Condense query functions
Browse files- functions/__init__.py +2 -2
- functions/chat_functions.py +5 -5
- functions/query_functions.py +18 -75
- templates/data_file.py +135 -135
- tools/tools.py +8 -8
functions/__init__.py
CHANGED
@@ -1,9 +1,9 @@
|
|
1 |
-
from .query_functions import
|
2 |
from .chart_functions import table_generation_func, scatter_chart_generation_func, \
|
3 |
line_chart_generation_func, bar_chart_generation_func, pie_chart_generation_func, histogram_generation_func, scatter_chart_fig
|
4 |
from .chat_functions import example_question_generator, chatbot_func
|
5 |
from .stat_functions import regression_func
|
6 |
|
7 |
-
__all__ = ["
|
8 |
"line_chart_generation_func","bar_chart_generation_func","regression_func", "pie_chart_generation_func", "histogram_generation_func",
|
9 |
"scatter_chart_fig","example_question_generator","chatbot_func"]
|
|
|
1 |
+
from .query_functions import graphql_schema_query, graphql_csv_query, query_func
|
2 |
from .chart_functions import table_generation_func, scatter_chart_generation_func, \
|
3 |
line_chart_generation_func, bar_chart_generation_func, pie_chart_generation_func, histogram_generation_func, scatter_chart_fig
|
4 |
from .chat_functions import example_question_generator, chatbot_func
|
5 |
from .stat_functions import regression_func
|
6 |
|
7 |
+
__all__ = ["query_func","graphql_schema_query","graphql_csv_query","table_generation_func","scatter_chart_generation_func",
|
8 |
"line_chart_generation_func","bar_chart_generation_func","regression_func", "pie_chart_generation_func", "histogram_generation_func",
|
9 |
"scatter_chart_fig","example_question_generator","chatbot_func"]
|
functions/chat_functions.py
CHANGED
@@ -62,7 +62,8 @@ def example_question_generator(session_hash, data_source, name, titles, schema):
|
|
62 |
return example_response["replies"][0].text
|
63 |
|
64 |
def system_message(data_source, titles, schema=""):
|
65 |
-
|
|
|
66 |
system_message_dict = {
|
67 |
'file_upload' : f"""You are a helpful and knowledgeable agent who has access to an SQLite database which has a table called 'data_source' that contains the following columns: {titles}.
|
68 |
You also have access to a function, called table_generation_func, that can take a query.csv file generated from our sql query and returns an iframe that we should display in our chat window.
|
@@ -111,13 +112,12 @@ def system_message(data_source, titles, schema=""):
|
|
111 |
return system_message_dict[data_source]
|
112 |
|
113 |
def chatbot_func(message, history, session_hash, data_source, titles, schema, *args):
|
114 |
-
from functions import
|
115 |
-
|
116 |
line_chart_generation_func,bar_chart_generation_func,pie_chart_generation_func,histogram_generation_func
|
117 |
import tools.tools as tools
|
118 |
|
119 |
-
available_functions = {"
|
120 |
-
"graphql_query_func": graphql_query_func,"graphql_schema_query": graphql_schema_query,"graphql_csv_query": graphql_csv_query,
|
121 |
"table_generation_func":table_generation_func,
|
122 |
"line_chart_generation_func":line_chart_generation_func,"bar_chart_generation_func":bar_chart_generation_func,
|
123 |
"scatter_chart_generation_func":scatter_chart_generation_func, "pie_chart_generation_func":pie_chart_generation_func,
|
|
|
62 |
return example_response["replies"][0].text
|
63 |
|
64 |
def system_message(data_source, titles, schema=""):
|
65 |
+
print("TITLES")
|
66 |
+
print(titles)
|
67 |
system_message_dict = {
|
68 |
'file_upload' : f"""You are a helpful and knowledgeable agent who has access to an SQLite database which has a table called 'data_source' that contains the following columns: {titles}.
|
69 |
You also have access to a function, called table_generation_func, that can take a query.csv file generated from our sql query and returns an iframe that we should display in our chat window.
