Spaces:
Runtime error
Runtime error
Update app.py
Browse files
app.py
CHANGED
@@ -19,6 +19,7 @@ from sentence_transformers import SentenceTransformer
|
|
19 |
import faiss
|
20 |
import numpy as np
|
21 |
from PIL import Image
|
|
|
22 |
|
23 |
# Configure logging
|
24 |
logging.basicConfig(
|
@@ -36,26 +37,12 @@ DEFAULT_PORT = 7860
|
|
36 |
MODEL_CACHE_DIR = Path("model_cache")
|
37 |
TEMPLATE_DIR = Path("templates")
|
38 |
TEMP_DIR = Path("temp")
|
39 |
-
DATABASE_PATH = Path("code_database.json")
|
40 |
-
|
41 |
|
42 |
# Ensure directories exist
|
43 |
for directory in [MODEL_CACHE_DIR, TEMPLATE_DIR, TEMP_DIR]:
|
44 |
directory.mkdir(exist_ok=True, parents=True)
|
45 |
|
46 |
-
|
47 |
-
@dataclass
|
48 |
-
class Template:
|
49 |
-
code: str
|
50 |
-
description: str
|
51 |
-
components: List[str]
|
52 |
-
metadata: Dict[str, Any] = field(default_factory=dict)
|
53 |
-
version: str = "1.0"
|
54 |
-
|
55 |
-
class TemplateManager:
|
56 |
-
# ... (TemplateManager remains the same) ...
|
57 |
-
|
58 |
-
|
59 |
class RAGSystem:
|
60 |
def __init__(self, model_name: str = "gpt2", device: str = "cuda" if torch.cuda.is_available() else "cpu", embedding_model="all-mpnet-base-v2"):
|
61 |
try:
|
@@ -65,89 +52,91 @@ class RAGSystem:
|
|
65 |
self.pipe = pipeline("text-generation", model=self.model, tokenizer=self.tokenizer, device=self.device)
|
66 |
self.embedding_model = SentenceTransformer(embedding_model)
|
67 |
self.load_database()
|
|
|
68 |
except Exception as e:
|
69 |
-
logger.error(f"Error loading language model or embedding model: {e}.
|
70 |
self.pipe = None
|
71 |
self.embedding_model = None
|
72 |
self.code_embeddings = None
|
73 |
|
74 |
-
|
75 |
def load_database(self):
|
76 |
-
|
77 |
-
|
78 |
-
|
79 |
-
|
80 |
-
|
81 |
-
|
82 |
-
|
83 |
-
|
84 |
-
|
85 |
-
|
86 |
-
|
87 |
|
88 |
-
|
89 |
-
|
90 |
-
|
91 |
-
|
92 |
|
93 |
-
|
94 |
-
|
95 |
-
|
96 |
-
|
97 |
-
|
98 |
-
|
99 |
-
|
100 |
-
|
101 |
-
|
|
|
102 |
|
103 |
def add_to_database(self, code: str):
|
104 |
"""Adds a code snippet to the database"""
|
105 |
try:
|
106 |
-
|
107 |
-
|
108 |
-
|
109 |
-
|
110 |
-
|
111 |
-
|
112 |
-
|
113 |
except Exception as e:
|
114 |
-
|
115 |
-
|
116 |
|
|
|
117 |
def save_database(self):
|
118 |
-
|
119 |
-
|
120 |
-
|
121 |
-
|
122 |
-
|
123 |
-
|
124 |
-
|
125 |
|
126 |
def rebuild_embeddings(self):
|
127 |
-
|
128 |
-
|
129 |
-
|
130 |
-
|
131 |
-
|
132 |
-
|
133 |
-
|
134 |
-
|
135 |
-
|
136 |
-
|
137 |
-
|
138 |
-
|
139 |
|
140 |
def retrieve_similar_code(self, description: str, top_k: int = 3) -> List[str]:
|
141 |
-
|
142 |
-
|
143 |
-
|
144 |
-
|
145 |
-
|
146 |
-
|
147 |
-
|
148 |
-
|
149 |
-
|
150 |
-
|
|
|
|
|
151 |
|
152 |
def generate_code(self, description: str, template_code: str) -> str:
|
153 |
retrieved_codes = self.retrieve_similar_code(description)
|
@@ -156,11 +145,13 @@ class RAGSystem:
|
|
156 |
try:
|
157 |
generated_text = self.pipe(prompt, max_length=500, num_return_sequences=1)[0]['generated_text']
|
158 |
generated_code = generated_text.split("Generated Code:")[1].strip().split('```')[0]
|
|
|
159 |
return generated_code
|
160 |
except Exception as e:
|
161 |
logger.error(f"Error generating code with language model: {e}. Returning template code.")
