{"id":32206269,"url":"https://github.com/psychbruce/fmat","last_synced_at":"2025-10-22T05:11:29.460Z","repository":{"id":160105933,"uuid":"589967946","full_name":"psychbruce/FMAT","owner":"psychbruce","description":"😷 The Fill-Mask Association Test (FMAT): Measuring Propositions in Natural Language.","archived":false,"fork":false,"pushed_at":"2025-08-20T13:07:36.000Z","size":10196,"stargazers_count":15,"open_issues_count":0,"forks_count":0,"subscribers_count":3,"default_branch":"main","last_synced_at":"2025-09-08T15:42:30.217Z","etag":null,"topics":["ai","artificial-intelligence","bert","bert-model","bert-models","contextualized-representation","fill-in-the-blank","fill-mask","huggingface","language-model","language-models","large-language-models","masked-language-models","natural-language-processing","natural-language-understanding","nlp","pretrained-models","transformer","transformers"],"latest_commit_sha":null,"homepage":"https://psychbruce.github.io/FMAT/","language":"R","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"gpl-3.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/psychbruce.png","metadata":{"files":{"readme":"README.md","changelog":"NEWS.md","contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2023-01-17T11:11:38.000Z","updated_at":"2025-08-20T13:04:24.000Z","dependencies_parsed_at":"2024-05-19T06:23:23.092Z","dependency_job_id":"3f71d181-36e5-42fe-ba65-74829e3caba7","html_url":"https://github.com/psychbruce/FMAT","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/psychbruce/FMAT","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/psychbruce%2FFMAT","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/psychbruce%2FFMAT/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/psychbruce%2FFMAT/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/psychbruce%2FFMAT/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/psychbruce","download_url":"https://codeload.github.com/psychbruce/FMAT/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/psychbruce%2FFMAT/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":280383000,"owners_count":26321423,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","status":"online","status_checked_at":"2025-10-22T02:00:06.515Z","response_time":63,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["ai","artificial-intelligence","bert","bert-model","bert-models","contextualized-representation","fill-in-the-blank","fill-mask","huggingface","language-model","language-models","large-language-models","masked-language-models","natural-language-processing","natural-language-understanding","nlp","pretrained-models","transformer","transformers"],"created_at":"2025-10-22T05:11:27.729Z","updated_at":"2025-10-22T05:11:29.452Z","avatar_url":"https://github.com/psychbruce.png","language":"R","funding_links":[],"categories":[],"sub_categories":[],"readme":"# FMAT \u003cimg src=\"man/figures/logo.png\" align=\"right\" height=\"160\"/\u003e\n\n😷 The Fill-Mask Association Test (掩码填空联系测验).\n\nThe *Fill-Mask Association Test* (FMAT) is an integrative and probability-based method using [BERT Models] to measure conceptual associations (e.g., attitudes, biases, stereotypes, social norms, cultural values) as *propositions* in natural language ([Bao, 2024, *JPSP*](https://doi.org/10.1037/pspa0000396)).