Spaces:
Runtime error
Runtime error
fixed
Browse files
pages/1_Textiles_Disentanglement.py
CHANGED
@@ -35,7 +35,6 @@ concept_vectors = pd.read_csv('./data/stored_vectors/scores_colors_hsv.csv')
|
|
35 |
concept_vectors['vector'] = [np.array([float(xx) for xx in x]) for x in concept_vectors['vector'].str.split(', ')]
|
36 |
concept_vectors['score'] = concept_vectors['score'].astype(float)
|
37 |
concept_vectors = concept_vectors.sort_values('score', ascending=False).reset_index()
|
38 |
-
print(concept_vectors[['vector', 'score']], concept_vectors.loc[0, 'vector'], concept_vectors.loc[0, 'vector'].shape)
|
39 |
|
40 |
with dnnlib.util.open_url('./data/textile_model_files/network-snapshot-005000.pkl') as f:
|
41 |
model = legacy.load_network_pkl(f)['G_ema'].to('cpu') # type: ignore
|
@@ -77,7 +76,7 @@ with input_col_1:
|
|
77 |
st.session_state.image_id = int(image_id)
|
78 |
|
79 |
with input_col_2:
|
80 |
-
with st.form('
|
81 |
|
82 |
st.write('**Choose color to vary**')
|
83 |
type_col = st.selectbox('Color:', tuple(COLORS_LIST))
|
@@ -88,7 +87,7 @@ with input_col_2:
|
|
88 |
color_lambda = chosen_color_lambda_input.number_input('Lambda:', min_value=0, step=1, value=7)
|
89 |
color_lambda_button = st.form_submit_button('Choose the defined lambda')
|
90 |
|
91 |
-
if
|
92 |
st.session_state.concept_ids = type_col
|
93 |
st.session_state.space_id = space_id
|
94 |
|
@@ -97,13 +96,13 @@ with input_col_3:
|
|
97 |
|
98 |
st.write('**Saturation variation**')
|
99 |
chosen_saturation_lambda_input = st.empty()
|
100 |
-
saturation_lambda = chosen_saturation_lambda_input.number_input('Lambda:', min_value=0, step=1)
|
101 |
-
saturation_lambda_button = st.form_submit_button('Choose the defined lambda')
|
102 |
|
103 |
st.write('**Value variation**')
|
104 |
chosen_value_lambda_input = st.empty()
|
105 |
-
value_lambda = chosen_value_lambda_input.number_input('Lambda:', min_value=0, step=1)
|
106 |
-
value_lambda_button = st.form_submit_button('Choose the defined lambda')
|
107 |
|
108 |
# with input_col_4:
|
109 |
# with st.form('Network specifics:'):
|
@@ -140,9 +139,9 @@ with header_col_1:
|
|
140 |
st.write(f'Original image')
|
141 |
|
142 |
with header_col_2:
|
143 |
-
color_separation_vector, performance_color = concept_vectors[concept_vectors['color'] == st.session_state.concept_ids].loc[0, ['vector', 'score']]
|
144 |
-
saturation_separation_vector, performance_saturation = concept_vectors[concept_vectors['color'] == 'Saturation'].loc[0, ['vector', 'score']]
|
145 |
-
value_separation_vector, performance_value = concept_vectors[concept_vectors['color'] == 'Value'].loc[0, ['vector', 'score']]
|
146 |
st.write(f'Change in {st.session_state.concept_ids} of {np.round(color_lambda, 2)}, in saturation of {np.round(saturation_lambda, 2)}, in value of {np.round(value_lambda, 2)}. - Performance color vector: {performance_color}, saturation vector: {performance_saturation}, value vector: {performance_value}')
|
147 |
|
148 |
# ---------------------------- DISPLAY COL 2 ROW 1 ------------------------------
|
|
|
35 |
concept_vectors['vector'] = [np.array([float(xx) for xx in x]) for x in concept_vectors['vector'].str.split(', ')]
|
36 |
concept_vectors['score'] = concept_vectors['score'].astype(float)
|
37 |
concept_vectors = concept_vectors.sort_values('score', ascending=False).reset_index()
|
|
|
38 |
|
39 |
with dnnlib.util.open_url('./data/textile_model_files/network-snapshot-005000.pkl') as f:
|
40 |
model = legacy.load_network_pkl(f)['G_ema'].to('cpu') # type: ignore
|
|
|
76 |
st.session_state.image_id = int(image_id)
|
77 |
|
78 |
with input_col_2:
|
79 |
+
with st.form('text_form_1'):
|
80 |
|
81 |
st.write('**Choose color to vary**')
|
82 |
type_col = st.selectbox('Color:', tuple(COLORS_LIST))
|
|
|
87 |
color_lambda = chosen_color_lambda_input.number_input('Lambda:', min_value=0, step=1, value=7)
|
88 |
color_lambda_button = st.form_submit_button('Choose the defined lambda')
|
89 |
|
90 |
+
if colors_button:
|
91 |
st.session_state.concept_ids = type_col
|
92 |
st.session_state.space_id = space_id
|
93 |
|
|
|
96 |
|
97 |
st.write('**Saturation variation**')
|
98 |
chosen_saturation_lambda_input = st.empty()
|
99 |
+
saturation_lambda = chosen_saturation_lambda_input.number_input('Lambda:', min_value=0, step=1, key=0)
|
100 |
+
saturation_lambda_button = st.form_submit_button('Choose the defined lambda for Saturation')
|
101 |
|
102 |
st.write('**Value variation**')
|
103 |
chosen_value_lambda_input = st.empty()
|
104 |
+
value_lambda = chosen_value_lambda_input.number_input('Lambda:', min_value=0, step=1, key=1)
|
105 |
+
value_lambda_button = st.form_submit_button('Choose the defined lambda for Value')
|
106 |
|
107 |
# with input_col_4:
|
108 |
# with st.form('Network specifics:'):
|
|
|
139 |
st.write(f'Original image')
|
140 |
|
141 |
with header_col_2:
|
142 |
+
color_separation_vector, performance_color = concept_vectors[concept_vectors['color'] == st.session_state.concept_ids].reset_index().loc[0, ['vector', 'score']]
|
143 |
+
saturation_separation_vector, performance_saturation = concept_vectors[concept_vectors['color'] == 'Saturation'].reset_index().loc[0, ['vector', 'score']]
|
144 |
+
value_separation_vector, performance_value = concept_vectors[concept_vectors['color'] == 'Value'].reset_index().loc[0, ['vector', 'score']]
|
145 |
st.write(f'Change in {st.session_state.concept_ids} of {np.round(color_lambda, 2)}, in saturation of {np.round(saturation_lambda, 2)}, in value of {np.round(value_lambda, 2)}. - Performance color vector: {performance_color}, saturation vector: {performance_saturation}, value vector: {performance_value}')
|
146 |
|
147 |
# ---------------------------- DISPLAY COL 2 ROW 1 ------------------------------
|