |
+
share=True, # Optional: create a public link
|
411 |
+
debug=True # Show detailed errors
|
412 |
)
|
413 |
|
414 |
if __name__ == "__main__":
|