Spaces:
Running
Running
File size: 3,982 Bytes
da67f65 5f24bec 0b3b4fe da67f65 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 |
import datetime
from huggingface_hub import Repository
import os
import pandas as pd
import streamlit as st
import altair as alt
import numpy as np
import plotly.graph_objects as go
today = datetime.date.today()
year, week, _ = today.isocalendar()
DATASET_REPO_URL = (
"https://huggingface.co/datasets/huggingface/transformers-stats-space-data"
)
DATA_FILENAME = f"data_{week}_{year}.csv"
DATA_FILE = os.path.join("data", DATA_FILENAME)
MODELS_TO_TRACK = ["wav2vec2", "whisper"]
repo = Repository(local_dir="data", clone_from=DATASET_REPO_URL)
repo.git_pull()
valid_weeks = []
download_results = []
model_download_results = {model_name: [] for model_name in MODELS_TO_TRACK}
# loop over past data, finding where we have data saved (valid weeks) and tracking monthly downloads for each week
for i in range(1, week + 1)[::-1]:
data_filename = f"data_{i}_{year}.csv"
data_file = os.path.join("data", data_filename)
if os.path.exists(data_file):
valid_weeks.append(i)
dataframe = pd.read_csv(data_file)
df_audio = dataframe[dataframe["modality"] == "audio"]
audio_int_downloads = {model: int(x.replace(",", "")) for model, x in
zip(df_audio["model_names"], df_audio["num_downloads"].values)}
download_results.append(sum(audio_int_downloads.values()))
for model_name in MODELS_TO_TRACK:
model_download_results[model_name].append(audio_int_downloads.get(model_name))
last_year = year - 1
last_week = 52
data_filename = f"data_{last_week}_{last_year}.csv"
data_file = os.path.join("data", data_filename)
if os.path.exists(data_file):
valid_weeks.append(0)
dataframe = pd.read_csv(data_file)
df_audio = dataframe[dataframe["modality"] == "audio"]
audio_int_downloads = {model: int(x.replace(",", "")) for model, x in
zip(df_audio["model_names"], df_audio["num_downloads"].values)}
download_results.append(sum(audio_int_downloads.values()))
for model_name in MODELS_TO_TRACK:
model_download_results[model_name].append(audio_int_downloads.get(model_name))
fig = go.Figure()
fig.update_layout(
title="Monthly downloads",
xaxis_title="Week",
yaxis_title="Downloads",)
fig.add_trace(
go.Scatter(x=valid_weeks, y=download_results, mode='lines+markers', name="Total")
)
for model_name in MODELS_TO_TRACK:
fig.add_trace(
go.Scatter(x=valid_weeks, y=model_download_results[model_name], mode='lines+markers', name=model_name)
)
st.title("Audio Stats")
st.plotly_chart(fig)
week = st.selectbox(
"Week",
valid_weeks,
index=0,
help="Filter the download results by week"
)
DATA_FILENAME = f"data_{week}_{year}.csv"
DATA_FILE = os.path.join("data", DATA_FILENAME)
with open(DATA_FILE, "r") as f:
dataframe = pd.read_csv(DATA_FILE)
st.header(f"Stats for year {year} and week {week}")
# print audio
df_audio = dataframe[dataframe["modality"] == "audio"]
audio_int_downloads = np.array(
[int(x.replace(",", "")) for x in df_audio["num_downloads"].values]
)
source = pd.DataFrame(
{
"Number of total downloads": audio_int_downloads,
"Model architecture name": df_audio["model_names"].values,
}
)
bar_chart = (
alt.Chart(source)
.mark_bar()
.encode(
y="Number of total downloads",
x=alt.X("Model architecture name", sort=None),
)
)
st.subheader(f"Top audio downloads last 30 days")
st.altair_chart(bar_chart, use_container_width=True)
st.subheader("Audio stats last 30 days")
dataframe = dataframe[dataframe["modality"] == "audio"].drop("modality", axis=1)
dataframe.loc["Total"] = dataframe.sum(numeric_only=True)
total_audio_downloads = sum(audio_int_downloads)
# nice formatting
dataframe.at["Total", "num_downloads"] = "{:,}".format(total_audio_downloads)
dataframe.at["Total", "model_names"] = ""
dataframe.at["Total", "download_per_model"] = ""
st.table(dataframe) |