Spaces:
Sleeping
Sleeping
File size: 4,305 Bytes
a3c36fd db41bee 4207871 b7541ec db41bee 4207871 d710548 e57f927 31db34d bc45490 db41bee bc45490 f671264 bc45490 5c5efaf bc45490 b7541ec 5c5efaf 02bac7f 5c5efaf 02bac7f 5c5efaf bc45490 31db34d db41bee 31db34d db41bee e57f927 db41bee e57f927 bc45490 3cd939f bc45490 7e5c64b bc45490 33354da 2b58e9e eba77c5 2b58e9e eba77c5 bc45490 7e5c64b bc45490 33354da bc45490 7e5c64b 02bac7f bc45490 33354da bc45490 5c5efaf 02bac7f bc45490 7e5c64b 33354da 7e5c64b bc45490 4207871 bc45490 4207871 a3c36fd bc45490 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 |
import gradio as gr
import numpy as np
import os
import pandas as pd
import pysr
import tempfile
from typing import Optional
empty_df = pd.DataFrame(
{
"equation": [],
"loss": [],
"complexity": [],
}
)
def greet(
file_obj: Optional[tempfile._TemporaryFileWrapper],
col_to_fit: str,
niterations: int,
maxsize: int,
binary_operators: list,
unary_operators: list,
force_run: bool,
):
if col_to_fit == "":
return (
empty_df,
"Please enter a column to predict!",
)
if len(binary_operators) == 0 and len(unary_operators) == 0:
return (
empty_df,
"Please select at least one operator!",
)
if file_obj is None:
return (
empty_df,
"Please upload a CSV file!",
)
# Look at some statistics of the file:
df = pd.read_csv(file_obj.name)
if len(df) == 0:
return (
empty_df,
"The file is empty!",
)
if len(df.columns) == 1:
return (
empty_df,
"The file has only one column!",
)
if col_to_fit not in df.columns:
return (
empty_df,
f"The column to predict, {col_to_fit}, is not in the file!"
f"I found {df.columns}.",
)
if len(df) > 1000 and not force_run:
return (
empty_df,
"You have uploaded a file with more than 2000 rows. "
"This will take very long to run. "
"Please upload a subsample of the data, "
"or check the box 'Ignore Warnings'.",
)
binary_operators = str(binary_operators).replace("'", '"')
unary_operators = str(unary_operators).replace("'", '"')
df = pd.read_csv(file_obj)
y = np.array(df[col_to_fit])
X = df.drop([col_to_fit], axis=1)
model = pysr.PySRRegressor(
progress=False,
verbosity=0,
maxsize=maxsize,
niterations=niterations,
binary_operators=binary_operators,
unary_operators=unary_operators,
)
model.fit(X, y)
df = model.equations_[["equation", "loss", "complexity"]]
# Convert all columns to string type:
df = df.astype(str)
msg = (
"Success!\n"
f"You may run the model locally (faster) with "
f"the following parameters:"
+f"""
model = PySRRegressor(
niterations={niterations},
binary_operators={str(binary_operators)},
unary_operators={str(unary_operators)},
maxsize={maxsize},
)
model.fit(X, y)""")
df.to_csv("pysr_output.csv", index=False)
return df, msg
def main():
demo = gr.Interface(
fn=greet,
description="Symbolic Regression with PySR. Watch search progress by clicking 'See logs'!",
inputs=[
gr.inputs.File(label="Upload a CSV File"),
gr.inputs.Textbox(label="Column to Predict", placeholder="y"),
gr.inputs.Slider(
minimum=1,
maximum=1000,
default=40,
label="Number of Iterations",
step=1,
),
gr.inputs.Slider(
minimum=7,
maximum=35,
default=20,
label="Maximum Complexity",
step=1,
),
gr.inputs.CheckboxGroup(
choices=["+", "-", "*", "/", "^"],
label="Binary Operators",
default=["+", "-", "*", "/"],
),
gr.inputs.CheckboxGroup(
choices=[
"sin",
"cos",
"exp",
"log",
"square",
"cube",
"sqrt",
"abs",
"tan",
],
label="Unary Operators",
default=[],
),
gr.inputs.Checkbox(
default=False,
label="Ignore Warnings",
),
],
outputs=[
"dataframe",
gr.outputs.Textbox(label="Error Log"),
],
)
# Add file to the demo:
demo.launch()
if __name__ == "__main__":
main()
|