|
|
|
112 |
return system_message_dict[data_source]
|
113 |
|
114 |
def chatbot_func(message, history, session_hash, data_source, titles, schema, *args):
|
115 |
+
from functions import table_generation_func, regression_func, scatter_chart_generation_func, \
|
116 |
+
query_func, graphql_schema_query, graphql_csv_query, \
|
117 |
line_chart_generation_func,bar_chart_generation_func,pie_chart_generation_func,histogram_generation_func
|
118 |
import tools.tools as tools
|
119 |
|
120 |
+
available_functions = {"query_func":query_func,"graphql_schema_query": graphql_schema_query,"graphql_csv_query": graphql_csv_query,
|
|
|
121 |
"table_generation_func":table_generation_func,
|
122 |
"line_chart_generation_func":line_chart_generation_func,"bar_chart_generation_func":bar_chart_generation_func,
|
123 |
"scatter_chart_generation_func":scatter_chart_generation_func, "pie_chart_generation_func":pie_chart_generation_func,
|
functions/query_functions.py
CHANGED
@@ -35,28 +35,6 @@ class SQLiteQuery:
|
|
35 |
self.connection.close()
|
36 |
return {"results": results, "queries": queries, "csv_columns": column_names}
|
37 |
|
38 |
-
|
39 |
-
|
40 |
-
def sqlite_query_func(queries: List[str], session_hash, **kwargs):
|
41 |
-
dir_path = TEMP_DIR / str(session_hash)
|
42 |
-
sql_query = SQLiteQuery(f'{dir_path}/file_upload/data_source.db')
|
43 |
-
try:
|
44 |
-
result = sql_query.run(queries, session_hash)
|
45 |
-
if len(result["results"][0]) > 1000:
|
46 |
-
print("QUERY TOO LARGE")
|
47 |
-
return {"reply": f"""query result too large to be processed by llm, the query results are in our query.csv file.
|
48 |
-
The column names of this query.csv file are: {result["csv_columns"]}.
|
49 |
-
If you need to display the results directly, perhaps use the table_generation_func function."""}
|
50 |
-
else:
|
51 |
-
return {"reply": result["results"][0]}
|
52 |
-
|
53 |
-
except Exception as e:
|
54 |
-
reply = f"""There was an error running the SQL Query = {queries}
|
55 |
-
The error is {e},
|
56 |
-
You should probably try again.
|
57 |
-
"""
|
58 |
-
return {"reply": reply}
|
59 |
-
|
60 |
@component
|
61 |
class PostgreSQLQuery:
|
62 |
|
@@ -82,30 +60,6 @@ class PostgreSQLQuery:
|
|
82 |
results.append(f"{result}")
|
83 |
self.connection.close()
|
84 |
return {"results": results, "queries": queries, "csv_columns": column_names}
|
85 |
-
|
86 |
-
|
87 |
-
|
88 |
-
def sql_query_func(queries: List[str], session_hash, args, **kwargs):
|
89 |
-
sql_query = PostgreSQLQuery(args[0], args[1], args[2], args[3], args[4])
|
90 |
-
try:
|
91 |
-
result = sql_query.run(queries, session_hash)
|
92 |
-
print("RESULT")
|
93 |
-
print(result)
|
94 |
-
if len(result["results"][0]) > 1000:
|
95 |
-
print("QUERY TOO LARGE")
|
96 |
-
return {"reply": f"""query result too large to be processed by llm, the query results are in our query.csv file.
|
97 |
-
The column names of this query.csv file are: {result["csv_columns"]}.
|
98 |
-
If you need to display the results directly, perhaps use the table_generation_func function."""}
|
99 |
-
else:
|
100 |
-
return {"reply": result["results"][0]}
|
101 |
-
|
102 |
-
except Exception as e:
|
103 |
-
reply = f"""There was an error running the SQL Query = {queries}
|
104 |
-
The error is {e},
|
105 |
-
You should probably try again.
|
106 |
-
"""
|
107 |
-
print(reply)
|
108 |
-
return {"reply": reply}
|
109 |
|
110 |
@component
|
111 |
class DocDBQuery:
|
@@ -155,29 +109,6 @@ class DocDBQuery:
|
|
155 |
self.client.close()
|
156 |
return {"results": results, "queries": aggregation_pipeline, "csv_columns": column_names}
|
157 |
|
158 |
-
|
159 |
-
|
160 |
-
def doc_db_query_func(aggregation_pipeline: List[str], db_collection: AnyStr, session_hash, args, **kwargs):
|
161 |
-
doc_db_query = DocDBQuery(args[0], args[1])
|
162 |
-
try:
|
163 |
-
result = doc_db_query.run(aggregation_pipeline, db_collection, session_hash)
|
164 |
-
print("RESULT")
|
165 |
-
if len(result["results"][0]) > 1000:
|
166 |
-
print("QUERY TOO LARGE")
|
167 |
-
return {"reply": f"""query result too large to be processed by llm, the query results are in our query.csv file.
|
168 |
-
The column names of this query.csv file are: {result["csv_columns"]}.
|
169 |
-
If you need to display the results directly, perhaps use the table_generation_func function."""}
|
170 |
-
else:
|
171 |
-
return {"reply": result["results"][0]}
|
172 |
-
|
173 |
-
except Exception as e:
|
174 |
-
reply = f"""There was an error running the NoSQL (Mongo) Query = {aggregation_pipeline}
|
175 |
-
The error is {e},
|
176 |
-
You should probably try again.
