Last active
February 23, 2024 23:20
-
-
Save bigsnarfdude/c19dbff531712ec1c5ee84e68d7785e4 to your computer and use it in GitHub Desktop.
process_transcripts.py
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
import os | |
import sys | |
def clean_duplicates(file_name): | |
duplicates = [] | |
cleaned = [] | |
with open(file_name, 'r') as f: | |
sentences = f.readlines() | |
for s in sentences: | |
if s in cleaned: | |
if s not in duplicates: | |
duplicates.append(s) | |
else: | |
cleaned.append(s) | |
return ' '.join(cleaned) | |
def process_directory(directory): | |
for file_name in os.listdir(directory): | |
if file_name.endswith('.txt'): | |
file_path = os.path.join(directory, file_name) | |
cleaned_text = clean_duplicates(file_path) | |
new_file_name = file_path.replace('.txt', '_cleaned.txt') | |
with open(new_file_name, 'w') as f: | |
f.write(cleaned_text) | |
if __name__ == "__main__": | |
if len(sys.argv) < 2: | |
print("Usage: python script_name.py <directory_path>") | |
sys.exit(1) | |
directory_path = sys.argv[1] | |
process_directory(directory_path) |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
import sys | |
from llama_cpp import Llama | |
from prompt import PromptText | |
from utils import read_file_to_string | |
from utils import get_filename | |
from utils import slugify | |
from utils import split_file | |
import time | |
input_file_path = sys.argv[1] if len(sys.argv) > 1 else sys.exit("No input file found") | |
output_file_path = get_filename(input_file_path) | |
output_file_path = slugify(output_file_path) | |
output_file_path = output_file_path+'_summary.txt' # built to go to temp but doesnt delete | |
mixtral = "/home/vincent/development/llama.cpp/models/mixtral-8x7b-instruct-v0.1.Q5_K_M.gguf" | |
mistral = "/home/vincent/development/llama.cpp/models/mistral-7b-instruct-v0.2.Q5_K_M.gguf" | |
prompt = PromptText().summary_text | |
file_chunks = split_file(input_file_path) | |
holder = [] | |
for item in file_chunks: | |
content = read_file_to_string(item) | |
content_word_max = 4096 | |
gpu_layers = 14 | |
generated_response = 200 | |
summary_submission = prompt+content+'[/INST]' | |
llm = Llama(model_path=mixtral, n_ctx=content_word_max, n_gpu_layers=gpu_layers) | |
output = llm(summary_submission, max_tokens=generated_response, stop=["</s>"]) | |
output_text = output['choices'][0]['text'] | |
holder.append(output_text) | |
llm = None | |
with open(output_file_path, 'w') as output_file: | |
output_file.write(''.join(holder)) |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
import os | |
import re | |
def read_file_to_string(file_path): | |
try: | |
with open(file_path, 'r') as file: | |
content = file.read() | |
return content | |
except FileNotFoundError: | |
return "File not found." | |
except Exception as e: | |
return f"An error occurred: {e}" | |
def slugify(value): | |
""" | |
Normalizes a string: converts to lowercase, removes non-alpha characters, | |
and converts dashes and spaces to underscores. | |
""" | |
value = re.sub(r"[^\w\s-]", "", value).strip().lower() | |
value = re.sub(r"[-\s]+", "_", value) | |
return re.sub(r"[^\x00-\x7f]", "", value) | |
def get_filename(file_path): | |
""" | |
Returns the filename without the filetype extension. | |
""" | |
#base = os.path.basename(file_path) | |
#filename_without_extension, _ = os.path.splitext(base) | |
return os.path.splitext(os.path.basename(file_path))[0] | |
def split_file(file_path, chunk_size=4096): | |
holder = [] | |
with open(file_path, 'r', encoding='utf-8') as file: | |
content = file.read() | |
chunks = [content[i:i+chunk_size] for i in range(0, len(content), chunk_size)] | |
for i, chunk in enumerate(chunks): | |
output_file_name = f'/tmp/{file_path}_chunk_{i+1}.txt' | |
with open(output_file_name, 'w', encoding='utf-8') as chunk_file: | |
chunk_file.write(chunk) | |
holder.append(output_file_name) | |
return holder | |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment