Spaces:
Running
Running
File size: 22,047 Bytes
43b66f1 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 |
import streamlit as st
import os
import pandas as pd
import numpy as np
import plotly.express as px
import json
from pathlib import Path
# Make sure necessary directories exist
os.makedirs('assets', exist_ok=True)
os.makedirs('database/data', exist_ok=True)
os.makedirs('fine_tuned_models', exist_ok=True)
# Page configuration
st.set_page_config(
page_title="ML Dataset & Code Generation Manager",
page_icon="🤗",
layout="wide",
initial_sidebar_state="expanded",
)
def load_css():
"""Load custom CSS styles"""
css_dir = Path("assets")
css_path = css_dir / "custom.css"
if not css_path.exists():
# Create assets directory if it doesn't exist
css_dir.mkdir(exist_ok=True)
# Create a basic CSS file if it doesn't exist
with open(css_path, "w") as f:
f.write("""
/* Custom styles for ML Dataset & Code Generation Manager */
@import url('https://fonts.googleapis.com/css2?family=Inter:wght@400;500;600;700&family=Space+Grotesk:wght@500;700&display=swap');
h1, h2, h3, h4, h5, h6 {
font-family: 'Space Grotesk', sans-serif;
font-weight: 700;
color: #1A1C1F;
}
body {
font-family: 'Inter', sans-serif;
color: #1A1C1F;
background-color: #F8F9FA;
}
.stButton button {
background-color: #2563EB;
color: white;
border-radius: 4px;
border: none;
padding: 0.5rem 1rem;
font-weight: 600;
}
.stButton button:hover {
background-color: #1D4ED8;
}
/* Card styling */
.card {
background-color: white;
border-radius: 8px;
padding: 1.5rem;
box-shadow: 0 1px 3px rgba(0, 0, 0, 0.1);
margin-bottom: 1rem;
}
/* Accent colors */
.accent-primary {
color: #2563EB;
}
.accent-secondary {
color: #84919A;
}
.accent-success {
color: #10B981;
}
.accent-warning {
color: #F59E0B;
}
.accent-danger {
color: #EF4444;
}
""")
# Load custom CSS
with open(css_path, "r") as f:
st.markdown(f"<style>{f.read()}</style>", unsafe_allow_html=True)
def render_finetune_ui():
"""
Renders the fine-tuning UI for code generation models.
"""
try:
from components.fine_tuning.finetune_ui import render_finetune_ui as ft_ui
ft_ui()
except ImportError as e:
st.error(f"Could not load fine-tuning UI: {e}")
# Create default fine-tuning UI component if not exists
os.makedirs("components/fine_tuning", exist_ok=True)
if not os.path.exists("components/fine_tuning/__init__.py"):
with open("components/fine_tuning/__init__.py", "w") as f:
f.write('"""\nFine-tuning package for code generation models.\n"""\n')
if not os.path.exists("components/fine_tuning/finetune_ui.py"):
with open("components/fine_tuning/finetune_ui.py", "w") as f:
f.write('''"""
Streamlit UI for fine-tuning code generation models.
"""
import streamlit as st
import pandas as pd
import os
def render_dataset_preparation():
"""
Render the dataset preparation interface.
"""
st.subheader("Dataset Preparation")
st.write("Prepare your dataset for fine-tuning code generation models.")
# Dataset upload
uploaded_file = st.file_uploader("Upload your dataset", type=["csv", "json"])
if uploaded_file is not None:
try:
if uploaded_file.name.endswith('.csv'):
df = pd.read_csv(uploaded_file)
else:
df = pd.read_json(uploaded_file)
st.write("Dataset Preview:")
st.dataframe(df.head())
# Example of data columns mapping
st.subheader("Column Mapping")
input_col = st.selectbox("Select input column (e.g., code)", df.columns)
target_col = st.selectbox("Select target column (e.g., comment)", df.columns)
# Sample transformation
if st.button("Apply Transformation"):
if input_col and target_col:
# Example transformation: simple trim/clean
df[input_col] = df[input_col].astype(str).str.strip()
df[target_col] = df[target_col].astype(str).str.strip()
st.write("Transformed Dataset:")
st.dataframe(df.head())
# Option to save processed dataset
if st.button("Save Processed Dataset"):
processed_path = os.path.join("datasets", "processed_dataset.csv")
os.makedirs("datasets", exist_ok=True)
df.to_csv(processed_path, index=False)
st.success(f"Dataset saved to {processed_path}")
except Exception as e:
st.error(f"Error processing dataset: {e}")
def render_model_training():
"""
Render the model training interface.
