Google Colab Code to test Word2Vec model to find cosine and OOV

gistfile1.txt

import gensim.downloader as api

# 1. Load pre-trained Google News Word2Vec model (~1.5 GB download)
# Note: For a faster download during testing, you can use "glove-wiki-gigaword-50"
print("Loading Word2Vec model...")
model = api.load("word2vec-google-news-300")
print("Model loaded successfully!\n")

# 2. List of test samples
test_words = [
    # --- User-Provided Samples ---
    "unhappines",     # Typo / missing subword "un"
    "tokenization",   # Rare/technical term absent in 2013 Google News
    "cat",            # Standard word (from "the cat")
    
    # --- 3 Additional Samples ---
    "quickkkly",      # Multi-letter typo
    "hyperparameter", # Specialized machine learning term
    "skibidi"         # Modern slang / new word
]

# 3. Test OOV status in Word2Vec
print(f"{'WORD / TOKEN':<18} | {'STATUS':<15} | {'SIMILAR / VECTOR ACTION'}")
print("-" * 65)

for word in test_words:
    if word in model.key_to_index:
        # Get most similar word to prove it has a vector
        similar = model.most_similar(word, topn=1)[0][0]
        print(f"{word:<18} | ✅ IN VOCAB    | Top Neighbor: '{similar}'")
    else:
        print(f"{word:<18} | ❌ OOV (KeyError) | Cannot process vector!")