Spaces:
Running
Running
File size: 4,387 Bytes
f39233d |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 |
import gradio as gr
import pandas as pd
from huggingface_hub import HfApi
# Initialize Hugging Face API
api = HfApi()
# Constants
GGUF_TAG = "gguf"
CHUNK_SIZE = 1000
# Clickable links function
def clickable(x, which_one):
if which_one == "models":
if x not in ["Not Found", "Unknown"]:
return f'<a target="_blank" href="https://huggingface.co/{x}" style="color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;">{x}</a>'
else:
return "Not Found"
else:
if x != "Not Found":
return f'<a target="_blank" href="https://huggingface.co/{which_one}/{x}" style="color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;">{x}</a>'
return "Not Found"
# Fetch models and return a DataFrame with clickable links
def fetch_models():
models = api.list_models(filter=GGUF_TAG, full=True)
data = []
for model in models:
model_id = model.id if model.id else "Not Found"
author = model.author if model.author else "Unknown"
data.append({
"Model ID": model_id,
"Author Name": author,
"Downloads (30d)": model.downloads or 0,
"Likes": model.likes or 0,
"Created At": model.created_at.isoformat() if model.created_at else "N/A",
"Last Modified": model.last_modified.isoformat() if model.last_modified else "N/A",
})
df = pd.DataFrame(data)
# Apply clickable links
df["Model ID"] = df["Model ID"].apply(lambda x: clickable(x, "models"))
df["Author Name"] = df["Author Name"].apply(lambda x: clickable(x, "models"))
return df
# Prepare authors DataFrame
def prepare_authors_df(models_df):
# Extract the actual author name from the link (if needed)
authors_df = models_df.copy()
authors_df["Clean Author Name"] = authors_df["Author Name"].str.extract(r'href="https://huggingface\.co/(.*?)"')
grouped = authors_df.groupby("Clean Author Name").agg(
Models_Count=("Model ID", "count"),
Total_Downloads=("Downloads (30d)", "sum")
).reset_index()
grouped.rename(columns={"Clean Author Name": "Author Name"}, inplace=True)
return grouped.sort_values(by="Models_Count", ascending=False)
all_models_df = fetch_models().sort_values(by="Downloads (30d)", ascending=False)
authors_df = prepare_authors_df(all_models_df)
def update_model_table(start_idx):
# Instead of messing with model_table.value, just return more rows from all_models_df
new_end = start_idx + CHUNK_SIZE
# Slice the global DataFrame that already has HTML formatting
combined_df = all_models_df.iloc[:new_end].copy()
return combined_df, new_end
with gr.Blocks() as demo:
gr.Markdown("""
# GGUF Tracker
Welcome to **GGUF Tracker**, a live-updating leaderboard for all things GGUF on Hugging Face. It’s simple: the stats refresh every hour, giving you the latest numbers on downloads, likes, and top contributors. Whether you're here to find models or just check out who's killing it, this is the place.
By the way, I’m Richard Erkhov, and you can check out more of what I’m working on at my [**github**](https://github.com/RichardErkhov), [**huggingface**](https://huggingface.co/RichardErkhov) or [**erkhov.com**](https://erkhov.com). Go take a look—I think you’ll like what you find.
""")
# Existing content
gr.Markdown("# GGUF Models and Authors Leaderboard")
with gr.Tabs():
with gr.TabItem("Models"):
# Initially load the first CHUNK_SIZE rows
model_table = gr.DataFrame(
value=all_models_df.iloc[:CHUNK_SIZE],
interactive=True,
label="GGUF Models",
wrap=True,
datatype=["markdown", "markdown", "number", "number", "str", "str"]
)
load_more_button = gr.Button("Load More Models")
start_idx = gr.State(value=CHUNK_SIZE)
load_more_button.click(fn=update_model_table, inputs=start_idx, outputs=[model_table, start_idx])
with gr.TabItem("Authors"):
gr.DataFrame(
value=authors_df,
interactive=False,
label="Authors",
wrap=True,
datatype=["str", "number", "number"]
)
demo.launch()
|