|
177 |
-
"""
|
178 |
-
print(reply)
|
179 |
-
return {"reply": reply}
|
180 |
-
|
181 |
@component
|
182 |
class GraphQLQuery:
|
183 |
|
@@ -214,12 +145,23 @@ class GraphQLQuery:
|
|
214 |
results.append(f"{response_frame}")
|
215 |
return {"results": results, "queries": graphql_query, "csv_columns": column_names}
|
216 |
|
217 |
-
|
218 |
-
|
219 |
-
def graphql_query_func(graphql_query: AnyStr, session_hash, args, **kwargs):
|
220 |
-
graphql_object = GraphQLQuery()
|
221 |
try:
|
222 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
223 |
print("RESULT")
|
224 |
if len(result["results"][0]) > 1000:
|
225 |
print("QUERY TOO LARGE")
|
@@ -230,7 +172,7 @@ def graphql_query_func(graphql_query: AnyStr, session_hash, args, **kwargs):
|
|
230 |
return {"reply": result["results"][0]}
|
231 |
|
232 |
except Exception as e:
|
233 |
-
reply = f"""There was an error running the
|
234 |
The error is {e},
|
235 |
You should probably try again.
|
236 |
"""
|
@@ -266,6 +208,7 @@ def graphql_csv_query(csv_query: AnyStr, session_hash, **kwargs):
|
|
266 |
query = pd.read_csv(f'{dir_path}/graphql/query.csv')
|
267 |
query.Name = 'query'
|
268 |
print("GRAPHQL CSV QUERY")
|
|
|
269 |
queried_df = sqldf(csv_query, locals())
|
270 |
print(queried_df)
|
271 |
column_names = list(queried_df.columns)
|
|
|
35 |
self.connection.close()
|
36 |
return {"results": results, "queries": queries, "csv_columns": column_names}
|
37 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
38 |
@component
|
39 |
class PostgreSQLQuery:
|
40 |
|
|
|
60 |
results.append(f"{result}")
|
61 |
self.connection.close()
|
62 |
return {"results": results, "queries": queries, "csv_columns": column_names}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
63 |
|
64 |
@component
|
65 |
class DocDBQuery:
|
|
|
109 |
self.client.close()
|
110 |
return {"results": results, "queries": aggregation_pipeline, "csv_columns": column_names}
|
111 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
112 |
@component
|
113 |
class GraphQLQuery:
|
114 |
|
|
|
145 |
results.append(f"{response_frame}")
|
146 |
return {"results": results, "queries": graphql_query, "csv_columns": column_names}
|
147 |
|
148 |
+
def query_func(queries:List[str], session_hash, session_folder, args, **kwargs):
|
|
|
|
|
|
|
149 |
try:
|
150 |
+
print("QUERY")
|
151 |
+
print(queries)
|
152 |
+
if session_folder == "file_upload":
|
153 |
+
dir_path = TEMP_DIR / str(session_hash)
|
154 |
+
sql_query = SQLiteQuery(f'{dir_path}/file_upload/data_source.db')
|
155 |
+
result = sql_query.run(queries, session_hash)
|
156 |
+
elif session_folder == "sql":
|
157 |
+
sql_query = PostgreSQLQuery(args[0], args[1], args[2], args[3], args[4])
|
158 |
+
result = sql_query.run(queries, session_hash)
|
159 |
+
elif session_folder == 'doc_db':
|
160 |
+
doc_db_query = DocDBQuery(args[0], args[1])
|
161 |
+
result = doc_db_query.run(queries, kwargs['db_collection'], session_hash)
|
162 |
+
elif session_folder == 'graphql':
|
163 |
+
graphql_object = GraphQLQuery()
|
164 |
+
result = graphql_object.run(queries, args[0], args[1], args[2], session_hash)
|
165 |
print("RESULT")
|
166 |
if len(result["results"][0]) > 1000:
|
167 |
print("QUERY TOO LARGE")
|
|
|
172 |
return {"reply": result["results"][0]}
|
173 |
|
174 |
except Exception as e:
|
175 |
+
reply = f"""There was an error running the {session_folder} Query = {queries}
|
176 |
The error is {e},
|
177 |
You should probably try again.
