Robo2000 commited on
Commit
fa64f12
1 Parent(s): e5f42ca

Upload app.py

Browse files
Files changed (1) hide show
  1. app.py +99 -0
app.py ADDED
@@ -0,0 +1,99 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ from typing import List, Dict
2
+ import httpx
3
+ import gradio as gr
4
+ import pandas as pd
5
+
6
+ async def get_splits(dataset_name: str) -> Dict[str, List[Dict]]:
7
+ URL = f"https://datasets-server.huggingface.co/splits?dataset={dataset_name}"
8
+ async with httpx.AsyncClient() as session:
9
+ response = await session.get(URL)
10
+ return response.json()
11
+
12
+ async def get_valid_datasets() -> Dict[str, List[str]]:
13
+ URL = f"https://datasets-server.huggingface.co/valid"
14
+ async with httpx.AsyncClient() as session:
15
+ response = await session.get(URL)
16
+ datasets = response.json()["valid"]
17
+ return gr.Dropdown.update(choices=datasets, value="awacke1/ChatbotMemory.csv")
18
+ # The one to watch: https://huggingface.co/rungalileo
19
+ # rungalileo/medical_transcription_40
20
+
21
+ async def get_first_rows(dataset: str, config: str, split: str) -> Dict[str, Dict[str, List[Dict]]]:
22
+ URL = f"https://datasets-server.huggingface.co/first-rows?dataset={dataset}&config={config}&split={split}"
23
+ async with httpx.AsyncClient() as session:
24
+ response = await session.get(URL)
25
+ print(URL)
26
+ gr.Markdown(URL)
27
+ return response.json()
28
+
29
+ def get_df_from_rows(api_output):
30
+ dfFromSort = pd.DataFrame([row["row"] for row in api_output["rows"]])
31
+ try:
32
+ dfFromSort.sort_values(by=1, axis=1, ascending=True, inplace=False, kind='mergesort', na_position='last', ignore_index=False, key=None)
33
+ except:
34
+ print("Exception sorting due to keyerror?")
35
+ return dfFromSort
36
+
37
+ async def update_configs(dataset_name: str):
38
+ splits = await get_splits(dataset_name)
39
+ all_configs = sorted(set([s["config"] for s in splits["splits"]]))
40
+ return (gr.Dropdown.update(choices=all_configs, value=all_configs[0]),
41
+ splits)
42
+
43
+ async def update_splits(config_name: str, state: gr.State):
44
+ splits_for_config = sorted(set([s["split"] for s in state["splits"] if s["config"] == config_name]))
45
+ dataset_name = state["splits"][0]["dataset"]
46
+ dataset = await update_dataset(splits_for_config[0], config_name, dataset_name)
47
+ return (gr.Dropdown.update(choices=splits_for_config, value=splits_for_config[0]), dataset)
48
+
49
+ async def update_dataset(split_name: str, config_name: str, dataset_name: str):
50
+ rows = await get_first_rows(dataset_name, config_name, split_name)
51
+ df = get_df_from_rows(rows)
52
+ return df
53
+
54
+ # Guido von Roissum: https://www.youtube.com/watch?v=-DVyjdw4t9I
55
+ async def update_URL(dataset: str, config: str, split: str) -> str:
56
+ URL = f"https://datasets-server.huggingface.co/first-rows?dataset={dataset}&config={config}&split={split}"
57
+ URL = f"https://huggingface.co/datasets/{split}"
58
+ return (URL)
59
+
60
+ async def openurl(URL: str) -> str:
61
+ html = f"<a href={URL} target=_blank>{URL}</a>"
62
+ return (html)
63
+
64
+ with gr.Blocks() as demo:
65
+ gr.Markdown("<h1><center>🥫Datasets🎨</center></h1>")
66