|
162 |
return template_code
|
163 |
else:
|
|
|
164 |
return f"# Placeholder code generation. Description: {description}\n{template_code}"
|
165 |
|
166 |
def generate_interface(self, screenshot: Optional[Image.Image], description: str) -> str:
|
@@ -173,16 +164,23 @@ class RAGSystem:
|
|
173 |
try:
|
174 |
generated_text = self.pipe(prompt, max_length=500, num_return_sequences=1)[0]['generated_text']
|
175 |
generated_code = generated_text.split("```")[1].strip()
|
|
|
176 |
return generated_code
|
177 |
except Exception as e:
|
178 |
logger.error(f"Error generating interface with language model: {e}. Returning placeholder.")
|
179 |
return "import gradio as gr\n\ndemo = gr.Interface(fn=lambda x:x, inputs='text', outputs='text')\ndemo.launch()"
|
180 |
else:
|
|
|
181 |
return "import gradio as gr\n\ndemo = gr.Interface(fn=lambda x:x, inputs='text', outputs='text')\ndemo.launch()"
|
182 |
|
183 |
class PreviewManager:
|
184 |
-
|
|
|
185 |
|
|
|
|
|
|
|
|
|
186 |
|
187 |
class GradioInterface:
|
188 |
def __init__(self):
|
@@ -194,23 +192,18 @@ class GradioInterface:
|
|
194 |
|
195 |
def _extract_components(self, code: str) -> List[str]:
|
196 |
"""Extract components from the code."""
|
197 |
-
# This logic should analyze the code and extract components.
|
198 |
-
# For example, you might look for function definitions, classes, etc.
|
199 |
components = []
|
200 |
-
|
201 |
-
function_matches = re.findall(r'def (\w+)', code)
|
202 |
-
components.extend(function_matches)
|
203 |
-
|
204 |
-
# Simple regex to find class definitions
|
205 |
class_matches = re.findall(r'class (\w+)', code)
|
206 |
components.extend(class_matches)
|
207 |
-
|
208 |
-
# You can add more sophisticated logic here as needed
|
209 |
return components
|
210 |
|
211 |
def _get_template_choices(self) -> List[str]:
|
212 |
"""Get available template choices."""
|
213 |
-
|
|
|
|
|
214 |
|
215 |
def launch(self, **kwargs):
|
216 |
with gr.Blocks() as interface:
|
@@ -221,37 +214,35 @@ class GradioInterface:
|
|
221 |
template_choice = gr.Dropdown(label="Select Template", choices=self._get_template_choices(), value=None)
|
222 |
save_button = gr.Button("Save as Template")
|
223 |
|
224 |
-
# Generate code button action
|
225 |
generate_button.click(
|
226 |
fn=self.generate_code,
|
227 |
-
inputs=description_input,
|
228 |
outputs=code_output
|
229 |
)
|
230 |
|
231 |
-
# Save template button action
|
232 |
save_button.click(
|
233 |
fn=self.save_template,
|
234 |
inputs=[code_output, template_choice, description_input],
|
235 |
outputs=code_output
|
236 |
)
|
237 |
|
238 |
-
# Additional UI elements can be added here
|
239 |
gr.Markdown("### Preview")
|
240 |
preview_output = gr.Textbox(label="Preview", interactive=False)
|
241 |
-
self.preview_manager.update_preview(code_output)
|
242 |
|
243 |
-
# Update preview when code is generated
|
244 |
generate_button.click(
|
245 |
fn=lambda code: self.preview_manager.update_preview(code),
|
246 |
inputs=code_output,
|
247 |
outputs=preview_output
|
248 |
)
|
249 |
|
|
|
250 |
interface.launch(**kwargs)
|
251 |
|
252 |
-
def generate_code(self, description: str) -> str:
|
253 |
-
"""Generate code based on the description."""
|
254 |
-
template_code = "" #
|
|
|
255 |
return self.rag_system.generate_code(description, template_code)
|
256 |
|
257 |
def save_template(self, code: str, name: str, description: str) -> str:
|
@@ -261,25 +252,16 @@ class GradioInterface:
|
|
261 |
template = Template(code=code, description=description, components=components)
|
262 |
if self.template_manager.save_template(name, template):
|
263 |
self.rag_system.add_to_database(code) # Add code to the database
|
|
|
264 |
return f"✅ Template '{name}' saved successfully."
