Update Model/MultimodelNER/VLSP2016/MNER_2016.py

Model/MultimodelNER/VLSP2016/MNER_2016.py

@@ -1,106 +1,106 @@
-import streamlit as st
-from spacy import displacy
-from Model.NER.VLSP2021.Predict_Ner import ViTagger
-import re
-from thunghiemxuly import save_uploaded_image,convert_text_to_txt,add_string_to_txt
-
-import os
-from transformers import AutoTokenizer, BertConfig
-from Model.MultimodelNER.VLSP2016.train_umt_2016 import load_model,predict
-from Model.MultimodelNER.Ner_processing import format_predictions,process_predictions,combine_entities,remove_B_prefix,combine_i_tags
-
-from Model.MultimodelNER.predict import get_test_examples_predict
-from Model.MultimodelNER import resnet as resnet
-from Model.MultimodelNER.resnet_utils import myResnet
-import torch
-import numpy as np
-from Model.MultimodelNER.VLSP2016.dataset_roberta import MNERProcessor_2016
-
-
-CONFIG_NAME = 'bert_config.json'
-WEIGHTS_NAME = 'pytorch_model.bin'
-device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
-
-
-net = getattr(resnet, 'resnet152')()
-net.load_state_dict(torch.load(os.path.join('E:/demo_datn/pythonProject1/Model/Resnet/', 'resnet152.pth')))
-encoder = myResnet(net, True, device)
-def process_text(text):
-    # Loại bỏ dấu cách thừa và dấu cách ở đầu và cuối văn bản
-    processed_text = re.sub(r'\s+', ' ', text.strip())
-    return processed_text
-
-
-
-def show_mner_2016():
-    multimodal_text = st.text_area("Enter your text for MNER:", height=300)
-    multimodal_text = process_text(multimodal_text)  # Xử lý văn bản
-    image = st.file_uploader("Upload an image (only jpg):", type=["jpg"])
-    if st.button("Process Multimodal NER"):
-            save_image = 'E:/demo_datn/pythonProject1/Model/MultimodelNER/VLSP2016/Image'
-            save_txt = 'E:/demo_datn/pythonProject1/Model/MultimodelNER/VLSP2016/Filetxt/test.txt'
-            image_name = image.name
-            save_uploaded_image(image, save_image)
-            convert_text_to_txt(multimodal_text, save_txt)
-            add_string_to_txt(image_name, save_txt)
-            st.image(image, caption="Uploaded Image", use_column_width=True)
-
-            bert_model='vinai/phobert-base-v2'
-            output_dir='E:/demo_datn/pythonProject1/Model/MultimodelNER/VLSP2016/best_model'
-            output_model_file = os.path.join(output_dir, WEIGHTS_NAME)
-            output_encoder_file = os.path.join(output_dir, "pytorch_encoder.bin")
-            processor = MNERProcessor_2016()
-            label_list = processor.get_labels()
-            auxlabel_list = processor.get_auxlabels()
-            num_labels = len(label_list) + 1
-            auxnum_labels = len(auxlabel_list) + 1
-            trans_matrix = np.zeros((auxnum_labels, num_labels), dtype=float)
-            trans_matrix[0, 0] = 1  # pad to pad
-            trans_matrix[1, 1] = 1  # O to O
-            trans_matrix[2, 2] = 0.25  # B to B-MISC
-            trans_matrix[2, 4] = 0.25  # B to B-PER
-            trans_matrix[2, 6] = 0.25  # B to B-ORG
-            trans_matrix[2, 8] = 0.25  # B to B-LOC
-            trans_matrix[3, 3] = 0.25  # I to I-MISC
-            trans_matrix[3, 5] = 0.25  # I to I-PER
-            trans_matrix[3, 7] = 0.25  # I to I-ORG
-            trans_matrix[3, 9] = 0.25  # I to I-LOC
-            trans_matrix[4, 10] = 1  # X to X
-            trans_matrix[5, 11] = 1  # [CLS] to [CLS]
-            trans_matrix[6, 12] = 1
-            tokenizer = AutoTokenizer.from_pretrained(bert_model, do_lower_case=False)
-            model_umt, encoder_umt = load_model(output_model_file, output_encoder_file, encoder,num_labels,auxnum_labels)
-            eval_examples = get_test_examples_predict('E:/demo_datn/pythonProject1/Model/MultimodelNER/VLSP2016/Filetxt/')
-
-            y_pred, a = predict(model_umt, encoder_umt, eval_examples, tokenizer, device,save_image,trans_matrix)
-            formatted_output = format_predictions(a, y_pred[0])
-            final = process_predictions(formatted_output)
-            final2 = combine_entities(final)
-            final3 = remove_B_prefix(final2)
-            final4 = combine_i_tags(final3)
-            words_and_labels = final4
-            # Tạo danh sách từ
-            words = [word for word, _ in words_and_labels]
-            # Tạo danh sách thực thể và nhãn cho mỗi từ, loại bỏ nhãn 'O'
-            entities = [{'start': sum(len(word) + 1 for word, _ in words_and_labels[:i]),
-                         'end': sum(len(word) + 1 for word, _ in words_and_labels[:i + 1]), 'label': label} for
-                        i, (word, label)
-                        in enumerate(words_and_labels) if label != 'O']
-            # print(entities)
-
-            # Render the visualization without color for 'O' labels
-            html = displacy.render(
-                {"text": " ".join(words), "ents": entities, "title": None},
-                style="ent",
-                manual=True,
-                options={"colors": {"MISC": "#806699",
-                                    "ORG": "#ff6666",
-                                    "LOC": "#66cc66",
-                                    "PER": "#bf80ff",
-                                    "O": None}}
-            )
-            # print(html)
