Duplicate from drift-ai/faq-website

Co-authored-by: claes <vincentclaes@users.noreply.huggingface.co>

.gitignore

@@ -0,0 +1,129 @@
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[cod]
+*$py.class
+
+# C extensions
+*.so
+
+# Distribution / packaging
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+pip-wheel-metadata/
+share/python-wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+MANIFEST
+
+# PyInstaller
+#  Usually these files are written by a python script from a template
+#  before PyInstaller builds the exe, so as to inject date/other infos into it.
+*.manifest
+*.spec
+
+# Installer logs
+pip-log.txt
+pip-delete-this-directory.txt
+
+# Unit test / coverage reports
+htmlcov/
+.tox/
+.nox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+*.cover
+*.py,cover
+.hypothesis/
+.pytest_cache/
+
+# Translations
+*.mo
+*.pot
+
+# Django stuff:
+*.log
+local_settings.py
+db.sqlite3
+db.sqlite3-journal
+
+# Flask stuff:
+instance/
+.webassets-cache
+
+# Scrapy stuff:
+.scrapy
+
+# Sphinx documentation
+docs/_build/
+
+# PyBuilder
+target/
+
+# Jupyter Notebook
+.ipynb_checkpoints
+
+# IPython
+profile_default/
+ipython_config.py
+
+# pyenv
+.python-version
+
+# pipenv
+#   According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control.
+#   However, in case of collaboration, if having platform-specific dependencies or dependencies
+#   having no cross-platform support, pipenv may install dependencies that don't work, or not
+#   install all needed dependencies.
+#Pipfile.lock
+
+# PEP 582; used by e.g. github.com/David-OConnor/pyflow
+__pypackages__/
+
+# Celery stuff
+celerybeat-schedule
+celerybeat.pid
+
+# SageMath parsed files
+*.sage.py
+
+# Environments
+.env
+.venv
+env/
+venv/
+ENV/
+env.bak/
+venv.bak/
+
+# Spyder project settings
+.spyderproject
+.spyproject
+
+# Rope project settings
+.ropeproject
+
+# mkdocs documentation
+/site
+
+# mypy
+.mypy_cache/
+.dmypy.json
+dmypy.json
+
+# Pyre type checker
+.pyre/

README.md

@@ -0,0 +1,15 @@
+---
+title: FAQ a Website
+emoji: 🦙
+colorFrom: white
+colorTo: gray
+sdk: gradio
+sdk_version: 3.21.0
+app_file: app.py
+pinned: false
+license: apache-2.0
+duplicated_from: drift-ai/faq-website
+---
+
+# Faq A website
+repo for the code to QA content form a website

