VoucherVision / vouchervision /LLM_MistralAI.py
phyloforfun's picture
Major update. Support for 15 LLMs, World Flora Online taxonomy validation, geolocation, 2 OCR methods, significant UI changes, stability improvements, consistent JSON parsing
9d06861
raw
history blame
7.3 kB
import os, time, random, torch, json
from langchain_mistralai.chat_models import ChatMistralAI
from langchain.output_parsers import RetryWithErrorOutputParser
from langchain.prompts import PromptTemplate
from langchain_core.output_parsers import JsonOutputParser
from vouchervision.utils_LLM import SystemLoadMonitor, count_tokens, save_individual_prompt
from vouchervision.utils_LLM_JSON_validation import validate_and_align_JSON_keys_with_template
from vouchervision.utils_taxonomy_WFO import validate_taxonomy_WFO
from vouchervision.utils_geolocate_HERE import validate_coordinates_here
from vouchervision.tool_wikipedia import WikipediaLinks
class MistralHandler:
RETRY_DELAY = 2 # Wait 10 seconds before retrying
MAX_RETRIES = 5 # Maximum number of retries
STARTING_TEMP = 0.1
TOKENIZER_NAME = None
VENDOR = 'mistral'
RANDOM_SEED = 2023
def __init__(self, logger, model_name, JSON_dict_structure):
self.logger = logger
self.monitor = SystemLoadMonitor(logger)
self.has_GPU = torch.cuda.is_available()
self.model_name = model_name
self.JSON_dict_structure = JSON_dict_structure
self.starting_temp = float(self.STARTING_TEMP)
self.temp_increment = float(0.2)
self.adjust_temp = self.starting_temp
# Set up a parser
self.parser = JsonOutputParser()
# Define the prompt template
self.prompt = PromptTemplate(
template="Answer the user query.\n{format_instructions}\n{query}\n",
input_variables=["query"],
partial_variables={"format_instructions": self.parser.get_format_instructions()},
)
self._set_config()
def _set_config(self):
self.config = {'max_tokens': 1024,
'temperature': self.starting_temp,
'random_seed': self.RANDOM_SEED,
'safe_mode': False,
'top_p': 1,
}
self._build_model_chain_parser()
def _adjust_config(self):
new_temp = self.adjust_temp + self.temp_increment
self.config['random_seed'] = random.randint(1, 1000)
self.json_report.set_text(text_main=f'Incrementing temperature from {self.adjust_temp} to {new_temp} and random_seed to {self.config.get("random_seed")}')
self.logger.info(f'Incrementing temperature from {self.adjust_temp} to {new_temp} and random_seed to {self.config.get("random_seed")}')
self.adjust_temp += self.temp_increment
self.config['temperature'] = self.adjust_temp
def _reset_config(self):
self.json_report.set_text(text_main=f'Resetting temperature from {self.adjust_temp} to {self.starting_temp} and random_seed to {self.RANDOM_SEED}')
self.logger.info(f'Incrementing temperature from {self.adjust_temp} to {self.starting_temp} and random_seed to {self.RANDOM_SEED}')
self.adjust_temp = self.starting_temp
self.config['temperature'] = self.starting_temp
self.config['random_seed'] = self.RANDOM_SEED
def _build_model_chain_parser(self):
# Initialize MistralAI
self.llm_model = ChatMistralAI(mistral_api_key=os.environ.get("MISTRAL_API_KEY"),
model=self.model_name,
max_tokens=self.config.get('max_tokens'),
safe_mode=self.config.get('safe_mode'),
top_p=self.config.get('top_p'))
# Set up the retry parser with the runnable
self.retry_parser = RetryWithErrorOutputParser.from_llm(parser=self.parser, llm=self.llm_model, max_retries=self.MAX_RETRIES)
self.chain = self.prompt | self.llm_model
def call_llm_api_MistralAI(self, prompt_template, json_report, paths):
_____, ____, _, __, ___, json_file_path_wiki, txt_file_path_ind_prompt = paths
self.json_report = json_report
self.json_report.set_text(text_main=f'Sending request to {self.model_name}')
self.monitor.start_monitoring_usage()
nt_in = 0
nt_out = 0
ind = 0
while ind < self.MAX_RETRIES:
ind += 1
try:
model_kwargs = {"temperature": self.adjust_temp, "random_seed": self.config.get("random_seed")}
# Invoke the chain to generate prompt text
response = self.chain.invoke({"query": prompt_template, "model_kwargs": model_kwargs})
# Use retry_parser to parse the response with retry logic
output = self.retry_parser.parse_with_prompt(response.content, prompt_value=prompt_template)
if output is None:
self.logger.error(f'[Attempt {ind}] Failed to extract JSON from:\n{response}')
self._adjust_config()
else:
nt_in = count_tokens(prompt_template, self.VENDOR, self.TOKENIZER_NAME)
nt_out = count_tokens(response.content, self.VENDOR, self.TOKENIZER_NAME)
output = validate_and_align_JSON_keys_with_template(output, self.JSON_dict_structure)
if output is None:
self.logger.error(f'[Attempt {ind}] Failed to extract JSON from:\n{response}')
self._adjust_config()
else:
self.monitor.stop_inference_timer() # Starts tool timer too
json_report.set_text(text_main=f'Working on WFO, Geolocation, Links')
output, WFO_record = validate_taxonomy_WFO(output, replace_if_success_wfo=False) ###################################### make this configurable
output, GEO_record = validate_coordinates_here(output, replace_if_success_geo=False) ###################################### make this configurable
Wiki = WikipediaLinks(json_file_path_wiki)
Wiki.gather_wikipedia_results(output)
save_individual_prompt(Wiki.sanitize(prompt_template), txt_file_path_ind_prompt)
self.logger.info(f"Formatted JSON:\n{json.dumps(output,indent=4)}")
usage_report = self.monitor.stop_monitoring_report_usage()
if self.adjust_temp != self.starting_temp:
self._reset_config()
json_report.set_text(text_main=f'LLM call successful')
return output, nt_in, nt_out, WFO_record, GEO_record, usage_report
except Exception as e:
self.logger.error(f'JSON Parsing Error (LangChain): {e}')
self._adjust_config()
time.sleep(self.RETRY_DELAY)
self.logger.info(f"Failed to extract valid JSON after [{ind}] attempts")
self.json_report.set_text(text_main=f'Failed to extract valid JSON after [{ind}] attempts')
usage_report = self.monitor.stop_monitoring_report_usage()
self._reset_config()
json_report.set_text(text_main=f'LLM call failed')
return None, nt_in, nt_out, None, None, usage_report