File size: 6,937 Bytes
3f46b34
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
# type: ignore -- ignores linting import issues when using multiple virtual environments
import streamlit.components.v1 as components
import streamlit as st
import pandas as pd
import logging
from deeploy import Client
from shap import TreeExplainer


# reset Plotly theme after streamlit import
import plotly.io as pio


pio.templates.default = "plotly"

logging.basicConfig(level=logging.INFO)

st.set_page_config(layout="wide")

st.title("Your title")

st.markdown(
    """
    <style>
        section[data-testid="stSidebar"] {
            width: 300px !important;
        }
    </style>
    """,
    unsafe_allow_html=True,
) # Set the side bar width to fit the Deeploy logo


def get_model_url():
    """Function to get Deeploy model URL and split it into workspace and deployment ID."""
    model_url = st.text_area(
        "Model URL (without the /explain endpoint, default is the demo deployment)",
        "https://api.app.deeploy.ml/workspaces/708b5808-27af-461a-8ee5-80add68384c7/deployments/9155091a-0abb-45b3-8b3b-24ac33fa556b/",
        height=125,
    )
    elems = model_url.split("/")
    try:
        workspace_id = elems[4]
        deployment_id = elems[6]
    except IndexError:
        workspace_id = ""
        deployment_id = ""
    return model_url, workspace_id, deployment_id


def ChangeButtonColour(widget_label, font_color, background_color="transparent"):
    """Function to change the color of a button (after it is defined)."""
    htmlstr = f"""
        <script>
            var elements = window.parent.document.querySelectorAll('button');
            for (var i = 0; i < elements.length; ++i) {{ 
                if (elements[i].innerText == '{widget_label}') {{ 
                    elements[i].style.color ='{font_color}';
                    elements[i].style.background = '{background_color}'
                }}
            }}
        </script>
        """
    components.html(f"{htmlstr}", height=0, width=0)

def predict():
    with st.spinner("Loading prediction and explanation..."):
        try:
            # Call the explain endpoint as it also includes the prediction
            exp = client.predict(
                request_body=request_body, deployment_id=deployment_id
            )
        except Exception as e:
            logging.error(e)
            st.error(
                "Failed to get prediction."
                + "Check whether you are using the right model URL and token for predictions. "
                + "Contact Deeploy if the problem persists."
            )
            return
        st.session_state.exp = exp
    st.session_state.evaluation_submitted = False    
    hide_expander()

def hide_expander():
    st.session_state.expander_toggle = False

def show_expander():
    st.session_state.expander_toggle = True

def submit_and_clear(evaluation: str):
    if evaluation == "yes":
        st.session_state.evaluation_input["result"] = 0 # Agree with the prediction
    else:
        desired_output = not predictions[0]
        st.session_state.evaluation_input["result"] = 1
        st.session_state.evaluation_input["value"] = {"predictions": [desired_output]}
    try:
        # Call the explain endpoint as it also includes the prediction
        client.evaluate(
            deployment_id, request_log_id, prediction_log_id, st.session_state.evaluation_input
        )
        st.session_state.evaluation_submitted = True
        st.session_state.exp = None
        show_expander()
    except Exception as e:
        logging.error(e)
        st.error(
            "Failed to submit feedback."
            + "Check whether you are using the right model URL and token for evaluations. "
            + "Contact Deeploy if the problem persists."
        )


# Define defaults for the session state
if "expander_toggle" not in st.session_state:
    st.session_state.expander_toggle = True

if "exp" not in st.session_state:
    st.session_state.exp = None

if "evaluation_submitted" not in st.session_state:
    st.session_state.evaluation_submitted = False

# Define sidebar for configuration of Deeploy connection
with st.sidebar:
    st.image("deeploy_logo_wide.png", width=250)

    # Ask for model URL and token
    host = st.text_input("Host (Changing is optional)", "app.deeploy.ml")
    model_url, workspace_id, deployment_id = get_model_url()
    deployment_token = st.text_input("Deeploy Model Token", "my-secret-token")
    if deployment_token == "my-secret-token":
        st.warning("Please enter Deeploy API token.")

    # In case you need to debug the workspace and deployment ID:
    # st.write("Values below are for debug only:")
    # st.write("Workspace ID: ", workspace_id)
    # st.write("Deployment ID: ", deployment_id)

    client_options = {
        "host": host,
        "deployment_token": deployment_token,
        "workspace_id": workspace_id,
    }
    client = Client(**client_options)

# For debugging the session state you can uncomment the following lines:
# with st.expander("Debug session state", expanded=False):
#     st.write(st.session_state)

# Input (for IRIS dataset)
with st.expander("Input values for prediction", expanded=st.session_state.expander_toggle):
    st.write("Please input the values for the model.")
    col1, col2 = st.columns(2)
    with col1:
        sep_len = st.number_input("Sepal length", value=1.0, step=0.1, key="Sepal length")
        sep_wid = st.number_input("Sepal width", value=1.0, step=0.1, key="Sepal width")
    with col2:
        pet_len = st.number_input("Petal length", value=1.0, step=0.1, key="Petal length")
        pet_wid = st.number_input("Petal width", value=1.0, step=0.1, key="Petal width")

request_body = {
    "instances": [
        [
            sep_len,
            sep_wid,
            pet_len,
            pet_wid,
        ],
    ]
}

# Predict and explain
predict_button = st.button("Predict", on_click=predict)
if st.session_state.exp is not None:
    st.write(st.session_state.exp)
    # predictions = st.session_state.exp["predictions"]
    # request_log_id = exp["requestLogId"]
    # prediction_log_id = exp["predictionLogIds"][0]

    # # exp_df = pd.DataFrame(
    # #     [exp["explanations"][0]["shap_values"]], columns=exp["featureLabels"]
    # # )
    # st.write("Predictions:", predictions)

    # # Evaluation
    # if st.session_state.evaluation_submitted is False:
    #     evaluation = st.radio("Do you agree with the prediction?", ("yes", "no"))
    #     if evaluation == "no":
    #         desired_output = # TODO
    #         st.session_state.evaluation_input = {
    #             "result": 1,
    #             "value": {"predictions": [desired_output]},
    #         }
    #     else:
    #         st.session_state.evaluation_input = {"result": 0}

    #     submit_button = st.button("Submit evaluation", on_click=submit_and_clear, args=(evaluation,))
    # else:
    #     st.success("Evaluation submitted successfully.")