|
178 |
"""
|
|
|
208 |
query = pd.read_csv(f'{dir_path}/graphql/query.csv')
|
209 |
query.Name = 'query'
|
210 |
print("GRAPHQL CSV QUERY")
|
211 |
+
print(csv_query)
|
212 |
queried_df = sqldf(csv_query, locals())
|
213 |
print(queried_df)
|
214 |
column_names = list(queried_df.columns)
|
templates/data_file.py
CHANGED
@@ -1,136 +1,136 @@
|
|
1 |
-
import gradio as gr
|
2 |
-
from functions import example_question_generator, chatbot_func
|
3 |
-
from data_sources import process_data_upload
|
4 |
-
from utils import message_dict
|
5 |
-
import ast
|
6 |
-
|
7 |
-
def run_example(input):
|
8 |
-
return input
|
9 |
-
|
10 |
-
def example_display(input):
|
11 |
-
if input == None:
|
12 |
-
display = True
|
13 |
-
else:
|
14 |
-
display = False
|
15 |
-
return [gr.update(visible=display),gr.update(visible=display),gr.update(visible=display),gr.update(visible=display)]
|
16 |
-
|
17 |
-
with gr.Blocks() as demo:
|
18 |
-
description = gr.HTML("""
|
19 |
-
<!-- Header -->
|
20 |
-
<div class="max-w-4xl mx-auto mb-12 text-center">
|
21 |
-
<div class="bg-blue-50 border border-blue-200 rounded-lg max-w-2xl mx-auto">
|
22 |
-
<h2 class="font-semibold text-blue-800 ">
|
23 |
-
<i class="fas fa-info-circle mr-2"></i>Supported Files
|
24 |
-
</h2>
|
25 |
-
<div class="flex flex-wrap justify-center gap-3 pb-4 text-blue-700">
|
26 |
-
<span class="tooltip">
|
27 |
-
<i class="fas fa-file-csv mr-1"></i>CSV
|
28 |
-
<span class="tooltip-text">Comma-separated values</span>
|
29 |
-
</span>
|
30 |
-
<span class="tooltip">
|
31 |
-
<i class="fas fa-file-alt mr-1"></i>TSV
|
32 |
-
<span class="tooltip-text">Tab-separated values</span>
|
33 |
-
</span>
|
34 |
-
<span class="tooltip">
|
35 |
-
<i class="fas fa-file-alt mr-1"></i>TXT
|
36 |
-
<span class="tooltip-text">Text files</span>
|
37 |
-
</span>
|
38 |
-
<span class="tooltip">
|
39 |
-
<i class="fas fa-file-excel mr-1"></i>XLS/XLSX
|
40 |
-
<span class="tooltip-text">Excel spreadsheets</span>
|
41 |
-
</span>
|
42 |
-
<span class="tooltip">
|
43 |
-
<i class="fas fa-file-code mr-1"></i>XML
|
44 |
-
<span class="tooltip-text">XML documents</span>
|
45 |
-
</span>
|
46 |
-
<span class="tooltip">
|
47 |
-
<i class="fas fa-file-code mr-1"></i>JSON
|
48 |
-
<span class="tooltip-text">JSON data files</span>
|
49 |
-
</span>
|
50 |
-
</div>
|
51 |
-
</div>
|
52 |
-
</div>
|
53 |
-
""", elem_classes="description_component")
|
54 |
-
example_file_1 = gr.File(visible=False, value="samples/bank_marketing_campaign.csv")
|
55 |
-
example_file_2 = gr.File(visible=False, value="samples/online_retail_data.csv")
|
56 |
-
example_file_3 = gr.File(visible=False, value="samples/tb_illness_data.csv")
|
57 |
-
with gr.Row():
|
58 |
-
example_btn_1 = gr.Button(value="Try Me: bank_marketing_campaign.csv", elem_classes="sample-btn bg-gradient-to-r from-purple-500 to-indigo-600 text-white p-6 rounded-lg text-left hover:shadow-lg", size="md", variant="primary")
|
59 |
-
example_btn_2 = gr.Button(value="Try Me: online_retail_data.csv", elem_classes="sample-btn bg-gradient-to-r from-purple-500 to-indigo-600 text-white p-6 rounded-lg text-left hover:shadow-lg", size="md", variant="primary")
|
60 |
-
example_btn_3 = gr.Button(value="Try Me: tb_illness_data.csv", elem_classes="sample-btn bg-gradient-to-r from-purple-500 to-indigo-600 text-white p-6 rounded-lg text-left hover:shadow-lg", size="md", variant="primary")
|
61 |
-
|
62 |
-
file_output = gr.File(label="Data File (CSV, TSV, TXT, XLS, XLSX, XML, JSON)", show_label=True, elem_classes="file_marker drop-zone border-2 border-dashed border-gray-300 rounded-lg hover:border-primary cursor-pointer bg-gray-50 hover:bg-blue-50 transition-colors duration-300", file_types=['.csv','.xlsx','.txt','.json','.ndjson','.xml','.xls','.tsv'])
|
63 |
-
example_btn_1.click(fn=run_example, inputs=example_file_1, outputs=file_output)
|
64 |
-
example_btn_2.click(fn=run_example, inputs=example_file_2, outputs=file_output)
|
65 |
-
example_btn_3.click(fn=run_example, inputs=example_file_3, outputs=file_output)
|
66 |
-
file_output.change(fn=example_display, inputs=file_output, outputs=[example_btn_1, example_btn_2, example_btn_3, description])
|
67 |
-
|
68 |
-
@gr.render(inputs=file_output)
|
69 |
-
def data_options(filename, request: gr.Request):
|
70 |
-
print(filename)
|
71 |
