Spaces:
Runtime error
Runtime error
File size: 4,207 Bytes
ac6c40f 57616af ac6c40f 88dffd6 ac6c40f 57616af ac6c40f 88dffd6 57616af 28dd726 57616af 396d1e7 57616af 396d1e7 57616af 28dd726 57616af 28dd726 57616af 28dd726 9999db9 28dd726 57616af ac6c40f 396d1e7 9994065 396d1e7 |
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 |
import streamlit as st
from .streamlit_utils import (
make_multiselect,
make_selectbox,
make_text_area,
make_text_input,
make_radio,
)
N_FIELDS = 7
def results_page():
st.session_state.card_dict["results"] = st.session_state.card_dict.get(
"results", {}
)
with st.expander("Previous Results", expanded=False):
key_pref = ["results", "results"]
st.session_state.card_dict["results"]["results"] = st.session_state.card_dict[
"results"
].get("results", {})
make_text_area(
label="What aspect of model ability can be measured with this dataset?",
key_list=key_pref + ["model-abilities"],
help="What kind of abilities should a model exhibit that performs well on the task of this dataset (e.g., reasoning capability, morphological inflection)?.",
)
make_multiselect(
label="What metrics are typically used for this task?",
key_list=key_pref + ["metrics"],
options=[
"BERT-Score",
"BLEU",
"BLEURT",
"ChrF",
"Entailment",
"FeQA",
"METEOR", "MoverScore",
"QAGS",
"ROUGE",
"WER",
"Other: Other Metrics"
],
help="Select all metrics that are typically used when evaluating models for this task.",
)
if "Other: Other Metrics" in st.session_state.card_dict["results"]["results"].get("metrics", []):
make_text_area(
label="Definitions of other metrics",
key_list=key_pref + ["other-metrics-definitions"],
help="If the evaluation strategies in the previous questions go beyond the list of metrics above, add descriptions and/or definitions for each metric.",
)
else:
st.session_state.card_dict["results"]["results"]["other-metrics-definitions"] = "N/A"
make_text_area(
label="List and describe the purpose of the metrics and evaluation methodology (including human evaluation) that the dataset creators used when introducing this task.",
key_list=key_pref + ["original-evaluation"],
help="When the generation task was not evaluated when this dataset was introduced, write N/A.",
)
make_radio(
label="Are previous results available?",
options=["no", "yes"],
key_list=key_pref + ["has-previous-results"],
help="Have papers evaluated models on this task? If no, write N/A for the following three questions.",
)
if st.session_state.card_dict["results"]["results"]["has-previous-results"] == "yes":
make_text_area(
label="What evaluation approaches have others used?",
key_list=key_pref + ["current-evaluation"],
help="If the current evaluation strategy diverts from the original, describe how models are being evaluated.",
)
make_text_area(
label="What are the most relevant previous results for this task/dataset?",
key_list=key_pref + ["previous-results"],
help="List and describe the source and performance metrics for models on this dataset.",
)
else:
st.session_state.card_dict["results"]["results"]["current-evaluation"] = "N/A"
st.session_state.card_dict["results"]["results"]["previous-results"] = "N/A"
def results_summary():
total_filled = sum(
[len(dct) for dct in st.session_state.card_dict.get("results", {}).values()]
)
with st.expander(
f"Previous Results Completion - {total_filled} of {N_FIELDS}", expanded=False
):
completion_markdown = ""
completion_markdown += (
f"- **Overall completion:**\n - {total_filled} of {N_FIELDS} fields\n"
)
completion_markdown += f"- **Sub-section - Previous Results:**\n - {len(st.session_state.card_dict.get('results', {}).get('results', {}))} of {N_FIELDS} fields\n"
st.markdown(completion_markdown)
|