network_visualzation_dz9

def measures_scores(embs, cats):
    test_sizes = [0.99, 0.95, 0.9, 0.8, 0.7]
    scores_results = []

    for test_size in test_sizes:
      X_train, X_val, y_train, y_val = train_test_split(embs, cats, test_size=test_size)
      gb_clf = GradientBoostingClassifier()
      gb_clf.fit(X_train, y_train)
      score = f1_score(y_val, gb_clf.predict(X_val), average='micro')
      scores_results.append(score)

    return scores_results


def embeddings_score_evaluate(_graph):
    graph = _graph.copy()
    laplacian_emb, svd_emb, deep_walk_emb, walklets_emb = xy_embeddings(graph)

    results = []
    results.append(measures_scores(laplacian_emb, category_id))
    results.append(measures_scores(svd_emb, category_id))
    results.append(measures_scores(deep_walk_emb, category_id))
    results.append(measures_scores(walklets_emb, category_id))

    return results

scores = embeddings_score_evaluate(gcc_cora)

def embeddings_score(gcc_cora):
  return scores