-            st.markdown(html, unsafe_allow_html=True)
-
-
+import streamlit as st
+from spacy import displacy
+from Model.NER.VLSP2021.Predict_Ner import ViTagger
+import re
+from thunghiemxuly import save_uploaded_image,convert_text_to_txt,add_string_to_txt
+
+import os
+from transformers import AutoTokenizer, BertConfig
+from Model.MultimodelNER.VLSP2016.train_umt_2016 import load_model,predict
+from Model.MultimodelNER.Ner_processing import format_predictions,process_predictions,combine_entities,remove_B_prefix,combine_i_tags
+
+from Model.MultimodelNER.predict import get_test_examples_predict
+from Model.MultimodelNER import resnet as resnet
+from Model.MultimodelNER.resnet_utils import myResnet
+import torch
+import numpy as np
+from Model.MultimodelNER.VLSP2016.dataset_roberta import MNERProcessor_2016
+
+
+CONFIG_NAME = 'bert_config.json'
+WEIGHTS_NAME = 'pytorch_model.bin'
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+
+
+net = getattr(resnet, 'resnet152')()
+net.load_state_dict(torch.load(os.path.join('/Model/Resnet/', 'resnet152.pth')))
+encoder = myResnet(net, True, device)
+def process_text(text):
+    # Loại bỏ dấu cách thừa và dấu cách ở đầu và cuối văn bản
+    processed_text = re.sub(r'\s+', ' ', text.strip())
+    return processed_text
+
+
+
+def show_mner_2016():
+    multimodal_text = st.text_area("Enter your text for MNER:", height=300)
+    multimodal_text = process_text(multimodal_text)  # Xử lý văn bản
+    image = st.file_uploader("Upload an image (only jpg):", type=["jpg"])
+    if st.button("Process Multimodal NER"):
+            save_image = '/Model/MultimodelNER/VLSP2016/Image'
+            save_txt = '/Model/MultimodelNER/VLSP2016/Filetxt/test.txt'
+            image_name = image.name
+            save_uploaded_image(image, save_image)
+            convert_text_to_txt(multimodal_text, save_txt)
+            add_string_to_txt(image_name, save_txt)
+            st.image(image, caption="Uploaded Image", use_column_width=True)
+
+            bert_model='vinai/phobert-base-v2'
+            output_dir='/Model/MultimodelNER/VLSP2016/best_model'
+            output_model_file = os.path.join(output_dir, WEIGHTS_NAME)
+            output_encoder_file = os.path.join(output_dir, "pytorch_encoder.bin")
+            processor = MNERProcessor_2016()
+            label_list = processor.get_labels()
+            auxlabel_list = processor.get_auxlabels()
+            num_labels = len(label_list) + 1
+            auxnum_labels = len(auxlabel_list) + 1
+            trans_matrix = np.zeros((auxnum_labels, num_labels), dtype=float)
+            trans_matrix[0, 0] = 1  # pad to pad
+            trans_matrix[1, 1] = 1  # O to O
+            trans_matrix[2, 2] = 0.25  # B to B-MISC
+            trans_matrix[2, 4] = 0.25  # B to B-PER
+            trans_matrix[2, 6] = 0.25  # B to B-ORG
+            trans_matrix[2, 8] = 0.25  # B to B-LOC
+            trans_matrix[3, 3] = 0.25  # I to I-MISC
+            trans_matrix[3, 5] = 0.25  # I to I-PER
+            trans_matrix[3, 7] = 0.25  # I to I-ORG
+            trans_matrix[3, 9] = 0.25  # I to I-LOC
+            trans_matrix[4, 10] = 1  # X to X
+            trans_matrix[5, 11] = 1  # [CLS] to [CLS]
+            trans_matrix[6, 12] = 1
+            tokenizer = AutoTokenizer.from_pretrained(bert_model, do_lower_case=False)
+            model_umt, encoder_umt = load_model(output_model_file, output_encoder_file, encoder,num_labels,auxnum_labels)
+            eval_examples = get_test_examples_predict('/Model/MultimodelNER/VLSP2016/Filetxt/')
+
+            y_pred, a = predict(model_umt, encoder_umt, eval_examples, tokenizer, device,save_image,trans_matrix)
+            formatted_output = format_predictions(a, y_pred[0])
+            final = process_predictions(formatted_output)
+            final2 = combine_entities(final)
+            final3 = remove_B_prefix(final2)
+            final4 = combine_i_tags(final3)
+            words_and_labels = final4
+            # Tạo danh sách từ
+            words = [word for word, _ in words_and_labels]
+            # Tạo danh sách thực thể và nhãn cho mỗi từ, loại bỏ nhãn 'O'
+            entities = [{'start': sum(len(word) + 1 for word, _ in words_and_labels[:i]),
+                         'end': sum(len(word) + 1 for word, _ in words_and_labels[:i + 1]), 'label': label} for
+                        i, (word, label)
+                        in enumerate(words_and_labels) if label != 'O']
+            # print(entities)
+
+            # Render the visualization without color for 'O' labels
+            html = displacy.render(
+                {"text": " ".join(words), "ents": entities, "title": None},
+                style="ent",
+                manual=True,
+                options={"colors": {"MISC": "#806699",
+                                    "ORG": "#ff6666",
+                                    "LOC": "#66cc66",
+                                    "PER": "#bf80ff",
+                                    "O": None}}
+            )
+            # print(html)
+            st.markdown(html, unsafe_allow_html=True)
+
+
 ###Ví dụ 1 : Một trận hỗn chiến đã xảy ra tại trận đấu khúc côn cầu giữa  Penguins và Islanders ở Mỹ (image:penguin)