"""
st.subheader("Model Training")
st.write("Configure and start training your model.")
# Model selection
model_options = [
"Salesforce/codet5-small",
"Salesforce/codet5-base",
"microsoft/codebert-base",
"microsoft/graphcodebert-base"
]
selected_model = st.selectbox("Select base model", model_options)
# Training parameters
col1, col2 = st.columns(2)
with col1:
batch_size = st.number_input("Batch size", min_value=1, max_value=64, value=8)
epochs = st.number_input("Number of epochs", min_value=1, max_value=100, value=3)
learning_rate = st.number_input("Learning rate", min_value=0.00001, max_value=0.1, value=0.0001, format="%.5f")
with col2:
max_input_length = st.number_input("Max input length", min_value=32, max_value=512, value=128)
max_target_length = st.number_input("Max target length", min_value=32, max_value=512, value=128)
task_type = st.selectbox("Task type", ["Code to Comment", "Comment to Code"])
# Training button (placeholder)
if st.button("Start Training"):
st.info("Training would start here. This is a placeholder.")
# In a real implementation, this would call the training function
# and display a progress bar or redirect to a training monitoring page
def render_model_testing():
"""
Render the model testing interface.
"""
st.subheader("Model Testing")
st.write("Test your fine-tuned model with custom inputs.")
# Model selection
st.selectbox("Select fine-tuned model", ["No models available yet"])
# Test input
if st.selectbox("Task type", ["Code to Comment", "Comment to Code"]) == "Code to Comment":
test_input = st.text_area("Enter code to generate a comment",
value="def fibonacci(n):\\n if n <= 1:\\n return n\\n else:\\n return fibonacci(n-1) + fibonacci(n-2)")
placeholder = "# This function implements the Fibonacci sequence recursively..."
else:
test_input = st.text_area("Enter comment to generate code",
value="# A function that calculates the factorial of a number recursively")
placeholder = "def factorial(n):\\n if n == 0:\\n return 1\\n else:\\n return n * factorial(n-1)"
# Generate button (placeholder)
if st.button("Generate"):
st.code(placeholder, language="python")
# In a real implementation, this would call the model inference function
def render_finetune_ui():
"""
Render the fine-tuning UI for code generation models.
"""
st.title("Fine-Tune Code Generation Models")
tabs = st.tabs(["Dataset Preparation", "Model Training", "Model Testing"])
with tabs[0]:
render_dataset_preparation()
with tabs[1]:
render_model_training()
with tabs[2]:
render_model_testing()
''')
# Try again after creating the files
try:
from components.fine_tuning.finetune_ui import render_finetune_ui as ft_ui
ft_ui()
except ImportError as e:
st.error(f"Still could not load fine-tuning UI after creating files: {e}")
st.info("Please restart the app to initialize the components.")
def render_code_quality_ui():
"""
Renders the code quality tools UI.
"""
try:
from components.code_quality import render_code_quality_tools
render_code_quality_tools()
except ImportError:
st.error("Code quality tools not found. Implementing basic version.")
st.title("Code Quality Tools")
st.write("This section will provide tools for code linting, formatting, and testing.")
# Tabs for different code quality tools
tabs = st.tabs(["Linting", "Formatting", "Type Checking", "Testing"])
with tabs[0]:
st.subheader("Code Linting")
st.write("Tools for checking code quality and style.")
st.code("# Coming soon: PyLint and Flake8 integration")
with tabs[1]:
st.subheader("Code Formatting")
st.write("Tools for formatting code according to style guides.")
st.code("# Coming soon: Black and isort integration")
with tabs[2]:
st.subheader("Type Checking")
st.write("Tools for checking type annotations.")
st.code("# Coming soon: MyPy integration")
with tabs[3]:
st.subheader("Testing")
st.write("Tools for running tests and checking code coverage.")
st.code("# Coming soon: PyTest integration")
def render_dataset_management_ui():
"""
Renders the dataset management UI.