|
265 |
else:
|
|
|
266 |
return "❌ Failed to save template."
|
267 |
except Exception as e:
|
268 |
logger.error(f"Error saving template: {e}")
|
269 |
return f"❌ Error saving template: {str(e)}"
|
270 |
|
271 |
-
|
272 |
def main():
|
273 |
-
# Configure logging
|
274 |
-
logging.basicConfig(
|
275 |
-
level=logging.INFO,
|
276 |
-
format='%(asctime)s - %(name)s - %(levelname)s - %(message)s',
|
277 |
-
handlers=[
|
278 |
-
logging.StreamHandler(),
|
279 |
-
logging.FileHandler('gradio_builder.log')
|
280 |
-
]
|
281 |
-
)
|
282 |
-
logger = logging.getLogger(__name__)
|
283 |
logger.info("=== Application Startup ===")
|
284 |
|
285 |
try:
|
|
|
19 |
import faiss
|
20 |
import numpy as np
|
21 |
from PIL import Image
|
22 |
+
from templates import TemplateManager, Template # Import TemplateManager and Template
|
23 |
|
24 |
# Configure logging
|
25 |
logging.basicConfig(
|
|
|
37 |
MODEL_CACHE_DIR = Path("model_cache")
|
38 |
TEMPLATE_DIR = Path("templates")
|
39 |
TEMP_DIR = Path("temp")
|
40 |
+
DATABASE_PATH = Path("code_database.json") # Path for our simple database
|
|
|
41 |
|
42 |
# Ensure directories exist
|
43 |
for directory in [MODEL_CACHE_DIR, TEMPLATE_DIR, TEMP_DIR]:
|
44 |
directory.mkdir(exist_ok=True, parents=True)
|
45 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
46 |
class RAGSystem:
|
47 |
def __init__(self, model_name: str = "gpt2", device: str = "cuda" if torch.cuda.is_available() else "cpu", embedding_model="all-mpnet-base-v2"):
|
48 |
try:
|
|
|
52 |
self.pipe = pipeline("text-generation", model=self.model, tokenizer=self.tokenizer, device=self.device)
|
53 |
self.embedding_model = SentenceTransformer(embedding_model)
|
54 |
self.load_database()
|
55 |
+
logger.info("RAG system initialized successfully.")
|
56 |
except Exception as e:
|
57 |
+
logger.error(f"Error loading language model or embedding model: {e}. Falling back to placeholder generation.")
|
58 |
self.pipe = None
|
59 |
self.embedding_model = None
|
60 |
self.code_embeddings = None
|
61 |
|
|
|
62 |
def load_database(self):
|
63 |
+
"""Loads or creates the code database"""
|
64 |
+
if DATABASE_PATH.exists():
|
65 |
+
try:
|
66 |
+
with open(DATABASE_PATH, 'r', encoding='utf-8') as f:
|
67 |
+
self.database = json.load(f)
|
68 |
+
self.code_embeddings = np.array(self.database['embeddings'])
|
69 |
+
logger.info("Loaded code database from file.")
|
70 |
+
except (json.JSONDecodeError, KeyError) as e:
|
71 |
+
logger.error(f"Error loading code database: {e}. Creating new database.")
|
72 |
+
self.database = {'codes': [], 'embeddings': []}
|
73 |
+
self.code_embeddings = np.array([])
|
74 |
|
75 |
+
else:
|
76 |
+
logger.info("Code database does not exist. Creating new database.")
|
77 |
+
self.database = {'codes': [], 'embeddings': []}
|
78 |
+
self.code_embeddings = np.array([])
|
79 |
|
80 |
+
if self.embedding_model and len(self.database['codes']) != len(self.database['embeddings']):
|
81 |
+
logger.warning("Mismatch between number of codes and embeddings, rebuilding embeddings.")
|
82 |
+
self.rebuild_embeddings()
|
83 |
+
elif self.embedding_model is None:
|
84 |
+
logger.warning("Embeddings are not supported in this context.")