\n\n⚠️ *Please update this package to version ≥ 2025.4 for faster and more robust functionality.*\n\n![](https://psychbruce.github.io/img/FMAT-Workflow.png)\n\n\u003c!-- badges: start --\u003e\n\n[![CRAN-Version](https://www.r-pkg.org/badges/version/FMAT?color=red)](https://CRAN.R-project.org/package=FMAT) [![GitHub-Version](https://img.shields.io/github/r-package/v/psychbruce/FMAT?label=GitHub\u0026color=orange)](https://github.com/psychbruce/FMAT) [![R-CMD-check](https://github.com/psychbruce/FMAT/actions/workflows/R-CMD-check.yaml/badge.svg)](https://github.com/psychbruce/FMAT/actions/workflows/R-CMD-check.yaml) [![CRAN-Downloads](https://cranlogs.r-pkg.org/badges/grand-total/FMAT)](https://CRAN.R-project.org/package=FMAT) [![GitHub-Stars](https://img.shields.io/github/stars/psychbruce/FMAT?style=social)](https://github.com/psychbruce/FMAT/stargazers)\n\n\u003c!-- badges: end --\u003e\n\n\u003cimg src=\"https://psychbruce.github.io/img/CC-BY-NC-SA.jpg\" width=\"120px\" height=\"42px\"/\u003e\n\n## Author\n\nBruce H. W. S. Bao 包寒吴霜\n\n📬 [baohws\\@foxmail.com](mailto:baohws@foxmail.com)\n\n📋 [psychbruce.github.io](https://psychbruce.github.io)\n\n## Citation\n\n### (1) FMAT Package\n\n-   Bao, H. W. S. (2023). *FMAT: The Fill-Mask Association Test*. \u003chttps://doi.org/10.32614/CRAN.package.FMAT\u003e\n\n### (2) FMAT Research Articles - Methodology\n\n-   Bao, H. W. S. (2024). The Fill-Mask Association Test (FMAT): Measuring propositions in natural language. *Journal of Personality and Social Psychology, 127*(3), 537–561. \u003chttps://doi.org/10.1037/pspa0000396\u003e\n\n### (3) FMAT Research Articles - Application\n\n-   Bao, H. W. S., \u0026 Gries, P. (2024). Intersectional race–gender stereotypes in natural language. *British Journal of Social Psychology, 63*(4), 1771–1786. \u003chttps://doi.org/10.1111/bjso.12748\u003e\n-   Bao, H. W. S., \u0026 Gries, P. (2025). Biases about Chinese people in English language use: Stereotypes, prejudice and discrimination. *China Quarterly*. \u003chttps://doi.org/10.1017/S0305741025100532\u003e\n-   Wang, Z., Xia, H., Bao, H. W. S., Jing, Y., \u0026 Gu, R. (2025). Artificial intelligence is stereotypically linked more with socially dominant groups in natural language. *Advanced Science*. \u003chttps://doi.org/10.1002/advs.202508623\u003e\n\n## Installation\n\nThe R package `FMAT` and three Python packages (`transformers`, `torch`, `huggingface-hub`) all need to be installed.\n\n### (1) R Package\n\n``` r\n## Method 1: Install from CRAN\ninstall.packages(\"FMAT\")\n\n## Method 2: Install from GitHub\ninstall.packages(\"devtools\")\ndevtools::install_github(\"psychbruce/FMAT\", force=TRUE)\n```\n\n### (2) Python Environment and Packages\n\nInstall [Anaconda](https://www.anaconda.com/download/success) (a recommended package manager that automatically installs Python, its IDEs like Spyder, and a large list of common Python packages).\n\nSpecify the Anaconda's Python interpreter in RStudio.\n\n\u003e RStudio → Tools → Global/Project Options\\\n\u003e → Python → Select → **Conda Environments**\\\n\u003e → Choose **\".../Anaconda3/python.exe\"**\n\nInstall specific versions of Python packages \"[transformers](https://pypi.org/project/transformers/#history)\", \"[torch](https://pypi.org/project/torch/#history)\", and \"[huggingface-hub](https://pypi.org/project/huggingface-hub/#history)\".