File size: 1,823 Bytes
7a7f67a |
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 |
import plotly.graph_objects as go
import plotly.express as px
import numpy as np
def plot_elo_mle(df):
fig = px.scatter(df, x="model", y="rating", error_y="error_y",
error_y_minus="error_y_minus",
# title="Bootstrap of Elo MLE Estimates (BigCodeBench-Complete)"
)
fig.update_layout(xaxis_title="Model",
yaxis_title="Rating",
autosize=True,
# width=1300,
# height=900,
)
return fig
def plot_solve_rate(df, task, rows=30, cols=38):
keys = df["task_id"]
values = df["solve_rate"]
values = np.array(values)
n = len(values)
values = np.pad(values, (0, rows * cols - n), 'constant', constant_values=np.nan).reshape((rows, cols))
keys = np.pad(keys, (0, rows * cols - n), 'constant', constant_values='').reshape((rows, cols))
hover_text = np.empty_like(values, dtype=object)
for i in range(rows):
for j in range(cols):
if not np.isnan(values[i, j]):
hover_text[i, j] = f"{keys[i, j]}<br>Solve Rate: {values[i, j]:.2f}"
else:
hover_text[i, j] = "NaN"
upper_solve_rate = round(np.count_nonzero(values)/n*100, 2)
fig = go.Figure(data=go.Heatmap(
z=values,
text=hover_text,
hoverinfo='text',
colorscale='teal',
zmin=0,
zmax=100
))
fig.update_layout(
title=f'BigCodeBench-{task}<br><i>Lowest Upper Limit: {upper_solve_rate}%</i>',
xaxis_nticks=cols,
yaxis_nticks=rows,
xaxis=dict(showticklabels=False),
yaxis=dict(showticklabels=False),
autosize=True,
# width=760,
# height=600,
)
return fig |