ruslanmv

Tutorial: Exploring a Python Package or Framework

When working with a new Python framework or library (let's call it a "package"), you often need to understand its structure: what modules it contains, what classes and functions are available, and how they are organized. This tutorial shows you several ways to do this using Python's built-in tools.

We'll use a hypothetical framework named beeai_framework as our main example, based on your input.

Prerequisites

1. Python Installed: You need a working Python interpreter.
2. Package Installed: The package you want to explore must be installed in your Python environment (e.g., using pip install package_name).

ruslanmv

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer, TrainingArguments, Trainer
from datasets import load_dataset

# Load teacher (DeepSeek-R1) and student (Granite-8B)
teacher_name = "deepseek-ai/deepseek-llm-7b-base"  # Replace with actual DeepSeek-R1 HF path
student_name = "ibm-granite/granite-3.1-8b-instruct"

# Load models and tokenizers
teacher = AutoModelForCausalLM.from_pretrained(teacher_name, torch_dtype=torch.bfloat16)

ruslanmv

import gradio as gr
import os
import spaces
from transformers import GemmaTokenizer, AutoModelForCausalLM
from transformers import AutoModelForCausalLM, AutoTokenizer, TextIteratorStreamer
from threading import Thread

# Set an environment variable
HF_TOKEN = os.environ.get("HF_TOKEN", None)

ruslanmv

# Import necessary libraries
import os
from ibm_watson_machine_learning.foundation_models import Model
from ibm_watson_machine_learning.metanames import GenTextParamsMetaNames as GenParams
from ibm_watson_machine_learning.foundation_models.utils.enums import ModelTypes, DecodingMethods

def get_model(model_type, api_key, project_id, max_tokens, min_tokens, decoding, temperature, top_k, top_p):
    generate_params = {
        GenParams.MAX_NEW_TOKENS: max_tokens,
        GenParams.MIN_NEW_TOKENS: min_tokens,

ruslanmv

#!/bin/bash
# Ollama Model Export Script
# Usage: bash ollama-export.sh vicuna:7b


set -e

echo "Ollama Model Export Script"

echo ""

ruslanmv

pip install "notebook<7"

###  Saving the Notebook to markdown
import IPython
#from IPython.lib import kernel
from ipykernel.kernelbase import Kernel
%%javascript
var kernel = IPython.notebook.kernel;
var thename = window.document.getElementById("notebook_name").innerHTML;
var command = "theNotebook = " + "'"+thename+"'";

ruslanmv

import os
import torch
import torch.multiprocessing as mp
# Distributed training setup
def init_distributed(rank, world_size):
    os.environ["MASTER_ADDR"] = "localhost"
    os.environ["MASTER_PORT"] = "12345"
    if rank == 0:
        print("Initializing distributed process group...")
    torch.distributed.init_process_group(backend='nccl', world_size=world_size, rank=rank)

ruslanmv

from datasets import load_dataset
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig, AutoTokenizer
from peft import LoraConfig
from trl import SFTTrainer
from transformers import TrainingArguments
from accelerate import PartialState

# Load the dataset
dataset_name = "ruslanmv/ai-medical-dataset"

ruslanmv

from IPython.display import FileLink
!tar -czvf lm-evaluation-harness.tar.gz lm-evaluation-harness

def create_download_link(filename):
    return FileLink(filename)

download_link = create_download_link("lm-evaluation-harness.tar.gz")
display(download_link)

ruslanmv

from datasets import load_dataset
from IPython.display import clear_output
import pandas as pd
import re
from dotenv import load_dotenv
import os
from ibm_watson_machine_learning.foundation_models.utils.enums import ModelTypes
from ibm_watson_machine_learning.metanames import GenTextParamsMetaNames as GenParams
from ibm_watson_machine_learning.foundation_models.utils.enums import DecodingMethods
from langchain.llms import WatsonxLLM