\\\n(RStudio Terminal / Anaconda Prompt / Windows Command)\n\nFor CPU users:\n\n```         \npip install transformers==4.40.2 torch==2.2.1 huggingface-hub==0.20.3\n```\n\nFor GPU (CUDA) users:\n\n```         \npip install transformers==4.40.2 huggingface-hub==0.20.3\npip install torch==2.2.1 --index-url https://download.pytorch.org/whl/cu121\n```\n\nTo use some models (e.g., `microsoft/deberta-v3-base`), \"You need to have sentencepiece installed to convert a slow tokenizer to a fast one\":\n\n```         \npip install sentencepiece\n```\n\n-   See [Guidance for GPU Acceleration] for installation guidance if you have an NVIDIA GPU device on your PC and want to use GPU to accelerate the pipeline.\n-   According to the May 2024 releases, \"transformers\" ≥ 4.41 depends on \"huggingface-hub\" ≥ 0.23. The suggested versions of \"transformers\" (4.40.2) and \"huggingface-hub\" (0.20.3) ensure the console display of progress bars when downloading BERT models while keeping these packages as new as possible.\n-   Proxy users may use the \"global mode\" (全局模式) to download models.\n-   If you find the error `HTTPSConnectionPool(host='huggingface.co', port=443)`, please try to (1) reinstall [Anaconda](https://www.anaconda.com/download/success) so that some unknown issues may be fixed, or (2) downgrade the \"[urllib3](https://pypi.org/project/urllib3/)\" package to version ≤ 1.25.11 (`pip install urllib3==1.25.11`) so that it will use HTTP proxies (rather than HTTPS proxies as in later versions) to connect to Hugging Face.\n\n## Guidance for FMAT\n\n### Step 1: Download BERT Models\n\nUse `BERT_download()` to download [BERT models]. Model files are saved in your local cache folder \"%USERPROFILE%/.cache/huggingface\". A full list of BERT models are available at [Hugging Face](https://huggingface.co/models?pipeline_tag=fill-mask).\n\nUse `BERT_info()` and `BERT_vocab()` to obtain detailed information of BERT models.\n\n### Step 2: Design FMAT Queries\n\nDesign queries that conceptually represent the constructs you would measure (see [Bao, 2024, *JPSP*](https://doi.org/10.1037/pspa0000396) for how to design queries).\n\nUse `FMAT_query()` and/or `FMAT_query_bind()` to prepare a `data.table` of queries.\n\n### Step 3: Run FMAT\n\nUse `FMAT_run()` to get raw data (probability estimates) for further analysis.\n\nSeveral steps of preprocessing have been included in the function for easier use (see `FMAT_run()` for details).\n\n-   For BERT variants using `\u003cmask\u003e` rather than `[MASK]` as the mask token, the input query will be *automatically* modified so that users can always use `[MASK]` in query design.\n-   For some BERT variants, special prefix characters such as `\\u0120` and `\\u2581` will be *automatically* added to match the whole words (rather than subwords) for `[MASK]`.\n\n### Notes\n\n-   Improvements are ongoing, especially for adaptation to more diverse (less popular) BERT models.\n-   If you find bugs or have problems using the functions, please report them at [GitHub Issues](https://github.com/psychbruce/FMAT/issues) or send me an email.\n\n## Guidance for GPU Acceleration\n\nBy default, the `FMAT` package uses CPU to enable the functionality for all users. But for advanced users who want to accelerate the pipeline with GPU, the `FMAT_run()` function now supports using a GPU device, about **3x faster** than CPU.