{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/frechet-chemnet-distance-a-metric-for","title":"Fréchet ChemNet Distance: A metric for generative models for molecules in drug discovery","arxiv_id":"1803.09518","date":"2018-03-26","proceeding":null,"authors":["Kristina Preuer","Philipp Renz","Thomas Unterthiner","Sepp Hochreiter","Günter Klambauer"],"abstract":"The new wave of successful generative models in machine learning has\nincreased the interest in deep learning driven de novo drug design. However,\nassessing the performance of such generative models is notoriously difficult.\nMetrics that are typically used to assess the performance of such generative\nmodels are the percentage of chemically valid molecules or the similarity to\nreal molecules in terms of particular descriptors, such as the partition\ncoefficient (logP) or druglikeness. However, method comparison is difficult\nbecause of the inconsistent use of evaluation metrics, the necessity for\nmultiple metrics, and the fact that some of these measures can easily be\ntricked by simple rule-based systems. We propose a novel distance measure\nbetween two sets of molecules, called Fr\\'echet ChemNet distance (FCD), that\ncan be used as an evaluation metric for generative models. The FCD is similar\nto a recently established performance metric for comparing image generation\nmethods, the Fr\\'echet Inception Distance (FID). Whereas the FID uses one of\nthe hidden layers of InceptionNet, the FCD utilizes the penultimate layer of a\ndeep neural network called ChemNet, which was trained to predict drug\nactivities. Thus, the FCD metric takes into account chemically and biologically\nrelevant information about molecules, and also measures the diversity of the\nset via the distribution of generated molecules. The FCD's advantage over\nprevious metrics is that it can detect if generated molecules are a) diverse\nand have similar b) chemical and c) biological properties as real molecules. We\nfurther provide an easy-to-use implementation that only requires the SMILES\nrepresentation of the generated molecules as input to calculate the FCD.\nImplementations are available at: https://www.github.com/bioinf-jku/FCD","url_abs":"http://arxiv.org/abs/1803.09518v3","url_pdf":"http://arxiv.org/pdf/1803.09518v3.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"frechet-chemnet-distance-a-metric-for","repo_url":"https://github.com/bioinf-jku/FCD","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"LGPL-3.0"}},{"paper_slug":"frechet-chemnet-distance-a-metric-for","repo_url":"https://github.com/insilicomedicine/fcd_torch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"drug-design","task_name":"Drug Design"},{"task_slug":"drug-discovery","task_name":"Drug Discovery"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1803.09518","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1803.09518"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/insilicomedicine/fcd_torch","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/bioinf-jku/FCD","reach":{"status":"ok","spdx":"LGPL-3.0"}}],"summary":{"unverified":5},"by_repo_kind":{"listed":{"samples":5,"ran":0,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"08c1bbf222daf46d","entry":"build_masked_loss","repo":"insilicomedicine/fcd_torch","repo_kind":"listed","path":"fcd_torch/keras2pytorch.py","file_url":"https://github.com/insilicomedicine/fcd_torch/blob/HEAD/fcd_torch/keras2pytorch.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"08c1bbf222daf46d"}},{"code_sha256_prefix":"dd804e3188fac887","entry":"calculate_frechet_distance","repo":"insilicomedicine/fcd_torch","repo_kind":"listed","path":"fcd_torch/utils.py","file_url":"https://github.com/insilicomedicine/fcd_torch/blob/HEAD/fcd_torch/utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"dd804e3188fac887"}},{"code_sha256_prefix":"0840fcb4bcdcde19","entry":"get_one_hot","repo":"insilicomedicine/fcd_torch","repo_kind":"listed","path":"fcd_torch/utils.py","file_url":"https://github.com/insilicomedicine/fcd_torch/blob/HEAD/fcd_torch/utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"0840fcb4bcdcde19"}},{"code_sha256_prefix":"020baa2a70cf8d4a","entry":"load_imported_model","repo":"insilicomedicine/fcd_torch","repo_kind":"listed","path":"fcd_torch/utils.py","file_url":"https://github.com/insilicomedicine/fcd_torch/blob/HEAD/fcd_torch/utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"020baa2a70cf8d4a"}},{"code_sha256_prefix":"27b3c0dc7c5ee22c","entry":"masked_accuracy","repo":"insilicomedicine/fcd_torch","repo_kind":"listed","path":"fcd_torch/keras2pytorch.py","file_url":"https://github.com/insilicomedicine/fcd_torch/blob/HEAD/fcd_torch/keras2pytorch.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"27b3c0dc7c5ee22c"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}