-
if request.session_hash not in message_dict:
|
72 |
-
message_dict[request.session_hash] = {}
|
73 |
-
message_dict[request.session_hash]['file_upload'] = None
|
74 |
-
if filename:
|
75 |
-
process_message = process_upload(filename, request.session_hash)
|
76 |
-
gr.HTML(value=process_message[1], padding=False)
|
77 |
-
if process_message[0] == "success":
|
78 |
-
if "bank_marketing_campaign" in filename:
|
79 |
-
example_questions = [
|
80 |
-
["Describe the dataset"],
|
81 |
-
["What levels of education have the highest and lowest average balance?"],
|
82 |
-
["What job is most and least common for a yes response from the individuals, not counting 'unknown'?"],
|
83 |
-
["Can you generate a bar chart of education vs. average balance?"],
|
84 |
-
["Can you generate a table of levels of education versus average balance, percent married, percent with a loan, and percent in default?"],
|
85 |
-
["Can we predict the relationship between the number of contacts performed before this campaign and the average balance?"],
|
86 |
-
["Can you plot the number of contacts performed before this campaign versus the duration and use balance as the size in a bubble chart?"]
|
87 |
-
]
|
88 |
-
elif "online_retail_data" in filename:
|
89 |
-
example_questions = [
|
90 |
-
["Describe the dataset"],
|
91 |
-
["What month had the highest revenue?"],
|
92 |
-
["Is revenue higher in the morning or afternoon?"],
|
93 |
-
["Can you generate a line graph of revenue per month?"],
|
94 |
-
["Can you generate a table of revenue per month?"],
|
95 |
-
["Can we predict how time of day affects transaction value in this data set?"],
|
96 |
-
["Can you plot revenue per month with size being the number of units sold that month in a bubble chart?"]
|
97 |
-
]
|
98 |
-
else:
|
99 |
-
try:
|
100 |
-
generated_examples = ast.literal_eval(example_question_generator(request.session_hash, 'file_upload', '', process_message[1], ''))
|
101 |
-
example_questions = [
|
102 |
-
["Describe the dataset"]
|
103 |
-
]
|
104 |
-
for example in generated_examples:
|
105 |
-
example_questions.append([example])
|
106 |
-
except Exception as e:
|
107 |
-
print("DATA FILE QUESTION GENERATION ERROR")
|
108 |
-
print(e)
|
109 |
-
example_questions = [
|
110 |
-
["Describe the dataset"],
|
111 |
-
["List the columns in the dataset"],
|
112 |
-
["What could this data be used for?"],
|
113 |
-
]
|
114 |
-
session_hash = gr.Textbox(visible=False, value=request.session_hash)
|
115 |
-
data_source = gr.Textbox(visible=False, value='file_upload')
|
116 |
-
schema = gr.Textbox(visible=False, value='')
|
117 |
-
titles = gr.Textbox(value=process_message[2], interactive=False, visible=False)
|
118 |
-
bot = gr.Chatbot(type='messages', label="CSV Chat Window", render_markdown=True, sanitize_html=False, show_label=True, render=False, visible=True, elem_classes="chatbot")
|
119 |
-
chat = gr.ChatInterface(
|
120 |
-
fn=chatbot_func,
|
121 |
-
type='messages',
|
122 |
-
chatbot=bot,
|
123 |
-
title="Chat with your data file",
|
124 |
-
concurrency_limit=None,
|
125 |
-
examples=example_questions,
|
126 |
-
additional_inputs=[session_hash, data_source, titles, schema]
|
127 |
-
)
|
128 |
-
|
129 |
-
def process_upload(upload_value, session_hash):
|
130 |
-
if upload_value:
|
131 |
-
process_message = process_data_upload(upload_value, session_hash)
|
132 |
-
return process_message
|
133 |
-
|
134 |
-
|
135 |
-
if __name__ == "__main__":
|
136 |
demo.launch()
|
|
|
1 |
+
import gradio as gr
|
2 |
+
from functions import example_question_generator, chatbot_func
|
3 |
+
from data_sources import process_data_upload
|
4 |
+
from utils import message_dict
|
5 |
+
import ast
|
6 |
+
|
7 |
+
def run_example(input):
|
8 |
+
return input
|
9 |
+
|
10 |
+
def example_display(input):
|
11 |
+
if input == None:
|
12 |
+
display = True
|
13 |
+
else:
|
14 |
+
display = False
|
15 |
+
return [gr.update(visible=display),gr.update(visible=display),gr.update(visible=display),gr.update(visible=display)]
|
16 |
+
|
17 |
+
with gr.Blocks() as demo:
|
18 |
+
description = gr.HTML("""
|
19 |
+
<!-- Header -->
|
20 |
+
<div class="max-w-4xl mx-auto mb-12 text-center">
|
21 |
+
<div class="bg-blue-50 border border-blue-200 rounded-lg max-w-2xl mx-auto">
|
22 |
+
<h2 class="font-semibold text-blue-800 ">
|
