File size: 13,436 Bytes
42da440 | 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 | import json
import os
import random
import re
import traceback
import sys
import asyncio
from augmentoolkit.generation_functions.generation_step_class import GenerationStep
from augmentoolkit.generation_functions.pipeline_step_class import (
PipelineStep,
filter_out_nonpresent_keys,
)
from augmentoolkit.utils.make_id import make_id
from augmentoolkit.utils.write_output_to_file import write_output_to_file
from tqdm import asyncio as tqdmasyncio
class RandomVariationStep(
PipelineStep
): # pipeline steps store the settings and the prompt, and prevent us from having to repeat the "read previous output" code among other things
def __init__(
self,
sampling_params=None,
output_file=None,
intermediate_output_path=None,
save_path=None,
use_stop=None,
completion_mode=None,
validation_function=lambda x, y: {"result": True, "message": "default message"},
max_retries=3,
result_key=None,
prompt_path=None,
regex=re.compile(r".*", re.DOTALL),
variation_generator_count=None,
output_processor=lambda x: x,
log_full_outputs=False,
method_overrides={},
details_key=None,
):
self.variation_generator_count = variation_generator_count
super().__init__(
sampling_params=sampling_params,
output_file=output_file, # Add output_file parameter for consolidated storage
intermediate_output_path=intermediate_output_path,
save_path=save_path,
use_stop=use_stop,
completion_mode=completion_mode,
validation_function=validation_function,
max_retries=max_retries,
result_key=result_key,
prompt_path=prompt_path,
output_processor=output_processor,
log_full_outputs=log_full_outputs,
method_overrides=method_overrides,
details_key=details_key,
)
async def read_previous_output(self, key, input_dict):
entry = input_dict.get(str(key), {})
# if not entry:
# print(f" Key '{str(key)}' not found in input_dict.")
# else:
# print(f" Found entry. Keys: {list(entry.keys())}")
current_variations = entry.get(self.result_key, [])
should_skip = False
if len(current_variations) >= self.variation_generator_count:
print(
f" Found {len(current_variations)} variations, which meets or exceeds target {self.variation_generator_count}. Will skip generation for key {key}."
)
should_skip = True
# else:
return should_skip
async def generate_data(
self,
processed_data,
engine_wrapper,
prompt_path,
prompt_folder,
default_prompt_folder,
completion_mode,
use_stop,
**kwargs,
):
# randomly select a yaml file from the prompt folder
# prompt_path is the name of the subfolder; prompt_folder is, per usual, the path to the overall PROMPTS folder.
search_dir = os.path.join(prompt_folder, prompt_path)
if not os.path.exists(search_dir):
search_dir = os.path.join(default_prompt_folder, prompt_path)
if not completion_mode:
prompt_files = [f for f in os.listdir(search_dir) if f.endswith(".yaml")]
else:
prompt_files = [
f for f in os.listdir(search_dir) if f.endswith(".yaml")
] # NOTE: This looks like a potential bug, it lists .yaml for completion mode too.
if not prompt_files:
error_msg = f" ERROR: No suitable prompt files found in {search_dir} for completion_mode={completion_mode}"
print(error_msg)
raise FileNotFoundError(error_msg)
random_prompt_file = random.choice(prompt_files)
prompt_filepath = os.path.join(search_dir, random_prompt_file)
try:
generator = GenerationStep(
prompt_path=prompt_filepath,
default_prompt_folder=default_prompt_folder,
sampling_params=self.sampling_params,
completion_mode=completion_mode,
engine_wrapper=engine_wrapper,
output_processor=self.output_processor,
retries=1,
logging_level=self.logging_level,
use_stop=use_stop,
prompt_folder=prompt_folder,
regex=self.regex,
)
# Note: We don't log the actual return values here as they could be large
result, full_output, full_response, full_input = await generator.generate(
**processed_data, **self.static_arguments, **kwargs
)
return result, full_output, full_response, full_input
except Exception as e:
print(f" ERROR during GenerationStep initialization or execution: {e}")
traceback.print_exc()
# Re-raise the exception to be handled by the caller (run method)
raise
def save(
self,
result=None,
key=None,
input_dict=None,
input_data=None,
full_response=None,
full_input=None,
completion_mode=False,
include_details=False,
):