app.py

@@ -0,0 +1,149 @@
+import torch
+from peft import PeftModel
+import transformers
+import gradio as gr
+from scrape_website import process_webpage
+assert (
+    "LlamaTokenizer" in transformers._import_structure["models.llama"]
+), "LLaMA is now in HuggingFace's main branch.\nPlease reinstall it: pip uninstall transformers && pip install git+https://github.com/huggingface/transformers.git"
+from transformers import LlamaTokenizer, LlamaForCausalLM, GenerationConfig
+
+tokenizer = LlamaTokenizer.from_pretrained("decapoda-research/llama-7b-hf")
+
+BASE_MODEL = "decapoda-research/llama-7b-hf"
+LORA_WEIGHTS = "tloen/alpaca-lora-7b"
+
+if torch.cuda.is_available():
+    device = "cuda"
+else:
+    device = "cpu"
+
+try:
+    if torch.backends.mps.is_available():
+        device = "mps"
+except:
+    pass
+
+if device == "cuda":
+    model = LlamaForCausalLM.from_pretrained(
+        BASE_MODEL,
+        load_in_8bit=False,
+        torch_dtype=torch.float16,
+        device_map="auto",
+    )
+    model = PeftModel.from_pretrained(
+        model, LORA_WEIGHTS, torch_dtype=torch.float16, force_download=True
+    )
+elif device == "mps":
+    model = LlamaForCausalLM.from_pretrained(
+        BASE_MODEL,
+        device_map={"": device},
+        torch_dtype=torch.float16,
+    )
+    model = PeftModel.from_pretrained(
+        model,
+        LORA_WEIGHTS,
+        device_map={"": device},
+        torch_dtype=torch.float16,
+    )
+else:
+    model = LlamaForCausalLM.from_pretrained(
+        BASE_MODEL, device_map={"": device}, low_cpu_mem_usage=True
+    )
+    model = PeftModel.from_pretrained(
+        model,
+        LORA_WEIGHTS,
+        device_map={"": device},
+    )
+
+
+def generate_prompt(instruction, input=None):
+    if input:
+        return f"""Below is an url that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request.
+### Instruction:
+{instruction}
+### Input:
+{input}
+### Response:"""
+    else:
+        return f"""Below is an url that describes a task. Write a response that appropriately completes the request.
+### Instruction:
+{instruction}
+### Response:"""
+
+if device != "cpu":
+    model.half()
+model.eval()
+if torch.__version__ >= "2":
+    model = torch.compile(model)
+
+
+def evaluate(
+    instruction,
+    url,
+    temperature=0.1,
+    top_p=0.75,
+    top_k=40,
+    num_beams=4,
+    max_new_tokens=128,
+    **kwargs,
+):
+    content = process_webpage(url=url)
+    # avoid GPU memory overflow
+    with torch.no_grad():
+        torch.cuda.empty_cache()
+        prompt = generate_prompt(instruction, content)
+        inputs = tokenizer(prompt, return_tensors="pt")
+        input_ids = inputs["input_ids"].to(device)
+        generation_config = GenerationConfig(
+            temperature=temperature,
+            top_p=top_p,
+            top_k=top_k,
+            num_beams=num_beams,
+            **kwargs,
+        )
+        generation_output = model.generate(
+            input_ids=input_ids,
+            generation_config=generation_config,
+            return_dict_in_generate=True,
+            output_scores=True,
+            max_new_tokens=max_new_tokens,
+        )
+        s = generation_output.sequences[0]
+        output = tokenizer.decode(s)
+    # avoid GPU memory overflow
+    torch.cuda.empty_cache()
+    return output.split("### Response:")[1].strip()
+
+
+g = gr.Interface(
+    fn=evaluate,
+    inputs=[
+        gr.components.Textbox(
+            lines=2, label="FAQ", placeholder="Ask me anything about this website?"
+        ),
+        gr.components.Textbox(lines=1, label="Website URL", placeholder="https://www.meet-drift.ai/"),
+        # gr.components.Slider(minimum=0, maximum=1, value=0.1, label="Temperature"),
+        # gr.components.Slider(minimum=0, maximum=1, value=0.75, label="Top p"),
+        # gr.components.Slider(minimum=0, maximum=100, step=1, value=40, label="Top k"),
+        # gr.components.Slider(minimum=1, maximum=4, step=1, value=4, label="Beams"),
+        # gr.components.Slider(
+        #     minimum=1, maximum=512, step=1, value=128, label="Max tokens"
+        # ),
+    ],
+    outputs=[
+        gr.inputs.Textbox(
+            lines=5,
+            label="Output",
+        )
+    ],
+    title="FAQ A Website",
+    examples=[
+        ["Can you list the capabilities this company has in bullet points?", "https://www.meet-drift.ai/"],
+        ["What's the name of the founder?", "https://www.meet-drift.ai/about"],
+        ["in 1 word what's the service the company is providing?", "https://www.meet-drift.ai/"],
+    ]
+    # description="Alpaca-LoRA is a 7B-parameter LLaMA model finetuned to follow instructions. It is trained on the [Stanford Alpaca](https://github.com/tatsu-lab/stanford_alpaca) dataset and makes use of the Huggingface LLaMA implementation. For more information, please visit [the project's website](https://github.com/tloen/alpaca-lora).",
+)
+g.queue(concurrency_count=1)
+g.launch()

requirements.txt

@@ -0,0 +1,10 @@
+beautifulsoup4
+requests
+datasets
+loralib
+sentencepiece
+git+https://github.com/huggingface/transformers.git
+accelerate
+bitsandbytes
+git+https://github.com/huggingface/peft.git
+gradio

scrape_website.py

@@ -0,0 +1,66 @@
+import requests
+from bs4 import BeautifulSoup
+
+TOKEN_CUT_OFF = 2500
+
+def process_webpage(url:str):
+    # A set to keep track of visited pages
+    visited_pages = set()
+
+    text_list = []
+
+    # A function to recursively get all child pages
+    def get_child_pages(url):
+        # Make a GET request to the page and get the HTML content
+        response = requests.get(url)
+        html_content = response.content
+
+        # Parse the HTML content using BeautifulSoup
+        soup = BeautifulSoup(html_content, "html.parser")
+
+        # Get all the text content from the relevant HTML tags
+        text_content = ""
+        for tag in ["p", "h1", "h2", "h3", "h4", "h5", "h6", "li"]:
+            for element in soup.find_all(tag):
+                text_content += element.get_text() + " "
+
+        # Add the page to the set of visited pages
+        text_content = f"page {url} contains: " + text_content
+        visited_pages.add(url)
+
+        # Find all the child links and recursively get their text content
+        for link in soup.find_all("a"):
+            href = link.get("href")
+            if href and href not in visited_pages and url in href:
+                get_child_pages(href)
+
+        text_list.append(text_content)
+
+    # Get the text content of the landing page
+    # get_child_pages(url)
+
+    # Make a GET request to the page and get the HTML content
+    response = requests.get(url)
+    html_content = response.content
+
+    # Parse the HTML content using BeautifulSoup
+    soup = BeautifulSoup(html_content, "html.parser")
+
+    # Get all the text content from the relevant HTML tags
+    text_content = ""
+    for tag in ["p", "h1", "h2", "h3", "h4", "h5", "h6", "li"]:
+        for element in soup.find_all(tag):
+            text_content += element.get_text() + " "
+
+    # # make main page as first item
+    # text_list.reverse()
+    # text_list_cut_off = text_list[:TOKEN_CUT_OFF]
+    # page_content = "\n".join(text_list_cut_off)
+    # # Print the text content of the landing page and all child pages
+    # print(page_content)
+    # return page_content
+    print(text_content)
+    return text_content
+
+if __name__ == '__main__':
+    process_webpage(url="https://www.meet-drift.ai/")