Custom document loaders in LangChain - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to import the base class for creating a custom document loader.

LangChain

from langchain.document_loaders import [1]

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

ABaseDocumentLoader

BCustomLoaderBase

CDocumentLoaderBase

DLoaderBase

Attempts:

3 left

💡 Hint

Common Mistakes

Using a class name that does not exist in the module.

Confusing the order of words in the class name.

The base class to create custom document loaders in LangChain is BaseDocumentLoader.

2fill in blank

medium

Complete the method name to override for loading documents in a custom loader class.

LangChain

class MyLoader(BaseDocumentLoader):
    def [1](self):
        # Your code to load documents
        pass

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

Aload_documents

Bfetch_documents

Cload

Dget_docs

Attempts:

3 left

💡 Hint

Common Mistakes

Using a method name that does not match the base class signature.

Using a generic name like 'load' instead of 'load_documents'.

The method to override in a custom document loader is load_documents, which returns the loaded documents.

3fill in blank

hard

Fix the error in the custom loader by completing the return statement to return a list of Document objects.

LangChain

from langchain.schema import Document

class MyLoader(BaseDocumentLoader):
    def load_documents(self):
        text = 'Sample text content'
        return [[1](page_content=text)]

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

ADoc

BDocumentLoader

CDocument

DDocObject

Attempts:

3 left

💡 Hint

Common Mistakes

Returning a list of strings instead of Document objects.

Using a wrong class name for the document wrapper.

The load_documents method must return a list of Document objects from langchain.schema.

4fill in blank

hard

Fill both blanks to create a custom loader that reads text from a file path and returns a Document.

LangChain

class FileLoader(BaseDocumentLoader):
    def __init__(self, file_path):
        self.file_path = file_path

    def load_documents(self):
        with open(self.file_path, [1]) as f:
            text = f.read()
        return [Document(page_content=[2])]

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

Ar

Btext

Cw

Dfile_path

Attempts:

3 left

💡 Hint

Common Mistakes

Using write mode 'w' instead of read mode 'r'.

Returning the file path string instead of the file content.

Open the file in read mode 'r' and return a Document with the read text.

5fill in blank

hard

Fill all three blanks to create a custom loader that filters lines containing a keyword and returns Documents for each matching line.

LangChain

class KeywordLoader(BaseDocumentLoader):
    def __init__(self, file_path, keyword):
        self.file_path = file_path
        self.keyword = keyword

    def load_documents(self):
        docs = []
        with open(self.file_path, [1]) as f:
            for line in f:
                if [2] in line:
                    docs.append(Document(page_content=[3]))
        return docs

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

Ar

Bself.keyword

Cline

Dw

Attempts:

3 left

💡 Hint

Common Mistakes

Opening the file in write mode instead of read mode.

Checking the wrong variable for the keyword.

Appending the keyword instead of the line content.

Open the file in read mode 'r', check if self.keyword is in each line, and create a Document with the matching line.