23 |
+
<i class="fas fa-info-circle mr-2"></i>Supported Files
|
24 |
+
</h2>
|
25 |
+
<div class="flex flex-wrap justify-center gap-3 pb-4 text-blue-700">
|
26 |
+
<span class="tooltip">
|
27 |
+
<i class="fas fa-file-csv mr-1"></i>CSV
|
28 |
+
<span class="tooltip-text">Comma-separated values</span>
|
29 |
+
</span>
|
30 |
+
<span class="tooltip">
|
31 |
+
<i class="fas fa-file-alt mr-1"></i>TSV
|
32 |
+
<span class="tooltip-text">Tab-separated values</span>
|
33 |
+
</span>
|
34 |
+
<span class="tooltip">
|
35 |
+
<i class="fas fa-file-alt mr-1"></i>TXT
|
36 |
+
<span class="tooltip-text">Text files</span>
|
37 |
+
</span>
|
38 |
+
<span class="tooltip">
|
39 |
+
<i class="fas fa-file-excel mr-1"></i>XLS/XLSX
|
40 |
+
<span class="tooltip-text">Excel spreadsheets</span>
|
41 |
+
</span>
|
42 |
+
<span class="tooltip">
|
43 |
+
<i class="fas fa-file-code mr-1"></i>XML
|
44 |
+
<span class="tooltip-text">XML documents</span>
|
45 |
+
</span>
|
46 |
+
<span class="tooltip">
|
47 |
+
<i class="fas fa-file-code mr-1"></i>JSON
|
48 |
+
<span class="tooltip-text">JSON data files</span>
|
49 |
+
</span>
|
50 |
+
</div>
|
51 |
+
</div>
|
52 |
+
</div>
|
53 |
+
""", elem_classes="description_component")
|
54 |
+
example_file_1 = gr.File(visible=False, value="samples/bank_marketing_campaign.csv")
|
55 |
+
example_file_2 = gr.File(visible=False, value="samples/online_retail_data.csv")
|
56 |
+
example_file_3 = gr.File(visible=False, value="samples/tb_illness_data.csv")
|
57 |
+
with gr.Row():
|
58 |
+
example_btn_1 = gr.Button(value="Try Me: bank_marketing_campaign.csv", elem_classes="sample-btn bg-gradient-to-r from-purple-500 to-indigo-600 text-white p-6 rounded-lg text-left hover:shadow-lg", size="md", variant="primary")
|
59 |
+
example_btn_2 = gr.Button(value="Try Me: online_retail_data.csv", elem_classes="sample-btn bg-gradient-to-r from-purple-500 to-indigo-600 text-white p-6 rounded-lg text-left hover:shadow-lg", size="md", variant="primary")
|
60 |
+
example_btn_3 = gr.Button(value="Try Me: tb_illness_data.csv", elem_classes="sample-btn bg-gradient-to-r from-purple-500 to-indigo-600 text-white p-6 rounded-lg text-left hover:shadow-lg", size="md", variant="primary")
|
61 |
+
|
62 |
+
file_output = gr.File(label="Data File (CSV, TSV, TXT, XLS, XLSX, XML, JSON)", show_label=True, elem_classes="file_marker drop-zone border-2 border-dashed border-gray-300 rounded-lg hover:border-primary cursor-pointer bg-gray-50 hover:bg-blue-50 transition-colors duration-300", file_types=['.csv','.xlsx','.txt','.json','.ndjson','.xml','.xls','.tsv'])
|
63 |
+
example_btn_1.click(fn=run_example, inputs=example_file_1, outputs=file_output)
|
64 |
+
example_btn_2.click(fn=run_example, inputs=example_file_2, outputs=file_output)
|
65 |
+
example_btn_3.click(fn=run_example, inputs=example_file_3, outputs=file_output)
|
66 |
+
file_output.change(fn=example_display, inputs=file_output, outputs=[example_btn_1, example_btn_2, example_btn_3, description])
|
67 |
+
|
68 |
+
@gr.render(inputs=file_output)
|
69 |
+
def data_options(filename, request: gr.Request):
|
70 |
+
print(filename)
|
71 |
+
if request.session_hash not in message_dict:
|
72 |
+
message_dict[request.session_hash] = {}
|
73 |
+
message_dict[request.session_hash]['file_upload'] = None
|
74 |
+
if filename:
|
75 |
+
process_message = process_upload(filename, request.session_hash)
|
76 |
+
gr.HTML(value=process_message[1], padding=False)
|
77 |
+
if process_message[0] == "success":
|
78 |
+
if "bank_marketing_campaign" in filename:
|
79 |
+
example_questions = [
|
80 |
+
["Describe the dataset"],
|
81 |
+
["What levels of education have the highest and lowest average balance?"],
|
82 |
+
["What job is most and least common for a yes response from the individuals, not counting 'unknown'?"],
|
83 |
+
["Can you generate a bar chart of education vs. average balance?"],
|
84 |
+
["Can you generate a table of levels of education versus average balance, percent married, percent with a loan, and percent in default?"],
|
85 |
+
["Can we predict the relationship between the number of contacts performed before this campaign and the average balance?"],
|
86 |
+
["Can you plot the number of contacts performed before this campaign versus the duration and use balance as the size in a bubble chart?"]