|
85 |
+
|
86 |
+
# Index the embeddings for efficient searching
|
87 |
+
if len(self.code_embeddings) > 0 and self.embedding_model:
|
88 |
+
self.index = faiss.IndexFlatL2(self.code_embeddings.shape[1]) # L2 distance
|
89 |
+
self.index.add(self.code_embeddings)
|
90 |
|
91 |
def add_to_database(self, code: str):
|
92 |
"""Adds a code snippet to the database"""
|
93 |
try:
|
94 |
+
embedding = self.embedding_model.encode(code)
|
95 |
+
self.database['codes'].append(code)
|
96 |
+
self.database['embeddings'].append(embedding.tolist())
|
97 |
+
self.code_embeddings = np.vstack((self.code_embeddings, embedding))
|
98 |
+
self.index.add(np.array([embedding])) # update FAISS index
|
99 |
+
self.save_database()
|
100 |
+
logger.info(f"Added code snippet to database. Total size: {len(self.database['codes'])}.")
|
101 |
except Exception as e:
|
102 |
+
logger.error(f"Error adding to database: {e}")
|
|
|
103 |
|
104 |
+
```python
|
105 |
def save_database(self):
|
106 |
+
"""Saves the database to a file"""
|
107 |
+
try:
|
108 |
+
with open(DATABASE_PATH, 'w', encoding='utf-8') as f:
|
109 |
+
json.dump(self.database, f, indent=2)
|
110 |
+
logger.info(f"Saved database to {DATABASE_PATH}.")
|
111 |
+
except Exception as e:
|
112 |
+
logger.error(f"Error saving database: {e}")
|
113 |
|
114 |
def rebuild_embeddings(self):
|
115 |
+
"""Rebuilds embeddings from the codes"""
|
116 |
+
try:
|
117 |
+
embeddings = self.embedding_model.encode(self.database['codes'])
|
118 |
+
self.code_embeddings = embeddings
|
119 |
+
self.database['embeddings'] = embeddings.tolist()
|
120 |
+
self.index = faiss.IndexFlatL2(embeddings.shape[1]) # L2 distance
|
121 |
+
self.index.add(embeddings)
|
122 |
+
self.save_database()
|
123 |
+
logger.info("Rebuilt and saved embeddings to the database.")
|
124 |
+
except Exception as e:
|
125 |
+
logger.error(f"Error rebuilding embeddings: {e}")
|
|
|
126 |
|
127 |
def retrieve_similar_code(self, description: str, top_k: int = 3) -> List[str]:
|
128 |
+
"""Retrieves similar code snippets from the database"""
|
129 |
+
if self.embedding_model is None:
|
130 |
+
logger.warning("Embedding model is not available. Cannot retrieve similar code.")
|
131 |
+
return []
|
132 |
+
try:
|
133 |
+
embedding = self.embedding_model.encode(description)
|
134 |
+
D, I = self.index.search(np.array([embedding]), top_k)
|
135 |
+
logger.info(f"Retrieved {top_k} similar code snippets for description: {description}.")
|
136 |
+
return [self.database['codes'][i] for i in I[0]]
|
137 |
+
except Exception as e:
|
138 |
+
logger.error(f"Error retrieving similar code: {e}")
|
139 |
+
return []
|
140 |
|
141 |
def generate_code(self, description: str, template_code: str) -> str:
|
142 |
retrieved_codes = self.retrieve_similar_code(description)
|
|
|
145 |
try:
|
146 |
generated_text = self.pipe(prompt, max_length=500, num_return_sequences=1)[0]['generated_text']
|
147 |
generated_code = generated_text.split("Generated Code:")[1].strip().split('```')[0]
|
148 |
+
logger.info("Code generated successfully.")
|
149 |
return generated_code
|
150 |
except Exception as e:
|
151 |
logger.error(f"Error generating code with language model: {e}. Returning template code.")
|
152 |
return template_code
|
153 |
else:
|
154 |
+
logger.warning("Text generation pipeline is not available. Returning placeholder code.")
|
155 |
return f"# Placeholder code generation. Description: {description}\n{template_code}"
|
156 |
|
157 |
def generate_interface(self, screenshot: Optional[Image.Image], description: str) -> str:
|
|
|
164 |
try:
|
165 |
generated_text = self.pipe(prompt, max_length=500, num_return_sequences=1)[0]['generated_text']
|
166 |
generated_code = generated_text.split("```")[1].strip()
|
167 |
+
logger.info("Interface code generated successfully.")
|
168 |
return generated_code
|
169 |
except Exception as e:
|
170 |
logger.error(f"Error generating interface with language model: {e}. Returning placeholder.")