\n\nTest results (on the developer's computer, depending on BERT model size):\n\n-   CPU (Intel 13th-Gen i7-1355U): 500\\~1000 queries/min\n-   GPU (NVIDIA GeForce RTX 2050): 1500\\~3000 queries/min\n\nChecklist:\n\n1.  Ensure that you have an NVIDIA GPU device (e.g., GeForce RTX Series) and an NVIDIA GPU driver installed on your system.\n2.  Install PyTorch (Python `torch` package) with CUDA support.\n    -   Find guidance for installation command at \u003chttps://pytorch.org/get-started/locally/\u003e.\n    -   CUDA is available only on Windows and Linux, but not on MacOS.\n    -   If you have installed a version of `torch` without CUDA support, please first uninstall it (command: `pip uninstall torch`) and then install the suggested one.\n    -   You may also install the corresponding version of CUDA Toolkit (e.g., for the `torch` version supporting CUDA 12.1, the same version of [CUDA Toolkit 12.1](https://developer.nvidia.com/cuda-12-1-0-download-archive) may also be installed).\n\nExample code for installing PyTorch with CUDA support:\\\n(RStudio Terminal / Anaconda Prompt / Windows Command)\n\n```         \npip install torch==2.2.1 --index-url https://download.pytorch.org/whl/cu121\n```\n\n## BERT Models\n\nThe reliability and validity of the following 12 BERT models in the FMAT have been established in our research, but future work is needed to examine the performance of other models.\n\n(model name on Hugging Face - model file size)\n\n1.  [bert-base-uncased](https://huggingface.co/bert-base-uncased) (420 MB)\n2.  [bert-base-cased](https://huggingface.co/bert-base-cased) (416 MB)\n3.  [bert-large-uncased](https://huggingface.co/bert-large-uncased) (1283 MB)\n4.  [bert-large-cased](https://huggingface.co/bert-large-cased) (1277 MB)\n5.  [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) (256 MB)\n6.  [distilbert-base-cased](https://huggingface.co/distilbert-base-cased) (251 MB)\n7.  [albert-base-v1](https://huggingface.co/albert-base-v1) (45 MB)\n8.  [albert-base-v2](https://huggingface.co/albert-base-v2) (45 MB)\n9.  [roberta-base](https://huggingface.co/roberta-base) (476 MB)\n10. [distilroberta-base](https://huggingface.co/distilroberta-base) (316 MB)\n11. [vinai/bertweet-base](https://huggingface.co/vinai/bertweet-base) (517 MB)\n12. [vinai/bertweet-large](https://huggingface.co/vinai/bertweet-large) (1356 MB)\n\nFor details about [BERT](https://arxiv.org/abs/1810.04805), see:\n\n-   [What is Fill-Mask? [HuggingFace]](https://huggingface.co/tasks/fill-mask)\n-   [An Explorable BERT [HuggingFace]](https://huggingface.co/spaces/exbert-project/exbert)\n-   [BERT Model Documentation [HuggingFace]](https://huggingface.co/docs/transformers/main/en/model_doc/bert)\n-   [Illustrated BERT](https://jalammar.github.io/illustrated-bert/)\n-   [Visual Guide to BERT](https://jalammar.github.io/a-visual-guide-to-using-bert-for-the-first-time/)\n\n``` r\nlibrary(FMAT)\nmodels = c(\n  \"bert-base-uncased\",\n  \"bert-base-cased\",\n  \"bert-large-uncased\",\n  \"bert-large-cased\",\n  \"distilbert-base-uncased\",\n  \"distilbert-base-cased\",\n  \"albert-base-v1\",\n  \"albert-base-v2\",\n  \"roberta-base\",\n  \"distilroberta-base\",\n  \"vinai/bertweet-base\",\n  \"vinai/bertweet-large\"\n)\nBERT_download(models)\n```\n\n``` {style=\"height: 500px\"}\nℹ Device Info:\n\nR Packages:\nFMAT          2024.5\nreticulate    1.36.1\n\nPython Packages:\ntransformers  4.40.2\ntorch         2.2.1+cu121\n\nNVIDIA GPU CUDA Support:\nCUDA Enabled: TRUE\nCUDA Version: 12.1\nGPU (Device): NVIDIA GeForce RTX 2050\n\n\n── Downloading model \"bert-base-uncased\" ──────────────────────────────────────────\n→ (1) Downloading configuration...