"""
st.title("Dataset Management")
# Tabs for different dataset operations
tabs = st.tabs(["Upload", "Preview", "Statistics", "Visualization", "Validation", "Version Control"])
with tabs[0]:
try:
from components.dataset_uploader import render_dataset_uploader
render_dataset_uploader()
except ImportError:
st.subheader("Dataset Upload")
st.write("Upload your datasets in CSV or JSON format.")
uploaded_file = st.file_uploader("Choose a file", type=["csv", "json"])
if uploaded_file is not None:
try:
if uploaded_file.name.endswith('.csv'):
df = pd.read_csv(uploaded_file)
dataset_type = "csv"
else:
df = pd.read_json(uploaded_file)
dataset_type = "json"
st.session_state["dataset"] = df
st.session_state["dataset_type"] = dataset_type
st.success(f"Successfully loaded {dataset_type.upper()} file with {df.shape[0]} rows and {df.shape[1]} columns.")
st.dataframe(df.head())
except Exception as e:
st.error(f"Error: {e}")
with tabs[1]:
if "dataset" in st.session_state:
try:
from components.dataset_preview import render_dataset_preview
render_dataset_preview(st.session_state["dataset"], st.session_state["dataset_type"])
except ImportError:
st.subheader("Dataset Preview")
st.dataframe(st.session_state["dataset"].head(10))
else:
st.info("Please upload a dataset first.")
with tabs[2]:
if "dataset" in st.session_state:
try:
from components.dataset_statistics import render_dataset_statistics
render_dataset_statistics(st.session_state["dataset"], st.session_state["dataset_type"])
except ImportError:
st.subheader("Dataset Statistics")
st.write("Basic statistics:")
st.write(st.session_state["dataset"].describe())
# Missing values
missing_data = st.session_state["dataset"].isnull().sum()
st.write("Missing values per column:")
st.write(missing_data[missing_data > 0])
else:
st.info("Please upload a dataset first.")
with tabs[3]:
if "dataset" in st.session_state:
try:
from components.dataset_visualization import render_dataset_visualization
render_dataset_visualization(st.session_state["dataset"], st.session_state["dataset_type"])
except ImportError:
st.subheader("Dataset Visualization")
# Only show for numerical columns
numeric_cols = st.session_state["dataset"].select_dtypes(include=[np.number]).columns.tolist()
if len(numeric_cols) > 0:
col1, col2 = st.columns(2)
with col1:
x_axis = st.selectbox("X-axis", numeric_cols)
with col2:
y_axis = st.selectbox("Y-axis", numeric_cols, index=min(1, len(numeric_cols)-1))
fig = px.scatter(st.session_state["dataset"], x=x_axis, y=y_axis)
st.plotly_chart(fig, use_container_width=True)
else:
st.write("No numerical columns available for visualization.")
else:
st.info("Please upload a dataset first.")
with tabs[4]:
if "dataset" in st.session_state:
try:
from components.dataset_validation import render_dataset_validation
render_dataset_validation(st.session_state["dataset"], st.session_state["dataset_type"])
except ImportError:
st.subheader("Dataset Validation")
# Simple validation checks
st.write("Dataset Shape:", st.session_state["dataset"].shape)
st.write("Duplicate Rows:", st.session_state["dataset"].duplicated().sum())
# Missing values percentage
missing_percent = (st.session_state["dataset"].isnull().sum() / len(st.session_state["dataset"])) * 100
st.write("Missing Values Percentage:")
st.write(missing_percent[missing_percent > 0])
else:
st.info("Please upload a dataset first.")
