wehub-resource-sync
97e91a83f3
Ruff / Ruff (push) Has been cancelled
Test / Core Tests (push) Has been cancelled
Test / Offline Coverage Tests (Python 3.10) (push) Has been cancelled
Test / Offline Coverage Tests (Python 3.11) (push) Has been cancelled
Test / Offline Coverage Tests (Python 3.12) (push) Has been cancelled
Test / Offline Coverage Tests (Python 3.13) (push) Has been cancelled
Test / Offline Coverage Tests (Python 3.9) (push) Has been cancelled
Test / Full Coverage (Python 3.11) (push) Has been cancelled
Test / Core Provider Tests (OpenAI) (push) Has been cancelled
Test / Core Provider Tests (Anthropic) (push) Has been cancelled
Test / Core Provider Tests (Google) (push) Has been cancelled
Test / Core Provider Tests (Other) (push) Has been cancelled
Test / Anthropic Tests (push) Has been cancelled
Test / Gemini Tests (push) Has been cancelled
Test / Google GenAI Tests (push) Has been cancelled
Test / Vertex AI Tests (push) Has been cancelled
Test / OpenAI Tests (push) Has been cancelled
Test / Writer Tests (push) Has been cancelled
Test / Auto Client Tests (push) Has been cancelled
ty / type-check (push) Has been cancelled
63 lines
1.8 KiB
Python
63 lines
1.8 KiB
Python
import streamlit as st
|
|
from stats_dict import stats_dict
|
|
|
|
# Sample data
|
|
query_data = {i: line.strip() for i, line in enumerate(open("test.jsonl"))}
|
|
|
|
# Initialize selected keys
|
|
selected_keys = {}
|
|
|
|
|
|
# Function to get lines
|
|
def get_lines(stats_key, keys):
|
|
indices = []
|
|
for key in keys:
|
|
indices.extend(stats_dict[stats_key]["_reverse_lookup"][key])
|
|
return "\n".join([query_data[i] for i in indices])
|
|
|
|
|
|
# Function to render dropdown and button
|
|
def render_dropdown_and_button(stats_key):
|
|
st.subheader(f"Stats for `{stats_key}`")
|
|
st.json(stats_dict[stats_key]["counter"])
|
|
st.json(
|
|
{k: v for k, v in stats_dict[stats_key].items() if isinstance(v, (int, float))}
|
|
)
|
|
st.subheader("Histogram")
|
|
st.bar_chart(stats_dict[stats_key]["counter"], use_container_width=True)
|
|
|
|
options = list(stats_dict[stats_key]["counter"].keys())
|
|
selected_keys[stats_key] = st.multiselect(
|
|
f"View samples with {stats_key}",
|
|
options,
|
|
default=selected_keys.get(stats_key, []),
|
|
)
|
|
st.code(get_lines(stats_key, selected_keys[stats_key]))
|
|
|
|
|
|
# Sidebar for navigation
|
|
st.sidebar.title("Navigation")
|
|
page = st.sidebar.selectbox(
|
|
"Select a page:",
|
|
["Validation Stats", "Individual Path Views"],
|
|
)
|
|
|
|
# Main Streamlit App
|
|
st.title("Structured Output Evaluation")
|
|
|
|
# Validation Stats
|
|
if page == "Validation Stats":
|
|
st.header("Validation Stats")
|
|
for key in [k for k in stats_dict.keys() if k.startswith("_")]:
|
|
render_dropdown_and_button(key)
|
|
|
|
# Individual Path Views
|
|
elif page == "Individual Path Views":
|
|
st.header("Individual Path Views")
|
|
path = st.selectbox(
|
|
"Choose a path:",
|
|
[key for key in stats_dict.keys() if not key.startswith("_")],
|
|
)
|
|
if "counter" in stats_dict[path]:
|
|
render_dropdown_and_button(path)
|