Spaces:
Running
Running
File size: 3,898 Bytes
441367e 1248b75 441367e |
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 |
import sys
import re
from importlib.metadata import version
import polars as pl
import gradio as gr
# Config
concurrency_limit = 5
title = "See ASR Outputs"
# https://www.tablesgenerator.com/markdown_tables
authors_table = """
## Authors
Follow them on social networks and **contact** if you need any help or have any questions:
| <img src="https://avatars.githubusercontent.com/u/7875085?v=4" width="100"> **Yehor Smoliakov** |
|-------------------------------------------------------------------------------------------------|
| https://t.me/smlkw in Telegram |
| https://x.com/yehor_smoliakov at X |
| https://github.com/egorsmkv at GitHub |
| https://huggingface.co/Yehor at Hugging Face |
| or use [email protected] |
""".strip()
examples = [
["evaluation_results.jsonl", False],
["evaluation_results_batch.jsonl", True],
]
description_head = f"""
# {title}
## Overview
See generated JSONL files made by ASR models as a dataframe.
""".strip()
description_foot = f"""
{authors_table}
""".strip()
metrics_value = """
Metrics will appear here.
""".strip()
tech_env = f"""
#### Environment
- Python: {sys.version}
""".strip()
tech_libraries = f"""
#### Libraries
- gradio: {version("gradio")}
- polars: {version("polars")}
""".strip()
def inference(file_name, _batch_mode):
if not file_name:
raise gr.Error("Please paste your JSON file.")
df = pl.read_ndjson(file_name)
required_columns = [
"filename",
"inference_start",
"inference_end",
"inference_total",
"duration",
"reference",
"prediction",
]
required_columns_batch = [
"inference_start",
"inference_end",
"inference_total",
"filenames",
"durations",
"references",
"predictions",
]
if _batch_mode:
if not all(col in df.columns for col in required_columns_batch):
raise gr.Error(
f"Please provide a JSONL file with the following columns: {required_columns_batch}"
)
else:
if not all(col in df.columns for col in required_columns):
raise gr.Error(
f"Please provide a JSONL file with the following columns: {required_columns}"
)
# exclude inference_start, inference_end
if _batch_mode:
df = df.drop(["inference_start", "inference_end", "filenames"])
else:
df = df.drop(["inference_start", "inference_end", "filename"])
# round "inference_total" field to 2 decimal places
df = df.with_columns(pl.col("inference_total").round(2))
return df
demo = gr.Blocks(
title=title,
analytics_enabled=False,
theme=gr.themes.Base(),
)
with demo:
gr.Markdown(description_head)
gr.Markdown("## Usage")
with gr.Row():
df = gr.DataFrame(
label="Dataframe",
)
with gr.Row():
with gr.Column():
jsonl_file = gr.File(label="A JSONL file")
batch_mode = gr.Checkbox(
label="Use batch mode",
)
gr.Button("Show").click(
inference,
concurrency_limit=concurrency_limit,
inputs=[jsonl_file, batch_mode],
outputs=df,
)
with gr.Row():
gr.Examples(
label="Choose an example",
inputs=[jsonl_file, batch_mode],
examples=examples,
)
gr.Markdown(description_foot)
gr.Markdown("### Gradio app uses:")
gr.Markdown(tech_env)
gr.Markdown(tech_libraries)
if __name__ == "__main__":
demo.queue()
demo.launch()
|