\nconfig.json: 100%|██████████| 570/570 [00:00\u003c00:00, 114kB/s]\n→ (2) Downloading tokenizer...\ntokenizer_config.json: 100%|██████████| 48.0/48.0 [00:00\u003c00:00, 23.9kB/s]\nvocab.txt: 100%|██████████| 232k/232k [00:00\u003c00:00, 1.50MB/s]\ntokenizer.json: 100%|██████████| 466k/466k [00:00\u003c00:00, 1.98MB/s]\n→ (3) Downloading model...\nmodel.safetensors: 100%|██████████| 440M/440M [00:36\u003c00:00, 12.1MB/s] \n✔ Successfully downloaded model \"bert-base-uncased\"\n\n── Downloading model \"bert-base-cased\" ────────────────────────────────────────────\n→ (1) Downloading configuration...\nconfig.json: 100%|██████████| 570/570 [00:00\u003c00:00, 63.3kB/s]\n→ (2) Downloading tokenizer...\ntokenizer_config.json: 100%|██████████| 49.0/49.0 [00:00\u003c00:00, 8.66kB/s]\nvocab.txt: 100%|██████████| 213k/213k [00:00\u003c00:00, 1.39MB/s]\ntokenizer.json: 100%|██████████| 436k/436k [00:00\u003c00:00, 10.1MB/s]\n→ (3) Downloading model...\nmodel.safetensors: 100%|██████████| 436M/436M [00:37\u003c00:00, 11.6MB/s] \n✔ Successfully downloaded model \"bert-base-cased\"\n\n── Downloading model \"bert-large-uncased\" ─────────────────────────────────────────\n→ (1) Downloading configuration...\nconfig.json: 100%|██████████| 571/571 [00:00\u003c00:00, 268kB/s]\n→ (2) Downloading tokenizer...\ntokenizer_config.json: 100%|██████████| 48.0/48.0 [00:00\u003c00:00, 12.0kB/s]\nvocab.txt: 100%|██████████| 232k/232k [00:00\u003c00:00, 1.50MB/s]\ntokenizer.json: 100%|██████████| 466k/466k [00:00\u003c00:00, 1.99MB/s]\n→ (3) Downloading model...\nmodel.safetensors: 100%|██████████| 1.34G/1.34G [01:36\u003c00:00, 14.0MB/s]\n✔ Successfully downloaded model \"bert-large-uncased\"\n\n── Downloading model \"bert-large-cased\" ───────────────────────────────────────────\n→ (1) Downloading configuration...\nconfig.json: 100%|██████████| 762/762 [00:00\u003c00:00, 125kB/s]\n→ (2) Downloading tokenizer...\ntokenizer_config.json: 100%|██████████| 49.0/49.0 [00:00\u003c00:00, 12.3kB/s]\nvocab.txt: 100%|██████████| 213k/213k [00:00\u003c00:00, 1.41MB/s]\ntokenizer.json: 100%|██████████| 436k/436k [00:00\u003c00:00, 5.39MB/s]\n→ (3) Downloading model...\nmodel.safetensors: 100%|██████████| 1.34G/1.34G [01:35\u003c00:00, 14.0MB/s]\n✔ Successfully downloaded model \"bert-large-cased\"\n\n── Downloading model \"distilbert-base-uncased\" ────────────────────────────────────\n→ (1) Downloading configuration...\nconfig.json: 100%|██████████| 483/483 [00:00\u003c00:00, 161kB/s]\n→ (2) Downloading tokenizer...\ntokenizer_config.json: 100%|██████████| 48.0/48.0 [00:00\u003c00:00, 9.46kB/s]\nvocab.txt: 100%|██████████| 232k/232k [00:00\u003c00:00, 16.5MB/s]\ntokenizer.json: 100%|██████████| 466k/466k [00:00\u003c00:00, 14.8MB/s]\n→ (3) Downloading model...