|
171 |
return "import gradio as gr\n\ndemo = gr.Interface(fn=lambda x:x, inputs='text', outputs='text')\ndemo.launch()"
|
172 |
else:
|
173 |
+
logger.warning("Text generation pipeline is not available. Returning placeholder interface code.")
|
174 |
return "import gradio as gr\n\ndemo = gr.Interface(fn=lambda x:x, inputs='text', outputs='text')\ndemo.launch()"
|
175 |
|
176 |
class PreviewManager:
|
177 |
+
def __init__(self):
|
178 |
+
self.preview_code = ""
|
179 |
|
180 |
+
def update_preview(self, code: str):
|
181 |
+
"""Update the preview with the generated code."""
|
182 |
+
self.preview_code = code
|
183 |
+
logger.info("Preview updated with new code.")
|
184 |
|
185 |
class GradioInterface:
|
186 |
def __init__(self):
|
|
|
192 |
|
193 |
def _extract_components(self, code: str) -> List[str]:
|
194 |
"""Extract components from the code."""
|
|
|
|
|
195 |
components = []
|
196 |
+
function_matches = re.findall(r'def (\w+)', code components.extend(function_matches)
|
|
|
|
|
|
|
|
|
197 |
class_matches = re.findall(r'class (\w+)', code)
|
198 |
components.extend(class_matches)
|
199 |
+
logger.info(f"Extracted components: {components}")
|
|
|
200 |
return components
|
201 |
|
202 |
def _get_template_choices(self) -> List[str]:
|
203 |
"""Get available template choices."""
|
204 |
+
choices = list(self.template_manager.templates.keys())
|
205 |
+
logger.info(f"Available template choices: {choices}")
|
206 |
+
return choices
|
207 |
|
208 |
def launch(self, **kwargs):
|
209 |
with gr.Blocks() as interface:
|
|
|
214 |
template_choice = gr.Dropdown(label="Select Template", choices=self._get_template_choices(), value=None)
|
215 |
save_button = gr.Button("Save as Template")
|
216 |
|
|
|
217 |
generate_button.click(
|
218 |
fn=self.generate_code,
|
219 |
+
inputs=[description_input, template_choice],
|
220 |
outputs=code_output
|
221 |
)
|
222 |
|
|
|
223 |
save_button.click(
|
224 |
fn=self.save_template,
|
225 |
inputs=[code_output, template_choice, description_input],
|
226 |
outputs=code_output
|
227 |
)
|
228 |
|
|
|
229 |
gr.Markdown("### Preview")
|
230 |
preview_output = gr.Textbox(label="Preview", interactive=False)
|
231 |
+
self.preview_manager.update_preview(code_output)
|
232 |
|
|
|
233 |
generate_button.click(
|
234 |
fn=lambda code: self.preview_manager.update_preview(code),
|
235 |
inputs=code_output,
|
236 |
outputs=preview_output
|
237 |
)
|
238 |
|
239 |
+
logger.info("Launching Gradio interface.")
|
240 |
interface.launch(**kwargs)
|
241 |
|
242 |
+
def generate_code(self, description: str, template_choice: Optional[str]) -> str:
|
243 |
+
"""Generate code based on the description and selected template."""
|
244 |
+
template_code = self.template_manager.get_template(template_choice) if template_choice else "" # Get template code if selected
|
245 |
+
logger.info(f"Generating code for description: {description} with template: {template_choice}")
|
246 |
return self.rag_system.generate_code(description, template_code)
|
247 |
|
248 |
def save_template(self, code: str, name: str, description: str) -> str:
|
|
|
252 |
template = Template(code=code, description=description, components=components)
|
253 |
if self.template_manager.save_template(name, template):
|
254 |
self.rag_system.add_to_database(code) # Add code to the database
|
255 |
+
logger.info(f"Template '{name}' saved successfully.")
|
256 |
return f"✅ Template '{name}' saved successfully."
|
257 |
else:
|
258 |
+
logger.error("Failed to save template.")
|
259 |
return "❌ Failed to save template."
|
260 |
except Exception as e:
|
261 |
logger.error(f"Error saving template: {e}")
|
262 |
return f"❌ Error saving template: {str(e)}"
|
263 |
|
|
|
264 |
def main():
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
265 |
logger.info("=== Application Startup ===")
|
266 |
|
267 |
try:
|