Connecting to open-source models in LangChain - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to import the OpenAI model from LangChain.

LangChain

from langchain.llms import [1]

Drag options to blanks, or click blank then click option'

AGPT4All

BOpenAI

CHuggingFaceHub

DCohere

Attempts:

3 left

💡 Hint

Common Mistakes

Importing GPT4All instead of OpenAI

Using incorrect module path

The OpenAI class is imported from langchain.llms to connect to OpenAI models.

2fill in blank

medium

Complete the code to create a GPT4All model instance with the model path.

LangChain

from langchain.llms import GPT4All
model = GPT4All(model=[1])

Drag options to blanks, or click blank then click option'

A'gpt4all-model.bin'

Bgpt4all-model.bin

C'openai'

D'gpt-3.5-turbo'

Attempts:

3 left

💡 Hint

Common Mistakes

Not using quotes around the model path

Using OpenAI model name instead of GPT4All model file

The model path must be a string with quotes, so use 'gpt4all-model.bin'.

3fill in blank

hard

Fix the error in the code to load a HuggingFaceHub model with the repo_id.

LangChain

from langchain.llms import HuggingFaceHub
model = HuggingFaceHub(repo_id=[1])

Drag options to blanks, or click blank then click option'

Agpt2

B"gpt2"

C'gpt2'

Dgpt-3

Attempts:

3 left

💡 Hint

Common Mistakes

Passing repo_id without quotes causing syntax error

Using incorrect model name like gpt-3

The repo_id must be a string, so it needs quotes like 'gpt2'.

4fill in blank

hard

Fill both blanks to create a GPT4All instance with streaming enabled and a callback.

LangChain

from langchain.callbacks.streaming_stdout import StreamingStdOutCallbackHandler
callbacks = [[1]()]
model = GPT4All(model='gpt4all-model.bin', streaming=[2], callbacks=callbacks)

Drag options to blanks, or click blank then click option'

AStreamingStdOutCallbackHandler

BTrue

CFalse

DStreamingCallback

Attempts:

3 left

💡 Hint

Common Mistakes

Using wrong callback class name

Setting streaming to False or missing it

Use StreamingStdOutCallbackHandler for callbacks and set streaming=True to enable streaming output.

5fill in blank

hard

Fill all three blanks to create a HuggingFaceHub model with a specific repo_id, task, and model_kwargs.

LangChain

model = HuggingFaceHub(repo_id=[1], task=[2], model_kwargs=[3])

Drag options to blanks, or click blank then click option'

A'google/flan-t5-small'

B'text2text-generation'

C{'temperature': 0.7}

D'gpt2'

Attempts:

3 left

💡 Hint

Common Mistakes

Using wrong repo_id or task strings

Passing model_kwargs as a string instead of dictionary

Use the correct repo_id string, specify the task as 'text2text-generation', and pass model_kwargs as a dictionary with temperature.