+ gr.Markdown("""<div align="center">Curated Datasets: <a href = "https://www.kaggle.com/datasets">Kaggle</a>. <a href="https://www.nlm.nih.gov/research/umls/index.html">NLM UMLS</a>. <a href="https://loinc.org/downloads/">LOINC</a>. <a href="https://www.cms.gov/medicare/icd-10/2022-icd-10-cm">ICD10 Diagnosis</a>. <a href="https://icd.who.int/dev11/downloads">ICD11</a>. <a href="https://paperswithcode.com/datasets?q=medical&v=lst&o=newest">Papers,Code,Datasets for SOTA in Medicine</a>. <a href="https://paperswithcode.com/datasets?q=mental&v=lst&o=newest">Mental</a>. <a href="https://paperswithcode.com/datasets?q=behavior&v=lst&o=newest">Behavior</a>. <a href="https://www.cms.gov/medicare-coverage-database/downloads/downloads.aspx">CMS Downloads</a>. <a href="https://www.cms.gov/medicare/fraud-and-abuse/physicianselfreferral/list_of_codes">CMS CPT and HCPCS Procedures and Services</a> """)
67
+
68
+ splits_data = gr.State()
69
+
70
+ with gr.Row():
71
+ dataset_name = gr.Dropdown(label="Dataset", interactive=True)
72
+ config = gr.Dropdown(label="Subset", interactive=True)
73
+ split = gr.Dropdown(label="Split", interactive=True)
74
+
75
+ with gr.Row():
76
+ #filterleft = gr.Textbox(label="First Column Filter",placeholder="Filter Column 1")
77
+ URLcenter = gr.Textbox(label="Dataset URL", placeholder="URL")
78
+ btn = gr.Button("Use Dataset")
79
+ #URLoutput = gr.Textbox(label="Output",placeholder="URL Output")
80
+ URLoutput = gr.HTML(label="Output",placeholder="URL Output")
81
+
82
+ with gr.Row():
83
+ dataset = gr.DataFrame(wrap=True, interactive=True)
84
+
85
+ demo.load(get_valid_datasets, inputs=None, outputs=[dataset_name])
86
+
87
+ dataset_name.change(update_configs, inputs=[dataset_name], outputs=[config, splits_data])
88
+ config.change(update_splits, inputs=[config, splits_data], outputs=[split, dataset])
89
+ split.change(update_dataset, inputs=[split, config, dataset_name], outputs=[dataset])
90
+
91
+ dataset_name.change(update_URL, inputs=[split, config, dataset_name], outputs=[URLcenter])
92
+
93
+ btn.click(openurl, [URLcenter], URLoutput)
94
+
95
+ demo.launch(debug=True)
96
+
97
+ # original: https://huggingface.co/spaces/freddyaboulton/dataset-viewer -- Freddy thanks! Your examples are the best.
98
+ # playlist on Gradio and Mermaid: https://www.youtube.com/watch?v=o7kCD4aWMR4&list=PLHgX2IExbFosW7hWNryq8hs2bt2aj91R-
99
+ # Link to Mermaid model and code: [![](https://mermaid.ink/img/pako:eNp1U8mO2zAM_RXCZ-eQpZccCmSZTIpOMQESIAdnDrRMx0JkydXSNDOYfy_lpUgD1AfBfnx8fCTlj0SYgpJ5UipzFRVaD4flSQM_YjwafcVJ9-FCfrbYVGA0ZQeLUkt9futiOM72pEh4QFijR9iTf2tzsx3Z0ti6hxslvb_Lm0TSNPvBDhQsg1TFXXAag7NBef_9hdDqFA6knbEbdgvGwu7mjRXVkDOLOV-yNXmytdQEsoROvTfi4EhK9XTSxUNz_mo4uVHm1lPyce-uR1k_n2RHymHRNPAvNXaTT7NVZYwjeDECVbS4UiYUAyc2lc-yFoPXxkujHaAl2G54PCjIpfBssZAGtsZ5KlLYkjWXkMLiuOfjPVhiymr3_x4qS7wicneTFuMW6Gdxlb6Cb7oJvt1LbEpMso08sza8MnqskA9jL27Ij72Jafb0G-tGkQNTdgKOy_XcFP5GDxFbWsJLV3FQid2LWfZsfpHVqAXBCBYa1e2dAHUBu5Ar6dgby0ghPWxQWk2Oh_L0M0h_S2Ep0YHUrXFHXD_msefo5XEkfFWBK8atdkA7mgfoalpATJI0qfnWoCz4b_iI0VPiK6rplMz5taASg_Kn5KQ_mYrBm_1Ni2TubaA0CU2BntYSeQl1Mi9ROfr8A8FBGds?type=png)](https://mermaid.live/edit#pako:eNp1U8mO2zAM_RXCZ-eQpZccCmSZTIpOMQESIAdnDrRMx0JkydXSNDOYfy_lpUgD1AfBfnx8fCTlj0SYgpJ5UipzFRVaD4flSQM_YjwafcVJ9-FCfrbYVGA0ZQeLUkt9futiOM72pEh4QFijR9iTf2tzsx3Z0ti6hxslvb_Lm0TSNPvBDhQsg1TFXXAag7NBef_9hdDqFA6knbEbdgvGwu7mjRXVkDOLOV-yNXmytdQEsoROvTfi4EhK9XTSxUNz_mo4uVHm1lPyce-uR1k_n2RHymHRNPAvNXaTT7NVZYwjeDECVbS4UiYUAyc2lc-yFoPXxkujHaAl2G54PCjIpfBssZAGtsZ5KlLYkjWXkMLiuOfjPVhiymr3_x4qS7wicneTFuMW6Gdxlb6Cb7oJvt1LbEpMso08sza8MnqskA9jL27Ij72Jafb0G-tGkQNTdgKOy_XcFP5GDxFbWsJLV3FQid2LWfZsfpHVqAXBCBYa1e2dAHUBu5Ar6dgby0ghPWxQWk2Oh_L0M0h_S2Ep0YHUrXFHXD_msefo5XEkfFWBK8atdkA7mgfoalpATJI0qfnWoCz4b_iI0VPiK6rplMz5taASg_Kn5KQ_mYrBm_1Ni2TubaA0CU2BntYSeQl1Mi9ROfr8A8FBGds)