Initial commit: Fine-tuning BERT on ABSA project

.gitattributes

@@ -0,0 +1 @@
+absa_webapp/fine_tuned_bert_absa/model.safetensors filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,4 @@
+# Fine-Tuning-Bert-on-ABSA
+In this project I have fine-tuned BERT on SemEval 2014 Task 4 dataset so it can work on ABSA and in the end I have deployed the model.
+
+I have attached the .ipynb file in this project and a saved model of this project so you can integrate it with streamlit or any other library to deploy it. I have also attached the streamlit source code file.

absa_webapp/app.py

@@ -0,0 +1,58 @@
+import streamlit as st
+from transformers import BertTokenizer, BertForSequenceClassification
+import torch
+
+# Load the fine-tuned model and tokenizer
+@st.cache_resource
+def load_model():
+    model = BertForSequenceClassification.from_pretrained('./fine_tuned_bert_absa')  # Path to your saved model
+    tokenizer = BertTokenizer.from_pretrained('./fine_tuned_bert_absa')
+    return model, tokenizer
+
+model, tokenizer = load_model()
+device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
+model.to(device)
+
+# Streamlit app title
+st.title("Aspect-Based Sentiment Analysis Web App")
+
+# Input fields for sentence and aspect
+sentence = st.text_area("Enter the review or sentence:")
+aspect = st.text_input("Enter the aspect to analyze (e.g., 'food', 'service'):")
+
+# Function to predict sentiment
+def predict_sentiment(sentence, aspect):
+    if not sentence or not aspect:
+        return "Please enter both a sentence and an aspect."
+    
+    # Preprocess the input
+    inputs = tokenizer.encode_plus(
+        aspect,
+        sentence,
+        add_special_tokens=True,
+        max_length=128,
+        padding='max_length',
+        truncation=True,
+        return_tensors='pt'
+    )
+
+    input_ids = inputs['input_ids'].to(device)
+    attention_mask = inputs['attention_mask'].to(device)
+
+    # Get predictions
+    model.eval()
+    with torch.no_grad():
+        outputs = model(input_ids, attention_mask=attention_mask)
+    
+    logits = outputs.logits
+    predicted_class = torch.argmax(logits, dim=1).item()
+
+    # Map class index to sentiment
+    label_map = {0: "Negative", 1: "Neutral", 2: "Positive"}
+    return label_map[predicted_class]
+
+# Button for prediction
+if st.button("Analyze Sentiment"):
+    result = predict_sentiment(sentence, aspect)
+    st.subheader(f"Aspect: {aspect}")
+    st.write(f"Predicted Sentiment: **{result}**")

absa_webapp/fine_tuned_bert_absa/config.json

@@ -0,0 +1,37 @@
+{
+  "_name_or_path": "bert-base-uncased",
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.46.2",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 30522
+}

absa_webapp/fine_tuned_bert_absa/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5ac521812c19b03803d45354d5dedc338466f6687a5d71fd762b2239796c3e24
+size 437961724

absa_webapp/fine_tuned_bert_absa/special_tokens_map.json

@@ -0,0 +1,7 @@
+{
+  "cls_token": "[CLS]",
+  "mask_token": "[MASK]",
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "unk_token": "[UNK]"
+}

absa_webapp/fine_tuned_bert_absa/tokenizer_config.json

@@ -0,0 +1,57 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "101": {
+      "content": "[CLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "102": {
+      "content": "[SEP]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "103": {
+      "content": "[MASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "[CLS]",
+  "do_basic_tokenize": true,
+  "do_lower_case": true,
+  "mask_token": "[MASK]",
+  "model_max_length": 512,
+  "never_split": null,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "unk_token": "[UNK]"
+}

absa_webapp/requirements.txt.txt

@@ -0,0 +1,3 @@
+streamlit
+transformers
+torch