Update .gitignore and Dockerfile, add

.gitignore

@@ -1,3 +1,4 @@
 .venv/
 __pycache__/
-.env
+.env
+cache

Dockerfile

@@ -4,8 +4,11 @@ FROM python:3.11.1
 # Set the working directory in the container
 WORKDIR /app
 
-# Install poetry
-# RUN pip3 install poetry==1.7.1
+# Run the container as root
+# USER root
+
+# Change permissions of the working directory
+RUN chmod 777 /app  # Gives all users read, write, and exec permissions in the app directory.
 
 # Copy the current directory contents into the container at /usr/src/app
 COPY . .

fine_tune_data/validation_data.csv

@@ -0,0 +1,2 @@
+text
+"### Human: What is the aerospace mechanisms symposia?### Assistant: An annual meeting of space mechanism experts. {'source': 'DM', 'page': 0}"

train_llm.py

@@ -12,10 +12,10 @@ from tqdm.notebook import tqdm
 
 # Load dataset
 dataset_name = 'ai-aerospace/ams_data_train_generic_v0.1_100'
-dataset=load_dataset(dataset_name,cache_dir='.')
+dataset=load_dataset(dataset_name,cache_dir='./cache')
 
 # Write dataset files into data directory
-data_directory = '../fine_tune_data/'
+data_directory = './fine_tune_data/'
 
 # Create the data directory if it doesn't exist
 os.makedirs(data_directory, exist_ok=True)