<\/span><\/p>\n# initialize artifacts\ntrain_data_artifact = comet_ml.Artifact(name=\"training_data\",\n artifact_type=\"dataset\",\n metadata={\"current_data\": datetime.datetime.utcnow().isoformat()})\ntrain_labels_artifact = comet_ml.Artifact(name=\"training_labels\",\n artifact_type=\"dataset\",\n metadata={\"current_data\": datetime.datetime.utcnow().isoformat()})\n\nvalidation_data_artifact = comet_ml.Artifact(name=\"validation_data\",\n artifact_type=\"dataset\",\n metadata={\"current_date\": datetime.datetime.utcnow().isoformat()})\n\nvalidation_labels_artifact = comet_ml.Artifact(name=\"validation_labels\",\n artifact_type=\"dataset\",\n metadata={\"current_date\": datetime.datetime.utcnow().isoformat()})<\/pre>\nNext, we start a for loop. This is done so we can generate five different neural network architectures, thus, conducting five experiments. Notice that the beginning of each loop initialized a new Experiment()<\/code> and Artifact()<\/code>instance. We do this to let Comet know we are starting a new experiment and would like to save a new model artifact \u2014 if we didn\u2019t do this it would update old versions. <\/span><\/p>\nfor i in range(5):\n experiment = comet_ml.Experiment(\n api_key=\"KGsg6IUGc5FpgH2lp3uF2gMbv\",\n project_name=\"testing-comet-experiments\",\n auto_param_logging=True,\n auto_histogram_weight_logging=True,\n auto_histogram_gradient_logging=True,\n auto_histogram_activation_logging=True,\n auto_histogram_epoch_rate=True\n )\n\n model_artifact = comet_ml.Artifact(name=f\"model_{i}\",\n artifact_type=\"model\",\n metadata={\"current_date\": datetime.datetime.utcnow().isoformat()})<\/pre>\nAfter that, we begin our training and validation experiments respectively. To add an artifact to an Artifact instance, we must pass either a file path or a file-like asset. This is why we saved our model after training using model.save()<\/code>, before adding it to the Artifact instance.<\/p>\nA similar procedure was followed for the train data\/labels and validation data\/labels \u2014 each array was saved with np.savetxt()<\/code>. But since each model is seeing the same data (train and validation), we set a condition to only log it once.<\/p>\n # model training\n with experiment.train():\n seed = np.random.randint(low=0, high=1000, dtype=int)\n\n np.random.seed(seed)\n tf.random.set_seed(seed)\n random.seed(seed)\n\n experiment.log_parameter(\"random_seed\", seed)\n\n model.fit(X_train, y_train, epochs=50, verbose=0)\n model.save(f\"my_model_{i}.h5\")\n\n # adding the artifact\n model_artifact.add(f\"my_model_{i}.h5\")\n experiment.log_artifact(model_artifact)\n\n if i==0:\n np.savetxt(\"training_data.txt\", X_train)\n np.savetxt(\"training_labels.txt\", y_train)\n np.savetxt(\"validation_data.txt\", X_val)\n np.savetxt(\"validation_labels.txt\", y_val)\n\n # adding data\n train_data_artifact.add(\"training_data.txt\")\n train_labels_artifact.add(\"training_labels.txt\")\n validation_data_artifact.add(\"validation_data.txt\")\n validation_labels_artifact.add(\"validation_labels.txt\")\n\n # logging each artifact\n experiment.log_artifact(train_data_artifact)\n experiment.log_artifact(train_labels_artifact)\n experiment.log_artifact(validation_data_artifact)\n experiment.log_artifact(validation_labels_artifact)<\/pre>\nThat\u2019s all that\u2019s happening.<\/p>\n
Retrieving a Logged Artifact<\/h1>\n In our last run, we stored several different Artifacts. At some point, we may be required to retrieve them from our workspace. This action is done using the get_artifact()<\/code> method on our Experiment()<\/code> instance. The method takes the unique Artifact name and workspace.<\/p>\nNote<\/strong>: you may also retrieve a specific artifact version by setting the version_or_alias attribute to the appropriate version. Leaving it blank will retrieve the latest version.<\/p><\/blockquote>\nexperiment = comet_ml.Experiment(api_key=\"KGsg6IUGc5FpgH2lp3uF2gMbv\")\n\nartifact_model_3 = experiment.get_artifact(artifact_name=\"model_3\",\n workspace=\"kurtispykes\")<\/pre>\nTo see all of the logged assets for an artifact version, we can use the assets<\/code>attribute of an Experiment()<\/code> instance as seen below. <\/span><\/p>\nfor asset in artifact_model_3.assets:\n print(f\"Logical path: {asset.logical_path}\\n\\\n Size: {asset.size}\\n\\\n Metadata: {asset.metadata}\\n\\\n Type: {asset.asset_type}\\n\\\n Artifact ID: {asset.artifact_id}\\n\\\n ID: {asset.id}\\n\\\n Version: {asset.artifact_version_id}\")<\/pre>\nAnd this will output the following:<\/p>\n