# Get existing entry or create a new one. Avoid copying input_data unless necessary.
entry = input_dict.get(str(key))
if entry is None:
# If the key truly doesn't exist, we MUST base it on input_data
# This case should ideally not happen often if execute_pipeline populates correctly
entry = input_data.copy() if input_data else {}
# else:
# Ensure the result key exists and is a list
if self.result_key not in entry or not isinstance(
entry.get(self.result_key), list
): # Use .get for safety
entry[self.result_key] = []
try:
entry[self.result_key].append(result)
except Exception as e:
print(f" ERROR appending result: {e}")
traceback.print_exc()
raise
if include_details:
# Ensure the details key exists and is a list
if self.details_key not in entry or not isinstance(
entry.get(self.details_key), list
): # Use .get for safety
entry[self.details_key] = []
detail_entry = {
"full_response": full_response,
"full_input": full_input,
"completion_mode": completion_mode,
}
try:
entry[self.details_key].append(detail_entry)
except Exception as e:
print(f" ERROR appending detail entry: {e}")
traceback.print_exc()
raise
# else:
try:
input_dict[str(key)] = entry
except Exception as e:
print(f" ERROR assigning entry to input_dict: {e}")
traceback.print_exc()
raise
return entry
async def run(
self,
key=None,
input_data=None,
engine_wrapper=None,
input_dict=None,
default_prompt_folder=None,
prompt_folder=None,
output_dir=None,
completion_mode=False,
use_stop=True,
include_details=False,
**kwargs,
):
should_skip = await self.read_previous_output(key, input_dict)
if should_skip:
return
# else:
processed_data, additional_kwargs = self.process_input_data(input_data)
# existing_variations = [] # This variable seems unused after the FileLock block was removed
# Determine current count and number needed based *only* on in-memory input_dict
current_variations_in_memory = input_dict.get(str(key), {}).get(
self.result_key, []
)
num_existing_in_memory = len(current_variations_in_memory)
num_variations_needed = self.variation_generator_count - num_existing_in_memory
if num_variations_needed <= 0:
# This case should technically be caught by read_previous_output, but added for robustness
return
# Generate remaining variations
for i in range(num_variations_needed):
error_message = ""
complete = False
max_retries = self.max_retries
while not complete and max_retries > 0:
try:
result, full_output, full_response, full_input = (
await self.generate_data(
processed_data,
engine_wrapper,
self.prompt_path,
prompt_folder,
default_prompt_folder,
completion_mode,
use_stop,
error_message=error_message,
**kwargs,
**additional_kwargs,
)
)
validation_passed = self.validation_function(result, input_data)
if validation_passed["result"]:
complete = True
else:
error_message = validation_passed["message"]
except Exception as e:
print(f" ERROR during generate_data or validation: {e}")
error_message = str(e)
traceback.print_exc()
max_retries -= 1
if not complete:
print(
f" Attempt failed or validation failed. Retries left: {max_retries}"
)
# End of while loop
if not complete:
print(
f" ERROR: Failed to generate a valid variation after {self.max_retries} attempts for key {key}, iteration {i+1}."
)
# We might want to remove the key from input_dict here or handle failure differently
return
# else:
self.save(
result=result,
key=key,
input_dict=input_dict,
input_data=input_dict.get(
str(key)
), # Pass the current state from input_dict
full_response=full_response,
full_input=full_input,
completion_mode=completion_mode,
include_details=include_details,
)
# End of for loop
async def execute_pipeline(
self,
input_dict={},
engine_wrapper=None,
rtwl=None,
default_prompt_folder=None,
prompt_folder=None,
output_dir=None,
completion_mode=None,
use_stop=None,
include_details=False,
**kwargs,
):
self.load_dataset(
input_dict=input_dict, output_dir=output_dir
) # if it is at the start and end of every pipeline execution, we get the same behavior as before, past thing will load and save just fine, without constant reads/writes.
try:
data_generations_tasks = [
self.run(
key=key,
input_data=value,
engine_wrapper=engine_wrapper,
input_dict=input_dict,
default_prompt_folder=default_prompt_folder,
prompt_folder=prompt_folder,
output_dir=output_dir,
completion_mode=completion_mode,
use_stop=use_stop,
include_details=include_details,
**kwargs,
)
for key, value in input_dict.items()
]
coroutines = [rtwl(task) for task in data_generations_tasks]
TASK_TIMEOUT_SECONDS = 600 # 10 minutes timeout
processed_count = 0
total_tasks = len(coroutines)
for future in tqdmasyncio.tqdm.as_completed(coroutines):
processed_count += 1
try:
await asyncio.wait_for(future, timeout=TASK_TIMEOUT_SECONDS)
except asyncio.TimeoutError:
print(
f"\nWARNING: Task {processed_count}/{total_tasks} for step '{self.output_file}' timed out after {TASK_TIMEOUT_SECONDS} seconds.",
file=sys.stderr,
)
# Task is cancelled by wait_for, loop continues
except Exception as e_inner: # Catch exceptions from the future itself
print(
f"\nError processing task {processed_count}/{total_tasks}: {e_inner}"
)
traceback.print_exc() # Optionally log traceback
filter_out_nonpresent_keys(input_dict, key_to_check=self.result_key)
except Exception as e:
print(f"Exception occurred during task execution: {e}")
# traceback.print_exc()
raise e
finally:
# Robust save with interrupt handling
self.save_dataset(input_dict=input_dict, output_dir=output_dir)
|