jacob-mathison/doczyai-pipelines
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Activity

Merged in DEV-DS (pull request #25)

Browse Source 
sso added again

Approved-by: Pankaj Katariya
This commit is contained in:
Mayank Aamseek
2024-03-11 10:50:39 +00:00
committed by Pankaj Katariya
parent 5fdfb1cc4d 28c984f90a
commit 128fd06bc6
3 changed files with 11 additions and 20 deletions
Download Patch File Download Diff File Expand all files Collapse all files
streamlit/interface_1.py
+2 -3
View File
@@ -25,9 +25,8 @@ st.set_page_config(layout = "wide")
# add_vertical_space(15)
# # st.write("Doczy")
# util.setup_page(REDIRECT_URI)
# if st.session_state.user_info['mail'] in user_list:
if 'maamseek@aarete.com' in user_list:
util.setup_page(REDIRECT_URI)
if st.session_state.user_info['mail'] in user_list:
s3_client = boto3.client('s3',
region_name="us-east-1",
streamlit/interface_2.py
+5 -6
View File
@@ -33,17 +33,16 @@ with st.sidebar:
add_vertical_space(15)
# st.write("Doczy")
# util.setup_page(REDIRECT_URI)
# if st.session_state.user_info['mail'] in user_list:
if 'maamseek@aarete.com' in user_list:
util.setup_page(REDIRECT_URI)
if st.session_state.user_info['mail'] in user_list:
fields = pd.read_csv('contract_fields.csv', encoding='unicode_escape', skipinitialspace=True)
fields = fields[fields['PRIORITY'] == 'A']
fields = fields.drop_duplicates(subset='Field Name', keep="first").sort_values('Field Name')
fields['Interrogation Question?'].fillna(' ', inplace=True)
fields['Interrogation Question?'] = fields['Interrogation Question?'].fillna(' ')
field_prompt_mapping = dict(zip(fields['Field Name'], fields['Interrogation Question?']))
def file_selector(folder_path='RAW_DOCUMENTS'):
def file_selector(folder_path=SOURCE_DIRECTORY):
filenames = os.listdir(folder_path)
selected_filename = st.selectbox('Select a file', filenames, label_visibility = "collapsed")
# return os.path.join(folder_path, selected_filename)
@@ -69,7 +68,7 @@ if 'maamseek@aarete.com' in user_list:
llm_selected = st.selectbox('Langauge Model',('Llama 2 Chat 13B', 'Llama 2 Chat 70B', 'Titan Text Express'), label_visibility = "collapsed")
page_list = []
with open(os.path.join(SOURCE_DIRECTORY, file_name[:-4]+'.txt'), 'r') as infile:
with open(os.path.join(SOURCE_DIRECTORY, file_name), 'r') as infile:
text = infile.read()
page_count = text.count('Start of Page No. = ')
for page in range(page_count+1):
streamlit/interface_3.py
+4 -11
View File
@@ -36,9 +36,9 @@ with st.sidebar:
add_vertical_space(15)
# st.write("Doczy")
# util.setup_page(REDIRECT_URI)
# if st.session_state.user_info['mail'] in user_list:
if 'maamseek@aarete.com' in user_list:
util.setup_page(REDIRECT_URI)
if st.session_state.user_info['mail'] in user_list:
fields = pd.read_csv('contract_fields.csv', encoding='unicode_escape', skipinitialspace=True)
fields = fields[fields['PRIORITY'].isin(['A','C'])]
field_values = pd.read_csv('contract_field_values.csv', encoding='unicode_escape', skipinitialspace=True)
@@ -122,13 +122,6 @@ if 'maamseek@aarete.com' in user_list:
, '(Internal) Carveout ID': 'Contract ID', column_name+'_PG': 'Original Page Number'}, inplace=True)
field_values = field_values.drop_duplicates(subset='Contract Name', keep="first").sort_values('Contract Name')
# sample = dict(zip(sample.Filename, sample.Answer))
# actual_value_list = [sample.get(contract.rsplit('.',1)[0]+'.txt', ' ') for contract in contract_list]
# AWS_ACCESS_KEY_ID = os.getenv('AWS_ACCESS_KEY_ID')
# AWS_SECRET_ACCESS_KEY = os.getenv('AWS_SECRET_ACCESS_KEY')
# AWS_SESSION_TOKEN=os.getenv('AWS_SESSION_TOKEN')
# Setup bedrock
bedrock_runtime = boto3.client(
service_name="bedrock-runtime",
@@ -344,7 +337,7 @@ if 'maamseek@aarete.com' in user_list:
accuracy = 'NA'
history.loc[len(history.index)] = [field, str(contract_count), None, datetime.now().strftime("%Y-%m-%d %H:%M:%S"), accuracy, attempt]
df.to_csv("RESULTS\\"+field.replace("?","").replace("/","_")+'-'+llm_selected+'.csv', index=False)
# df.to_csv("RESULTS\\"+field.replace("?","").replace("/","_")+'-'+llm_selected+'.csv', index=False)
history.to_csv('history.csv', index=False)
# df_copy = df.set_index(df.columns[0]).copy()