--- tags: - setfit - sentence-transformers - text-classification - generated_from_setfit_trainer widget: - text: >- TITLE: The Impact of the Mariel Boatlift on the Miami Labor Market ABSTRACT: Using data from the Current Population Survey, this paper describes the effect of the Mariel Boatlift of 1980 on the Miami labor market. The Mariel immigrants increased the Miami labor force by 7%, and the percentage increase in labor supply to less-skilled occupations and industries was even greater because most of the immigrants were relatively unskilled. Nevertheless, the Mariel influx appears to have had virtually no effect on the wages or unemployment rates of less-skilled workers, even among Cubans who had immigrated earlier. The author suggests that the ability of Miami's labor market to rapidly absorb the Mariel immigrants was largely owing to its adjustment to other large waves of immigrants in the two decades before the Mariel Boatlift. - text: >- TITLE: The Regularisation of Unauthorized Migrants: Literature Survey and Country Case Studies ABSTRACT: Regularisation programmes have emerged in the past 25 years or so as one of the mechanisms States use to account for and manage the undocumented immigrant population in their countries, and are usually implemented in concert with the internal and external strengthening of migration controls. This paper attempts to answer the questions arising from such programmes through a survey of nine regularisation programmes in the United States and the European Union. The first part of the survey offers a broad introduction to, overview and analysis of regularisation programmes through a review of available literature on the topic. The second part of the survey is an in-depth analysis of regularisation programmes in nine countries, and provides for each country a brief overview of their current migration policy, legal channels of immigration into the country, and the undocumented population in relation to the country's demographic profile. In order to provide a complete picture of each programme, throughout the survey an attempt has been made to draw on government, non-governmental and academic sources. - text: >- TITLE: When the Feds Hand You a “Light Envelope”: Exploring the Impact of the Federal Budget Sequester on the Build America Bond Program ABSTRACT: In 2009, the US federal government created the Build America Bond (BAB) program to improve access to the capital markets for state and local governments. While the federal government typically subsidizes subnational capital financing indirectly through tax exemption, the BAB program provided direct cash subsidies to governments. The Budget Control Act of 2011 implemented a sequester that reduced BAB subsidies by 5.7 to 8.7% each year since 2013. This paper estimates previous and potential future sequestration losses on all outstanding BABs as of the end of 2022. It delves into BAB use and sequestration by two issuers (governments in the state of Texas and City of Chicago). The paper concludes with a discussion of the current state of BABs as it relates to recent litigation and refinancing attempts. The loss estimates provide lessons for the provision and management of federal direct subsidies available to state and local governments. - text: >- TITLE: Does Personal Contact With Ethnic Minorities Affect Anti-Immigrant Sentiments? Evidence From a Field Experiment ABSTRACT: This article explores the causal effect of personal contact with ethnic minorities on majority members' views on immigration, immigrants' work ethics, and support for lower social assistance benefits to immigrants than to natives. Exogenous variation in personal contact is obtained by randomising soldiers into different rooms during the basic training period for conscripts in the Norwegian Army's North Brigade. Based on contact theory of majority--minority relations, the study spells out why the army can be regarded as an ideal contextual setting for exposure to reduce negative views on minorities. The study finds a substantive effect of contact on views on immigrants' work ethics, but small and insignificant effects on support for welfare dualism, as well as on views on whether immigration makes Norway a better place in which to live. - text: >- TITLE: The Contributions of Immigrants to American Culture ABSTRACT: The standard account of American immigration focuses on the acculturation and assimilation of immigrants and their children to American society. This analysis typically ignores the significant contributions of immigrants to the creation of American culture through the performing arts, sciences, and other cultural pursuits. Immigrants and their children are not born with more creative talents than native-born citizens, but their selectivity and marginality may have pushed and pulled those with ability into high-risk career paths that reward creative work. The presence of large numbers of talented immigrants in Hollywood, academia, and the high-tech industries has pushed American institutions to be more meritocratic and open to innovation than they would be otherwise. metrics: - accuracy - precision - recall - f1 pipeline_tag: text-classification library_name: setfit inference: true base_model: sentence-transformers/paraphrase-MiniLM-L3-v2 model-index: - name: SetFit with sentence-transformers/paraphrase-MiniLM-L3-v2 results: - task: type: text-classification name: Text Classification dataset: name: Unknown type: unknown split: test metrics: - type: accuracy value: 0.98125 name: Accuracy - type: precision value: 0.9933774834437086 name: Precision - type: recall value: 0.9868421052631579 name: Recall - type: f1 value: 0.9900990099009901 name: F1 license: apache-2.0 language: - en --- # paraphrase-MiniLM-L3-v2_immig This SetFit model was trained on 48 title-abstracts samples (24 per class) to differeniate between published studies related to immigration/migration research and those that are not. - **Model Type:** [SetFit](https://github.com/huggingface/setfit) - **Sentence Transformer body:** [sentence-transformers/paraphrase-MiniLM-L3-v2](https://huggingface.co/sentence-transformers/paraphrase-MiniLM-L3-v2) - **Classification head:** a [LogisticRegression](https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html) instance - **Train data/script repository:** [SetFit on GitHub](https://github.com/sumtxt/immigration-papers/model_training) ## Evaluation ### Metrics | Label | Accuracy | Precision | Recall | F1 | |:--------|:---------|:----------|:-------|:-------| | **all** | 0.9812 | 0.9934 | 0.9868 | 0.9901 | ## Uses ### Direct Use for Inference First install the SetFit library: ```bash pip install setfit ``` Then you can load this model and run inference. ```python from setfit import SetFitModel model = SetFitModel.from_pretrained("mmarbach/paraphrase-MiniLM-L3-v2_immig") preds = model("TITLE: ... ABSTRACT: ....") ``` ## Training Details ### Training Set Metrics | Training set | Min | Median | Max | |:-------------|:----|:---------|:----| | Word count | 97 | 155.6458 | 262 | | Label | Training Sample Count | |:------------------|:----------------------| | immigration_topic | 24 | | other_topic | 24 | ### Training Hyperparameters - batch_size: (16, 16) - num_epochs: (4, 4) - max_steps: -1 - sampling_strategy: oversampling - body_learning_rate: (2e-05, 1e-05) - head_learning_rate: 0.01 - loss: CosineSimilarityLoss - distance_metric: cosine_distance - margin: 0.25 - end_to_end: False - use_amp: False - warmup_proportion: 0.1 - l2_weight: 0.01 - seed: 42 - eval_max_steps: -1 - load_best_model_at_end: False ### Training Results | Epoch | Step | Training Loss | Validation Loss | |:------:|:----:|:-------------:|:---------------:| | 0.0133 | 1 | 0.288 | - | | 0.6667 | 50 | 0.1935 | - | | 1.0 | 75 | - | 0.0980 | | 1.3333 | 100 | 0.0472 | - | | 2.0 | 150 | 0.0118 | 0.0767 | | 2.6667 | 200 | 0.0057 | - | | 3.0 | 225 | - | 0.0719 | | 3.3333 | 250 | 0.0047 | - | | 4.0 | 300 | 0.0039 | 0.0718 | ### Framework Versions - Python: 3.12.11 - SetFit: 1.1.2 - Sentence Transformers: 5.0.0 - Transformers: 4.53.0 - PyTorch: 2.7.1 - Datasets: 3.6.0 - Tokenizers: 0.21.2 ```