|
87 |
+
]
|
88 |
+
elif "online_retail_data" in filename:
|
89 |
+
example_questions = [
|
90 |
+
["Describe the dataset"],
|
91 |
+
["What month had the highest revenue?"],
|
92 |
+
["Is revenue higher in the morning or afternoon?"],
|
93 |
+
["Can you generate a line graph of revenue per month?"],
|
94 |
+
["Can you generate a table of revenue per month?"],
|
95 |
+
["Can we predict how time of day affects transaction value in this data set?"],
|
96 |
+
["Can you plot revenue per month with size being the number of units sold that month in a bubble chart?"]
|
97 |
+
]
|
98 |
+
else:
|
99 |
+
try:
|
100 |
+
generated_examples = ast.literal_eval(example_question_generator(request.session_hash, 'file_upload', '', process_message[1], ''))
|
101 |
+
example_questions = [
|
102 |
+
["Describe the dataset"]
|
103 |
+
]
|
104 |
+
for example in generated_examples:
|
105 |
+
example_questions.append([example])
|
106 |
+
except Exception as e:
|
107 |
+
print("DATA FILE QUESTION GENERATION ERROR")
|
108 |
+
print(e)
|
109 |
+
example_questions = [
|
110 |
+
["Describe the dataset"],
|
111 |
+
["List the columns in the dataset"],
|
112 |
+
["What could this data be used for?"],
|
113 |
+
]
|
114 |
+
session_hash = gr.Textbox(visible=False, value=request.session_hash)
|
115 |
+
data_source = gr.Textbox(visible=False, value='file_upload')
|
116 |
+
schema = gr.Textbox(visible=False, value='')
|
117 |
+
titles = gr.Textbox(value=process_message[2], interactive=False, visible=False)
|
118 |
+
bot = gr.Chatbot(type='messages', label="CSV Chat Window", render_markdown=True, sanitize_html=False, show_label=True, render=False, visible=True, elem_classes="chatbot")
|
119 |
+
chat = gr.ChatInterface(
|
120 |
+
fn=chatbot_func,
|
121 |
+
type='messages',
|
122 |
+
chatbot=bot,
|
123 |
+
title="Chat with your data file",
|
124 |
+
concurrency_limit=None,
|
125 |
+
examples=example_questions,
|
126 |
+
additional_inputs=[session_hash, data_source, titles, schema]
|
127 |
+
)
|
128 |
+
|
129 |
+
def process_upload(upload_value, session_hash):
|
130 |
+
if upload_value:
|
131 |
+
process_message = process_data_upload(upload_value, session_hash)
|
132 |
+
return process_message
|
133 |
+
|
134 |
+
|
135 |
+
if __name__ == "__main__":
|
136 |
demo.launch()
|
tools/tools.py
CHANGED
@@ -10,7 +10,7 @@ def tools_call(session_hash, data_source, titles):
|
|
10 |
{
|
11 |
"type": "function",
|
12 |
"function": {
|
13 |
-
"name": "
|
14 |
"description": f"""This is a tool useful to query a SQLite table called 'data_source' with the following Columns: {titles_string}.
|
15 |
There may also be more columns in the table if the number of columns is too large to process.
|
16 |
This function also saves the results of the query to csv file called query.csv.""",
|
@@ -34,7 +34,7 @@ def tools_call(session_hash, data_source, titles):
|
|
34 |
{
|
35 |
"type": "function",
|
36 |
"function": {
|
37 |
-
"name": "
|
38 |
"description": f"""This is a tool useful to query a PostgreSQL database with the following tables, {titles_string}.
|
39 |
There may also be more tables in the database if the number of tables is too large to process.
