Spaces:
No application file
No application file
File size: 2,515 Bytes
8772b50 f646fc1 8772b50 f646fc1 e7d518b f646fc1 8772b50 f646fc1 8772b50 f646fc1 8772b50 f646fc1 8772b50 f646fc1 8772b50 7e055ad 8772b50 7e055ad 8772b50 f646fc1 e7d518b 8772b50 7e055ad 8772b50 f646fc1 e7d518b 8772b50 |
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 |
from typing import List
import gradio as gr
import pandas as pd
from app.backend.constant import LEADERBOARD_MAP
from app.backend.data_engine import DataEngine, COLUMNS_TYPES
class SubtabsComponent:
def __init__(self, data_engine: DataEngine):
self.data_engine = data_engine
def show(self, navigations: str = None, model_provides: List = None, evaluation_metrics: str = None,
embd_dtypes: str = None, embd_dims: List = None, similarity: str = None, max_tokens: int = None):
df_result = self.data_engine.jsons_to_df()
navigations = "Text" if navigations is None else navigations.split(" ", maxsplit=1)[0]
evaluation_metrics = "NDCG@10" if evaluation_metrics is None else evaluation_metrics
embd_dtype = "all" if embd_dtypes is None else embd_dtypes
embd_dims = [] if embd_dims is None else embd_dims
similarity = "all" if similarity is None else similarity
max_tokens = 0 if max_tokens is None else max_tokens
df_result = self.data_engine.filter_df(df_result, embd_dtype, embd_dims, similarity, max_tokens)
sort_col = evaluation_metrics.replace("@", '_at_').lower()
df_result = df_result.sort_values(by=sort_col, ascending=False)
items = []
for group_name,leaderboards in LEADERBOARD_MAP.items():
with gr.Column(visible=group_name.upper() == navigations.upper()) as column:
with gr.Tabs():
with gr.TabItem("overall"):
df_leaderboard = df_result[df_result["group_name"] == group_name]
gr_df = gr.Dataframe(
df_leaderboard,
datatype=COLUMNS_TYPES,
# interactive=True,
type="pandas"
)
items.append(gr_df)
for leaderboard in leaderboards:
with gr.TabItem(leaderboard):
df = df_leaderboard[df_leaderboard["leaderboard"] == leaderboard]
gr_df = gr.Dataframe(
df,
datatype=COLUMNS_TYPES,
# interactive=True,
height=500,
type="pandas"
)
items.append(gr_df)
items.append(column)
return items
|