full

history.csv

@@ -0,0 +1,17 @@
+review,sentiment
+its a superb movie,positive
+its a crap,negative
+"its somewhat okay
+",neutral
+The climax is wawa,neutral
+The climax is fantastic,positive
+the color of product is bullshit,negative
+the story is a scrap,negative
+its a  fantastic movie,positive
+the movie is bullshit,negative
+move is kind of okay,positive
+its an  above average movie,positive
+its okay,neutral
+its below average,negative
+its above average,neutral
+The product is entirely destructed\,negative

requirements.txt

@@ -1,3 +1,6 @@
-altair
-pandas
-streamlit
+transformers>=4.30.0
+datasets>=2.12.0
+pandas>=1.3.0
+scikit-learn>=1.0.0
+torch>=1.10.0
+streamlit>=1.20.0 

sentiment_model/config.json

@@ -0,0 +1,35 @@
+{
+  "_name_or_path": "distilbert-base-uncased",
+  "activation": "gelu",
+  "architectures": [
+    "DistilBertForSequenceClassification"
+  ],
+  "attention_dropout": 0.1,
+  "dim": 768,
+  "dropout": 0.1,
+  "hidden_dim": 3072,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2"
+  },
+  "initializer_range": 0.02,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2
+  },
+  "max_position_embeddings": 512,
+  "model_type": "distilbert",
+  "n_heads": 12,
+  "n_layers": 6,
+  "pad_token_id": 0,
+  "problem_type": "single_label_classification",
+  "qa_dropout": 0.1,
+  "seq_classif_dropout": 0.2,
+  "sinusoidal_pos_embds": false,
+  "tie_weights_": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.49.0",
+  "vocab_size": 30522
+}

sentiment_model/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9422fd567a5f50cc403e0a56a08831a6bb38c89235b836141a38f222801675ea
+size 267835644

sentiment_model/special_tokens_map.json

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

sentiment_model/tokenizer_config.json

@@ -0,0 +1,56 @@
+{
+  "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": false,
+  "cls_token": "[CLS]",
+  "do_lower_case": true,
+  "extra_special_tokens": {},
+  "mask_token": "[MASK]",
+  "model_max_length": 512,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "DistilBertTokenizer",
+  "unk_token": "[UNK]"
+}

streamlit_app.py

@@ -0,0 +1,130 @@
+import streamlit as st
+import os
+import torch
+from transformers import DistilBertTokenizerFast, DistilBertForSequenceClassification
+import pandas as pd
+
+# Set page config for a modern look
+title = "AI Sentiment Classifier"
+st.set_page_config(page_title=title, layout="centered")
+
+# Custom CSS for high-tech look
+st.markdown(
+    """
+    <style>
+    body {
+        background-color: #181A1B;
+    }
+    .main {
+        background-color: #23272A;
+        border-radius: 12px;
+        padding: 2rem 2rem 1.5rem 2rem;
+        box-shadow: 0 4px 32px 0 rgba(0,0,0,0.25);
+    }
+    .stTextArea textarea {
+        background-color: #23272A !important;
+        color: #F8F8F2 !important;
+        font-size: 1.1rem;
+        border-radius: 8px;
+    }
+    .stButton>button {
+        background: linear-gradient(90deg, #00C9FF 0%, #92FE9D 100%);
+        color: #181A1B;
+        font-weight: bold;
+        border-radius: 8px;
+        border: none;
+        font-size: 1.1rem;
+        padding: 0.5rem 1.5rem;
+    }
+    .sentiment-box {
+        background: #23272A;
+        border: 2px solid #00C9FF;
+        border-radius: 10px;
+        padding: 1.2rem;
+        margin-top: 1.5rem;
+        text-align: center;
+        font-size: 1.3rem;
+        color: #00C9FF;
+        font-weight: bold;
+        letter-spacing: 1px;
+        box-shadow: 0 2px 16px 0 rgba(0,201,255,0.10);
+    }
+    </style>
+    """,
+    unsafe_allow_html=True,
+)
+
+st.markdown(f"<h1 style='text-align:center; color:#00C9FF; margin-bottom:0.2em'>{title}</h1>", unsafe_allow_html=True)
+st.markdown("<h4 style='text-align:center; color:#F8F8F2; margin-top:0'>Classify the sentiment of your product or movie review instantly.</h4>", unsafe_allow_html=True)
+
+st.markdown("""
+<div class="main">
+""", unsafe_allow_html=True)
+
+review = st.text_area("Enter your review:", height=120, key="review_input")
+
+# Load model and tokenizer only once
+@st.cache_resource
+def load_model():
+    model_dir = './sentiment_model'
+    tokenizer = DistilBertTokenizerFast.from_pretrained(model_dir)
+    model = DistilBertForSequenceClassification.from_pretrained(model_dir)
+    model.eval()
+    return tokenizer, model
+
+tokenizer, model = load_model()
+
+sentiment = None
+show_result = False
+
+if st.button("Analyze Sentiment"):
+    if review.strip():
+        # Tokenize and predict
+        inputs = tokenizer([review], padding=True, truncation=True, max_length=128, return_tensors='pt')
+        with torch.no_grad():
+            outputs = model(input_ids=inputs['input_ids'], attention_mask=inputs['attention_mask'])
+            pred = torch.argmax(outputs.logits, dim=1).item()
+        sentiment_map = {0: "negative", 1: "neutral", 2: "positive"}
+        sentiment = sentiment_map.get(pred, "unknown")
+        color = {"positive": "#00FFB3", "neutral": "#FFD600", "negative": "#FF4B4B"}[sentiment]
+        st.markdown(f"<div class='sentiment-box' style='color:{color};border-color:{color}'>Sentiment: {sentiment.capitalize()}</div>", unsafe_allow_html=True)
+        show_result = True
+        # Save to local history file
+        history_file = "history.csv"
+        file_exists = os.path.isfile(history_file)
+        import csv
+        with open(history_file, mode="a", newline='', encoding="utf-8") as f:
+            writer = csv.writer(f)
+            if not file_exists:
+                writer.writerow(["review", "sentiment"])
+            writer.writerow([review, sentiment])
+    else:
+        st.warning("Please enter a review to analyze.")
+
+# Show history table
+st.markdown("<hr style='margin:2em 0 1em 0;border:1px solid #222;'>", unsafe_allow_html=True)
+st.subheader("Review History")
+# Clear History Button
+if st.button("Clear History 🗑️"):
+    history_file = "history.csv"
+    if os.path.isfile(history_file):
+        os.remove(history_file)
+    st.success("Review history cleared.")
+    st.experimental_rerun()
+
+history_file = "history.csv"
+if os.path.isfile(history_file):
+    try:
+        df = pd.read_csv(history_file)
+        if not df.empty:
+            st.dataframe(df[::-1].reset_index(drop=True), use_container_width=True)
+        else:
+            st.info("No review history yet.")
+    except Exception:
+        st.info("No review history yet.")
+else:
+    st.info("No review history yet.")
+
+st.markdown("""
+</div>
+""", unsafe_allow_html=True)