\nmodel.safetensors: 100%|██████████| 268M/268M [00:19\u003c00:00, 13.5MB/s] \n✔ Successfully downloaded model \"distilbert-base-uncased\"\n\n── Downloading model \"distilbert-base-cased\" ──────────────────────────────────────\n→ (1) Downloading configuration...\nconfig.json: 100%|██████████| 465/465 [00:00\u003c00:00, 233kB/s]\n→ (2) Downloading tokenizer...\ntokenizer_config.json: 100%|██████████| 49.0/49.0 [00:00\u003c00:00, 9.80kB/s]\nvocab.txt: 100%|██████████| 213k/213k [00:00\u003c00:00, 1.39MB/s]\ntokenizer.json: 100%|██████████| 436k/436k [00:00\u003c00:00, 8.70MB/s]\n→ (3) Downloading model...\nmodel.safetensors: 100%|██████████| 263M/263M [00:24\u003c00:00, 10.9MB/s] \n✔ Successfully downloaded model \"distilbert-base-cased\"\n\n── Downloading model \"albert-base-v1\" ─────────────────────────────────────────────\n→ (1) Downloading configuration...\nconfig.json: 100%|██████████| 684/684 [00:00\u003c00:00, 137kB/s]\n→ (2) Downloading tokenizer...\ntokenizer_config.json: 100%|██████████| 25.0/25.0 [00:00\u003c00:00, 3.57kB/s]\nspiece.model: 100%|██████████| 760k/760k [00:00\u003c00:00, 4.93MB/s]\ntokenizer.json: 100%|██████████| 1.31M/1.31M [00:00\u003c00:00, 13.4MB/s]\n→ (3) Downloading model...\nmodel.safetensors: 100%|██████████| 47.4M/47.4M [00:03\u003c00:00, 13.4MB/s]\n✔ Successfully downloaded model \"albert-base-v1\"\n\n── Downloading model \"albert-base-v2\" ─────────────────────────────────────────────\n→ (1) Downloading configuration...\nconfig.json: 100%|██████████| 684/684 [00:00\u003c00:00, 137kB/s]\n→ (2) Downloading tokenizer...\ntokenizer_config.json: 100%|██████████| 25.0/25.0 [00:00\u003c00:00, 4.17kB/s]\nspiece.model: 100%|██████████| 760k/760k [00:00\u003c00:00, 5.10MB/s]\ntokenizer.json: 100%|██████████| 1.31M/1.31M [00:00\u003c00:00, 6.93MB/s]\n→ (3) Downloading model...\nmodel.safetensors: 100%|██████████| 47.4M/47.4M [00:03\u003c00:00, 13.8MB/s]\n✔ Successfully downloaded model \"albert-base-v2\"\n\n── Downloading model \"roberta-base\" ───────────────────────────────────────────────\n→ (1) Downloading configuration...\nconfig.json: 100%|██████████| 481/481 [00:00\u003c00:00, 80.3kB/s]\n→ (2) Downloading tokenizer...\ntokenizer_config.json: 100%|██████████| 25.0/25.0 [00:00\u003c00:00, 6.25kB/s]\nvocab.json: 100%|██████████| 899k/899k [00:00\u003c00:00, 2.72MB/s]\nmerges.txt: 100%|██████████| 456k/456k [00:00\u003c00:00, 8.22MB/s]\ntokenizer.json: 100%|██████████| 1.36M/1.36M [00:00\u003c00:00, 8.56MB/s]\n→ (3) Downloading model...\nmodel.safetensors: 100%|██████████| 499M/499M [00:38\u003c00:00, 12.9MB/s] \n✔ Successfully downloaded model \"roberta-base\"\n\n── Downloading model \"distilroberta-base\" ─────────────────────────────────────────\n→ (1) Downloading configuration...\nconfig.json: 100%|██████████| 480/480 [00:00\u003c00:00, 96.4kB/s]\n→ (2) Downloading tokenizer...\ntokenizer_config.json: 100%|██████████| 25.0/25.0 [00:00\u003c00:00, 12.0kB/s]\nvocab.json: 100%|██████████| 899k/899k [00:00\u003c00:00, 6.59MB/s]\nmerges.txt: 100%|██████████| 456k/456k [00:00\u003c00:00, 9.46MB/s]\ntokenizer.json: 100%|██████████| 1.36M/1.36M [00:00\u003c00:00, 11.5MB/s]\n→ (3) Downloading model...\nmodel.safetensors: 100%|██████████| 331M/331M [00:25\u003c00:00, 13.0MB/s] \n✔ Successfully downloaded model \"distilroberta-base\"\n\n── Downloading model \"vinai/bertweet-base\" ────────────────────────────────────────\n→ (1) Downloading configuration...