with tabs[5]:
if "dataset" in st.session_state:
try:
from components.dataset_version_control import render_version_control_ui, render_save_version_ui, render_version_visualization
# If we have a dataset ID in session state, use it, otherwise prompt to save first
if "dataset_id" in st.session_state:
dataset_id = st.session_state["dataset_id"]
# Show dataset version control UI
render_version_control_ui(dataset_id, st.session_state.get("dataset"))
# Show save version UI
st.divider()
if st.session_state.get("dataset") is not None:
new_version = render_save_version_ui(dataset_id, st.session_state["dataset"])
if new_version:
st.success(f"Created new version: {new_version.version_id}")
# Show version visualization
st.divider()
render_version_visualization(dataset_id)
else:
# No dataset ID yet, so prompt to save the dataset first
st.info("To use version control, first save this dataset to the database.")
dataset_name = st.text_input("Dataset Name", value="My Dataset")
dataset_description = st.text_area("Dataset Description", value="Dataset uploaded for analysis")
if st.button("Save Dataset to Database"):
# Import database operations
from database.operations import DatasetOperations, DatasetVersionOperations
# Store dataset in database
dataset = DatasetOperations.store_dataframe_info(
df=st.session_state["dataset"],
name=dataset_name,
description=dataset_description,
source="local_upload"
)
# Store as initial version
initial_version = DatasetVersionOperations.create_version_from_dataframe(
dataset_id=dataset.id,
df=st.session_state["dataset"],
description="Initial version"
)
# Store dataset ID in session state
st.session_state["dataset_id"] = dataset.id
st.success(f"Dataset saved to database with ID: {dataset.id}")
st.success(f"Initial version created: {initial_version.version_id}")
# Rerun to show version control UI
st.experimental_rerun()
except ImportError as e:
st.subheader("Dataset Version Control")
st.error(f"Could not load version control components: {e}")
st.info("Please make sure all required components are installed.")
else:
st.info("Please upload a dataset first.")
def main():
"""
Main function to run the application.
"""
# Load custom CSS
load_css()
# Sidebar for navigation
st.sidebar.title("ML Dataset & Code Gen Manager")
# Navigation
page = st.sidebar.radio("Navigation", ["Home", "Dataset Management", "Fine-Tuning", "Code Quality Tools"])
# Display selected page
if page == "Home":
st.title("ML Dataset & Code Generation Manager")
st.write("Welcome to the ML Dataset & Code Generation Manager. This platform helps you manage ML datasets and fine-tune code generation models.")
# Main features in cards
col1, col2 = st.columns(2)
with col1:
st.markdown("""
<div class="card">
<h3>Dataset Management</h3>
<p>Upload, analyze, visualize, and validate your ML datasets.</p>
<ul>
<li>Support for CSV and JSON formats</li>
<li>Statistical analysis and visualization</li>
<li>Data validation and quality checks</li>
<li>Hugging Face Hub integration</li>
</ul>
</div>
""", unsafe_allow_html=True)
st.markdown("""
<div class="card">
<h3>Code Quality Tools</h3>
<p>Tools for ensuring high-quality code.</p>
<ul>
<li>Code linting with PyLint</li>
<li>Code formatting with Black and isort</li>
<li>Type checking with MyPy</li>
<li>Testing with PyTest</li>
</ul>
</div>
""", unsafe_allow_html=True)
with col2:
st.markdown("""
<div class="card">
<h3>Fine-Tuning</h3>
<p>Fine-tune code generation models on your custom datasets.</p>
<ul>
<li>Support for CodeT5, CodeBERT models</li>
<li>Code-to-comment and comment-to-code tasks</li>
<li>Custom dataset preparation</li>
<li>Model testing and evaluation</li>
</ul>
</div>
""", unsafe_allow_html=True)
st.markdown("""
<div class="card">
<h3>Hugging Face Integration</h3>
<p>Seamless integration with Hugging Face Hub.</p>
<ul>
<li>Search and load models and datasets</li>
<li>Deploy fine-tuned models to Hugging Face Spaces</li>
<li>Share and collaborate on models and datasets</li>
</ul>
</div>
""", unsafe_allow_html=True)
# Get started section
st.subheader("Get Started")
st.write("To get started, navigate to the Dataset Management page to upload your data, or explore the Fine-Tuning page to train code generation models.")
elif page == "Dataset Management":
render_dataset_management_ui()
elif page == "Fine-Tuning":
render_finetune_ui()
elif page == "Code Quality Tools":
render_code_quality_ui()
if __name__ == "__main__":
main() |