The skin is the largest organ in the human body. It also has several different layers, the epidermis (upper) and the dermis (lower). Skin cancer appears in three types of cells: squamous cells, basal cells, and melanocytes. Melanocytes make melanin and are found in the lower part of the epidermis. Melanin is the pigment that gives color to the skin. When the skin is exposed to sunlight, the melanocytes secrete more pigment and making the skin darker. Skin cancer is diagnosed by several methods such as clinical examination, skin tissue analysis, and dermatology. Early warning of skin cancer can help facilitate treatment and is a challenging task even for expert dermatologists in its early stages. Recent studies have found that deep learning and transfer learning are very useful for melanoma classification as well as medical diagnosis. This paper proposes an efficient method to classify melanoma using deep learning. The proposed model uses the ISIC dataset and can run on other databases. We screened the results for accuracy, sensitivity, and specificity. It will be helpful for dermatologists to classify the image as melanoma or benign, with ease, speed, and high accuracy, and does not require the high capabilities of the deep learning machine used