IlyasMoutawwakil's picture
update
5345cba
raw
history blame
5.07 kB
import gradio as gr
import pandas as pd
import plotly.express as px
QUANT_DATA = [
# open llm
"Model πŸ€—",
"DType πŸ“₯",
"Backend 🏭",
"Params (B)",
"Architecture πŸ›οΈ",
"Open LLM Score (%)",
# deployment settings
"DType πŸ“₯",
"Backend 🏭",
"Quantization πŸ—œοΈ",
"Quantization πŸ—œοΈ Custom Kernel",
# primary measurements
"Prefill (s)",
"Prefill (s) Custom Kernel",
"Decode (tokens/s)",
"Decode (tokens/s) Custom Kernel",
# speedups
"Prefill Speedup (%)",
"Decode Speedup (%)",
]
def get_quant_df(llm_perf_df):
copy_df = llm_perf_df.copy()
# seperate vanilla GPTQ experiments from Custom Kernel experiments
vanilla_df = copy_df[
(copy_df["Backend 🏭"] == "pytorch")
& (copy_df["DType πŸ“₯"] == "float16")
& (copy_df["Quantization πŸ—œοΈ"] == "Unquantized")
]
exllamav1_df = copy_df[(copy_df["Quantization πŸ—œοΈ"] == "GPTQ.4bit+ExllamaV1")]
exllamav2_df = copy_df[(copy_df["Quantization πŸ—œοΈ"] == "GPTQ.4bit+ExllamaV2")]
gemm_df = copy_df[(copy_df["Quantization πŸ—œοΈ"] == "AWQ.4bit+GEMM")]
gemv_df = copy_df[(copy_df["Quantization πŸ—œοΈ"] == "AWQ.4bit+GEMV")]
# merge the three dataframes
exllamav1_df = pd.merge(
vanilla_df,
exllamav1_df,
on=["Model πŸ€—"],
suffixes=["", " Custom Kernel"],
)
exllamav2_df = pd.merge(
vanilla_df,
exllamav2_df,
on=["Model πŸ€—"],
suffixes=["", " Custom Kernel"],
)
gemm_df = pd.merge(
vanilla_df,
gemm_df,
on=["Model πŸ€—"],
suffixes=["", " Custom Kernel"],
)
gemv_df = pd.merge(
vanilla_df,
gemv_df,
on=["Model πŸ€—"],
suffixes=["", " Custom Kernel"],
)
# concat the two dataframes row-wise
quant_df = pd.concat([exllamav1_df, exllamav2_df, gemm_df, gemv_df])
# compute speedups
quant_df["Prefill Speedup (%)"] = (
(quant_df["Prefill (s)"] / quant_df["Prefill (s) Custom Kernel"]) * 100
).round(2) - 100
quant_df["Decode Speedup (%)"] = (
(quant_df["Decode (tokens/s) Custom Kernel"] / quant_df["Decode (tokens/s)"])
* 100
).round(2) - 100
# filter speedups > 1000%
quant_df = quant_df[quant_df["Prefill Speedup (%)"] < 1000]
quant_df = quant_df[quant_df["Decode Speedup (%)"] < 1000]
return quant_df
def get_quant_decode_fig(llm_perf_df):
quant_df = get_quant_df(llm_perf_df)
# plot
decode_fig = px.box(
quant_df,
x="Architecture πŸ›οΈ",
y="Decode Speedup (%)",
color_discrete_sequence=px.colors.qualitative.Light24,
custom_data=QUANT_DATA,
color="Quantization πŸ—œοΈ Custom Kernel",
points="all",
)
# add hover data
decode_fig.update_traces(
hovertemplate="<br>".join(
[
f"<b>{column}:</b> %{{customdata[{i}]}}"
for i, column in enumerate(QUANT_DATA)
]
)
)
# add layout
decode_fig.update_layout(
title={
"text": "Decode Speedup per Architecture",
"y": 0.95,
"x": 0.5,
"xanchor": "center",
"yanchor": "top",
},
xaxis_title="LLM Architecture",
yaxis_title="Decode Speedup (%)",
legend_title="Quantization Scheme",
width=1200,
height=600,
)
return decode_fig
def get_quant_prefill_fig(llm_perf_df):
quant_df = get_quant_df(llm_perf_df)
# plot
prefill_fig = px.box(
quant_df,
x="Architecture πŸ›οΈ",
y="Prefill Speedup (%)",
color_discrete_sequence=px.colors.qualitative.Light24,
custom_data=QUANT_DATA,
color="Quantization πŸ—œοΈ Custom Kernel",
points="all",
)
# add hover data
prefill_fig.update_traces(
hovertemplate="<br>".join(
[
f"<b>{column}:</b> %{{customdata[{i}]}}"
for i, column in enumerate(QUANT_DATA)
]
)
)
# add layout
prefill_fig.update_layout(
title={
"text": "Prefill Speedup per Architecture",
"y": 0.95,
"x": 0.5,
"xanchor": "center",
"yanchor": "top",
},
xaxis_title="LLM Architecture",
yaxis_title="Prefill Speedup (%)",
legend_title="Quantization Scheme",
width=1200,
height=600,
)
return prefill_fig
def create_quant_plots(llm_perf_df):
# descriptive text
gr.HTML("πŸ‘† Hover over the points πŸ‘† for additional information.", elem_id="text")
# get figures
prefill_fig = get_quant_prefill_fig(llm_perf_df)
decode_fig = get_quant_decode_fig(llm_perf_df)
# create plots
prefill_plot = gr.components.Plot(
value=prefill_fig, elem_id="plot", show_label=False
)
decode_plot = gr.components.Plot(value=decode_fig, elem_id="plot", show_label=False)
return prefill_plot, decode_plot