One of the tragedies of the modern age is the pervasiveness of the thought of Immanuel Kant. Kant was expert in posturing his philosophy as scientific, but it was not only unscientific, but actually anti-scientific; Kant doesn't believe we can acutally know things in themself! Stanley Jaki writes very well about Kant's scientific shortcomings; see, for example, the book cited below.
The proper way to look at science is as a subset of a broader philosophy of nature. The following quotation is from a book recommended me by Benedict Ashley.
According to the view proposed here it would be inaccurate to regard mobile [i.e., physical] being as open to study by two distinct sciences, one considering it at its general level and the other considering it in a more proper and precise way. Knowledge that begins with general considerations concerning an object naturally tends to become more precise; unless it does so, it remains intrinsically imperfect. The same science that studies the first principles or ultimate causes in a given order of reality tends, but its intrinsic nature and not by any external force, to complete itself by the study of proximate causes. We do not change sciences in moving from a general to a particular level, so long as we do not change objects. What we do is to change the perfection of our knowledge of the object, we are like the man who recognized the distant object first as only a thing and later came to the much more perfect knowledge of it as a squirrel. Particular scientific knowledge is more perfect since it is more specific, although, from another angle, it is less perfect since it is less certain. General scientific knowledge is more perfect in being more certain but less perfect since it is not proper and precise.
Without previous general science, we would not be able to design instruments and arrange experiments to learn more about what we already know in preinstrumental and pre-experimental ways.
Logically prior and presupposed to refined knowledge through experiment and measurement, our general science of nature is thus immune from the revolutions that have taken place in specialized knowledge. Our means of attaining this general science are no better and no worse than Aristotle’s. They are in fact no different from his. The means is human reason itself, unaided by specialized and mathematical techniques, proceeding only by a logical analysis of general experience which our next chapter will explain.
The world we live in is a unified whole. The object of modern science is the same as that of natural philosophy (nature); the only difference is the methodology through which it looks at nature. Natural philosophy uses the ordinary data of our senses. Modern science uses instruments and specialized techniques. But notice that these instruments and specialized techniques rely on ordinary sense data; so we need to understand the general and ordinary sense data before we can use science's specialized ways of knowing. In other words, natural philosophy is foundational to modern science. But the advantage of modern science is that it is more precise.
Vincent E. Smith, The General Science of Nature (Milwaukee: The Bruce Publishing Company, 1958), 38, 41.
Immanuel Kant, Universal Natural History and Theory of the Heavens (Allgemeine Naturgeschichte und Theorie des Himmels, 1775) trans. Stanley Jaki (Scottish Academic Press, 1981), 302pp. Paperback reprint, 1992.
Jaki remarks: "The first full translation of a classic which shows Kant's ineptness in science and his weird ideas of denizens on other planets."