Logical <\/strong>path: my_model_3.h5\nSize<\/strong>: 33384\nMetadata<\/strong>: None\nType<\/strong>: unknown\nArtifact <\/strong>ID: 0c433884-b586-47a5-bf21-c841c8dc9733\nID<\/strong>: c27a83b616284beeafc91c69c1952df2\nVersion<\/strong>: 5fbcb375-bf52-47f9-a9f1-9e87fed09a25<\/span><\/pre>\nUpdating a Logged Artifact<\/h1>\n Now, let\u2019s say you’ve got some new data and you want to add it to the training data Artifact, how do we do that? Simple. We retrieve an existing Artifact, add the new file, bump the version, then log it \u2014 see the code below.<\/p>\n
# generating new instances\nX, y = make_circles(\n n_samples=200, noise=0.05, random_state=42\n)\n\n# retrieve old training data\nexperiment = comet_ml.Experiment(api_key=\"KGsg6IUGc5FpgH2lp3uF2gMbv\")\ntrain_data_artifact = experiment.get_artifact(artifact_name=\"training_data\",\n workspace=\"kurtispykes\")\ntrain_labels_artifact = experiment.get_artifact(artifact_name=\"training_labels\",\n workspace=\"kurtispykes\")\n\n# download non-remote assets to local disk\nlocal_train_data_artifact = train_data_artifact.download(\"new_data\")\nlocal_train_labels_artifact = train_labels_artifact.download(\"new_labels\")\n\n# loading the data from a txt file\ntrain_data = np.genfromtxt(\"new_data\/training_data.txt\")\ntrain_labels = np.genfromtxt(\"new_labels\/training_labels.txt\")\n\n# joining the new data to old data\ntrain_data_updated = np.concatenate((X_train, X), axis=0)\ntrain_labels_updated = np.concatenate((y_train, y), axis=0)\n\n# saving the updated data locally\nnp.savetxt(\"updated_training_data.txt\", X_train)\nnp.savetxt(\"updated_training_labels.txt\", y_train)\n\n# logging the updated data artifacts\nlocal_train_data_artifact.add(\"updated_training_data.txt\")\nlocal_train_labels_artifact.add(\"updated_training_labels.txt\")\n\n# bumping the version\nlocal_train_data_artifact.version = train_data_artifact.version.next_minor()\nlocal_train_labels_artifact.version = train_labels_artifact.version.next_minor()\n\n# logging the artifacts\nexperiment.log_artifact(local_train_data_artifact)\nexperiment.log_artifact(local_train_labels_artifact)<\/pre>\nWhen we visit our workspace, we will see there has been a new version added.<\/p>\n\n<\/picture><\/div>\nViewing the artifact we just added<\/strong><\/figcaption>\n<\/figure>\nThat\u2019s all folks!<\/p>\n
Thanks for reading.<\/em><\/p>\n<\/div>\n<\/div>\n<\/div>\n","protected":false},"excerpt":{"rendered":"Photo by Jeremy Bezanger on Unsplash All machine learning projects involve searching for the right model and features. As there is no way to know ahead of time what combination will best tackle a problem at hand, several experiments are conducted to try and find a solution worthy of being taken into production. At various […]<\/p>\n","protected":false},"author":8,"featured_media":0,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"customer_name":"","customer_description":"","customer_industry":"","customer_technologies":"","customer_logo":"","footnotes":""},"categories":[9,7],"tags":[],"coauthors":[138],"class_list":["post-8070","post","type-post","status-publish","format-standard","hentry","category-product","category-tutorials"],"yoast_head":"\n