\nconfig.json: 100%|██████████| 558/558 [00:00\u003c00:00, 187kB/s]\n→ (2) Downloading tokenizer...\nvocab.txt: 100%|██████████| 843k/843k [00:00\u003c00:00, 7.44MB/s]\nbpe.codes: 100%|██████████| 1.08M/1.08M [00:00\u003c00:00, 7.01MB/s]\ntokenizer.json: 100%|██████████| 2.91M/2.91M [00:00\u003c00:00, 9.10MB/s]\n→ (3) Downloading model...\npytorch_model.bin: 100%|██████████| 543M/543M [00:48\u003c00:00, 11.1MB/s] \n✔ Successfully downloaded model \"vinai/bertweet-base\"\n\n── Downloading model \"vinai/bertweet-large\" ───────────────────────────────────────\n→ (1) Downloading configuration...\nconfig.json: 100%|██████████| 614/614 [00:00\u003c00:00, 120kB/s]\n→ (2) Downloading tokenizer...\nvocab.json: 100%|██████████| 899k/899k [00:00\u003c00:00, 5.90MB/s]\nmerges.txt: 100%|██████████| 456k/456k [00:00\u003c00:00, 7.30MB/s]\ntokenizer.json: 100%|██████████| 1.36M/1.36M [00:00\u003c00:00, 8.31MB/s]\n→ (3) Downloading model...\npytorch_model.bin: 100%|██████████| 1.42G/1.42G [02:29\u003c00:00, 9.53MB/s]\n✔ Successfully downloaded model \"vinai/bertweet-large\"\n\n── Downloaded models: ──\n\n                           size\nalbert-base-v1            45 MB\nalbert-base-v2            45 MB\nbert-base-cased          416 MB\nbert-base-uncased        420 MB\nbert-large-cased        1277 MB\nbert-large-uncased      1283 MB\ndistilbert-base-cased    251 MB\ndistilbert-base-uncased  256 MB\ndistilroberta-base       316 MB\nroberta-base             476 MB\nvinai/bertweet-base      517 MB\nvinai/bertweet-large    1356 MB\n\n✔ Downloaded models saved at C:/Users/Bruce/.cache/huggingface/hub (6.52 GB)\n```\n\n``` r\nBERT_info(models)\n```\n\n```         \n                      model   size vocab  dims   mask\n                     \u003cfctr\u003e \u003cchar\u003e \u003cint\u003e \u003cint\u003e \u003cchar\u003e\n 1:       bert-base-uncased  420MB 30522   768 [MASK]\n 2:         bert-base-cased  416MB 28996   768 [MASK]\n 3:      bert-large-uncased 1283MB 30522  1024 [MASK]\n 4:        bert-large-cased 1277MB 28996  1024 [MASK]\n 5: distilbert-base-uncased  256MB 30522   768 [MASK]\n 6:   distilbert-base-cased  251MB 28996   768 [MASK]\n 7:          albert-base-v1   45MB 30000   128 [MASK]\n 8:          albert-base-v2   45MB 30000   128 [MASK]\n 9:            roberta-base  476MB 50265   768 \u003cmask\u003e\n10:      distilroberta-base  316MB 50265   768 \u003cmask\u003e\n11:     vinai/bertweet-base  517MB 64001   768 \u003cmask\u003e\n12:    vinai/bertweet-large 1356MB 50265  1024 \u003cmask\u003e\n```\n\n(Tested 2024-05-16 on the developer's computer: HP Probook 450 G10 Notebook PC)\n\n## Related Packages\n\nWhile the FMAT is an innovative method for the *computational intelligent* analysis of psychology and society, you may also seek for an integrative toolbox for other text-analytic methods. Another R package I developed---[PsychWordVec](https://psychbruce.github.io/PsychWordVec/)---is useful and user-friendly for word embedding analysis (e.g., the Word Embedding Association Test, WEAT). Please refer to its documentation and feel free to use it.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fpsychbruce%2Ffmat","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fpsychbruce%2Ffmat","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fpsychbruce%2Ffmat/lists"}