|
40 |
This function also saves the results of the query to csv file called query.csv.""",
|
@@ -58,14 +58,14 @@ def tools_call(session_hash, data_source, titles):
|
|
58 |
{
|
59 |
"type": "function",
|
60 |
"function": {
|
61 |
-
"name": "
|
62 |
"description": f"""This is a tool useful to build an aggregation pipeline to query a MongoDB NoSQL document database with the following collections, {titles_string}.
|
63 |
There may also be more collections in the database if the number of tables is too large to process.
|
64 |
This function also saves the results of the query to a csv file called query.csv.""",
|
65 |
"parameters": {
|
66 |
"type": "object",
|
67 |
"properties": {
|
68 |
-
"
|
69 |
"type": "string",
|
70 |
"description": "The MongoDB aggregation pipeline to use in the search. Infer this from the user's message. It should be a question or a statement."
|
71 |
},
|
@@ -74,7 +74,7 @@ def tools_call(session_hash, data_source, titles):
|
|
74 |
"description": "The MongoDB collection to use in the search. Infer this from the user's message. It should be a question or a statement.",
|
75 |
}
|
76 |
},
|
77 |
-
"required": ["
|
78 |
},
|
79 |
},
|
80 |
},
|
@@ -83,19 +83,19 @@ def tools_call(session_hash, data_source, titles):
|
|
83 |
{
|
84 |
"type": "function",
|
85 |
"function": {
|
86 |
-
"name": "
|
87 |
"description": f"""This is a tool useful to build a GraphQL query for a GraphQL API endpoint with the following types, {titles_string}.
|
88 |
There may also be more types in the GraphQL endpoint if the number of types is too large to process.
|
89 |
This function also saves the results of the query to a csv file called query.csv.""",
|
90 |
"parameters": {
|
91 |
"type": "object",
|
92 |
"properties": {
|
93 |
-
"
|
94 |
"type": "string",
|
95 |
"description": "The GraphQL query to use in the search. Infer this from the user's message. It should be a question or a statement."
|
96 |
}
|
97 |
},
|
98 |
-
"required": ["
|
99 |
},
|
100 |
},
|
101 |
},
|
|
|
10 |
{
|
11 |
"type": "function",
|
12 |
"function": {
|
13 |
+
"name": "query_func",
|
14 |
"description": f"""This is a tool useful to query a SQLite table called 'data_source' with the following Columns: {titles_string}.
|
15 |
There may also be more columns in the table if the number of columns is too large to process.
|
16 |
This function also saves the results of the query to csv file called query.csv.""",
|
|
|
34 |
{
|
35 |
"type": "function",
|
36 |
"function": {
|
37 |
+
"name": "query_func",
|
38 |
"description": f"""This is a tool useful to query a PostgreSQL database with the following tables, {titles_string}.
|
39 |
There may also be more tables in the database if the number of tables is too large to process.
|
40 |
This function also saves the results of the query to csv file called query.csv.""",
|
|
|
58 |
{
|
59 |
"type": "function",
|
60 |
"function": {
|
61 |
+
"name": "query_func",
|
62 |
"description": f"""This is a tool useful to build an aggregation pipeline to query a MongoDB NoSQL document database with the following collections, {titles_string}.
|
63 |
There may also be more collections in the database if the number of tables is too large to process.
|
64 |
This function also saves the results of the query to a csv file called query.csv.""",
|
65 |
"parameters": {
|
66 |
"type": "object",
|
67 |
"properties": {
|
68 |
+
"queries": {
|
69 |
"type": "string",
|
70 |
"description": "The MongoDB aggregation pipeline to use in the search. Infer this from the user's message. It should be a question or a statement."
|
71 |
},
|
|
|
74 |
"description": "The MongoDB collection to use in the search. Infer this from the user's message. It should be a question or a statement.",
|
75 |
}
|
76 |
},
|
77 |
+
"required": ["queries","db_collection"],
|
78 |
},
|
79 |
},
|
80 |
},
|
|
|
83 |
{
|
84 |
"type": "function",
|
85 |
"function": {
|
86 |
+
"name": "query_func",
|
87 |
"description": f"""This is a tool useful to build a GraphQL query for a GraphQL API endpoint with the following types, {titles_string}.
|
88 |
There may also be more types in the GraphQL endpoint if the number of types is too large to process.
|
89 |
This function also saves the results of the query to a csv file called query.csv.""",
|
90 |
"parameters": {
|
91 |
"type": "object",
|
92 |
"properties": {
|
93 |
+
"queries": {
|
94 |
"type": "string",
|
95 |
"description": "The GraphQL query to use in the search. Infer this from the user's message. It should be a question or a statement."
|
96 |
}
|
97 |
},
|
98 |
+
"required": ["queries"],
|
99 |
},
|
100 |
},
|
101 |
},
|