Tracking Artifacts With Comet ML - Comet<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/www.comet.com\/site\/blog\/tracking-artifacts-with-comet-ml\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Tracking Artifacts With Comet ML\" \/>\n<meta property=\"og:description\" content=\"Photo by Jeremy Bezanger on Unsplash All machine learning projects involve searching for the right model and features. As there is no way to know ahead of time what combination will best tackle a problem at hand, several experiments are conducted to try and find a solution worthy of being taken into production. At various […]\" \/>\n<meta property=\"og:url\" content=\"https:\/\/www.comet.com\/site\/blog\/tracking-artifacts-with-comet-ml\" \/>\n<meta property=\"og:site_name\" content=\"Comet\" \/>\n<meta property=\"article:publisher\" content=\"https:\/\/www.facebook.com\/cometdotml\" \/>\n<meta property=\"article:published_time\" content=\"2023-11-02T17:24:56+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2025-04-24T17:04:56+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/miro.medium.com\/v2\/resize:fit:700\/0*QOhDWF6sla2bkUj4\" \/>\n<meta name=\"author\" content=\"Kurtis Pykes\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:creator\" content=\"@Cometml\" \/>\n<meta name=\"twitter:site\" content=\"@Cometml\" \/>\n<meta name=\"twitter:label1\" content=\"Written by\" \/>\n\t<meta name=\"twitter:data1\" content=\"Kurtis Pykes\" \/>\n\t<meta name=\"twitter:label2\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data2\" content=\"6 minutes\" \/>\n<!-- \/ Yoast SEO Premium plugin. -->","yoast_head_json":{"title":"Tracking Artifacts With Comet ML - Comet","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/www.comet.com\/site\/blog\/tracking-artifacts-with-comet-ml","og_locale":"en_US","og_type":"article","og_title":"Tracking Artifacts With Comet ML","og_description":"Photo by Jeremy Bezanger on Unsplash All machine learning projects involve searching for the right model and features. As there is no way to know ahead of time what combination will best tackle a problem at hand, several experiments are conducted to try and find a solution worthy of being taken into production. At various […]","og_url":"https:\/\/www.comet.com\/site\/blog\/tracking-artifacts-with-comet-ml","og_site_name":"Comet","article_publisher":"https:\/\/www.facebook.com\/cometdotml","article_published_time":"2023-11-02T17:24:56+00:00","article_modified_time":"2025-04-24T17:04:56+00:00","og_image":[{"url":"https:\/\/miro.medium.com\/v2\/resize:fit:700\/0*QOhDWF6sla2bkUj4","type":"","width":"","height":""}],"author":"Kurtis Pykes","twitter_card":"summary_large_image","twitter_creator":"@Cometml","twitter_site":"@Cometml","twitter_misc":{"Written by":"Kurtis Pykes","Est. reading time":"6 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/www.comet.com\/site\/blog\/tracking-artifacts-with-comet-ml#article","isPartOf":{"@id":"https:\/\/www.comet.com\/site\/blog\/tracking-artifacts-with-comet-ml\/"},"author":{"name":"Team Comet Digital","@id":"https:\/\/www.comet.com\/site\/#\/schema\/person\/6266601170c60a7a82b3e0043fbe8ddf"},"headline":"Tracking Artifacts With Comet ML","datePublished":"2023-11-02T17:24:56+00:00","dateModified":"2025-04-24T17:04:56+00:00","mainEntityOfPage":{"@id":"https:\/\/www.comet.com\/site\/blog\/tracking-artifacts-with-comet-ml\/"},"wordCount":882,"publisher":{"@id":"https:\/\/www.comet.com\/site\/#organization"},"image":{"@id":"https:\/\/www.comet.com\/site\/blog\/tracking-artifacts-with-comet-ml#primaryimage"},"thumbnailUrl":"https:\/\/miro.medium.com\/v2\/resize:fit:700\/0*QOhDWF6sla2bkUj4","articleSection":["Product","Tutorials"],"inLanguage":"en-US"},{"@type":"WebPage","@id":"https:\/\/www.comet.com\/site\/blog\/tracking-artifacts-with-comet-ml\/","url":"https:\/\/www.comet.com\/site\/blog\/tracking-artifacts-with-comet-ml","name":"Tracking Artifacts With Comet ML - Comet","isPartOf":{"@id":"https:\/\/www.comet.com\/site\/#website"},"primaryImageOfPage":{"@id":"https:\/\/www.comet.com\/site\/blog\/tracking-artifacts-with-comet-ml#primaryimage"},"image":{"@id":"https:\/\/www.comet.com\/site\/blog\/tracking-artifacts-with-comet-ml#primaryimage"},"thumbnailUrl":"https:\/\/miro.medium.com\/v2\/resize:fit:700\/0*QOhDWF6sla2bkUj4","datePublished":"2023-11-02T17:24:56+00:00","dateModified":"2025-04-24T17:04:56+00:00","breadcrumb":{"@id":"https:\/\/www.comet.com\/site\/blog\/tracking-artifacts-with-comet-ml#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/www.comet.com\/site\/blog\/tracking-artifacts-with-comet-ml"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/www.comet.com\/site\/blog\/tracking-artifacts-with-comet-ml#primaryimage","url":"https:\/\/miro.medium.com\/v2\/resize:fit:700\/0*QOhDWF6sla2bkUj4","contentUrl":"https:\/\/miro.medium.com\/v2\/resize:fit:700\/0*QOhDWF6sla2bkUj4"},{"@type":"BreadcrumbList","@id":"https:\/\/www.comet.com\/site\/blog\/tracking-artifacts-with-comet-ml#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/www.comet.com\/site\/"},{"@type":"ListItem","position":2,"name":"Tracking Artifacts With Comet ML"}]},{"@type":"WebSite","@id":"https:\/\/www.comet.com\/site\/#website","url":"https:\/\/www.comet.com\/site\/","name":"Comet","description":"Build Better Models Faster","publisher":{"@id":"https:\/\/www.comet.com\/site\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/www.comet.com\/site\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"},{"@type":"Organization","@id":"https:\/\/www.comet.com\/site\/#organization","name":"Comet ML, Inc.","alternateName":"Comet","url":"https:\/\/www.comet.com\/site\/","logo":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/www.comet.com\/site\/#\/schema\/logo\/image\/","url":"https:\/\/www.comet.com\/site\/wp-content\/uploads\/2025\/01\/logo_comet_square.png","contentUrl":"https:\/\/www.comet.com\/site\/wp-content\/uploads\/2025\/01\/logo_comet_square.png","width":310,"height":310,"caption":"Comet ML, Inc."},"image":{"@id":"https:\/\/www.comet.com\/site\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/www.facebook.com\/cometdotml","https:\/\/x.com\/Cometml","https:\/\/www.youtube.com\/channel\/UCmN63HKvfXSCS-UwVwmK8Hw"]},{"@type":"Person","@id":"https:\/\/www.comet.com\/site\/#\/schema\/person\/6266601170c60a7a82b3e0043fbe8ddf","name":"Team Comet Digital","image":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/www.comet.com\/site\/#\/schema\/person\/image\/4f0c0a8cc7c0e87c636ff6a420a6647c","url":"https:\/\/www.comet.com\/site\/wp-content\/uploads\/2023\/08\/Screen-Shot-2023-08-12-at-8.58.50-AM-96x96.png","contentUrl":"https:\/\/www.comet.com\/site\/wp-content\/uploads\/2023\/08\/Screen-Shot-2023-08-12-at-8.58.50-AM-96x96.png","caption":"Team Comet Digital"},"sameAs":["https:\/\/www.comet.ml\/"],"url":"https:\/\/www.comet.com\/site\/blog\/author\/teamcometdigital\/"}]}},"_links":{"self":[{"href":"https:\/\/www.comet.com\/site\/wp-json\/wp\/v2\/posts\/8070","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.comet.com\/site\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.comet.com\/site\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.comet.com\/site\/wp-json\/wp\/v2\/users\/8"}],"replies":[{"embeddable":true,"href":"https:\/\/www.comet.com\/site\/wp-json\/wp\/v2\/comments?post=8070"}],"version-history":[{"count":1,"href":"https:\/\/www.comet.com\/site\/wp-json\/wp\/v2\/posts\/8070\/revisions"}],"predecessor-version":[{"id":15473,"href":"https:\/\/www.comet.com\/site\/wp-json\/wp\/v2\/posts\/8070\/revisions\/15473"}],"wp:attachment":[{"href":"https:\/\/www.comet.com\/site\/wp-json\/wp\/v2\/media?parent=8070"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.comet.com\/site\/wp-json\/wp\/v2\/categories?post=8070"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.comet.com\/site\/wp-json\/wp\/v2\/tags?post=8070"},{"taxonomy":"author","embeddable":true,"href":"https:\/\/www.comet.com\/site\/wp-json\/wp\/v2\/coauthors?post=8070"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}
![\"\"](\"https:\/\/miro.medium.com\/v2\/resize:fit:600\/1*r_9wyM5ScrTeBzxcNAeMmg.gif\")
<\/figure>![\"\"](\"https:\/\/miro.medium.com\/v2\/resize:fit:600\/1*OIgUkyPqof4Ihk9trgudTw.gif\")
<\/figure>