Add adapter bert-base-multilingual-uncased-hinglish-sentiment version 1

README.md

@@ -0,0 +1,64 @@
+---
+tags:
+- adapterhub:sentiment/hinglish-twitter-sentiment
+- text-classification
+- bert
+- adapter-transformers
+license: "apache-2.0"
+---
+
+# Adapter `bert-base-multilingual-uncased-hinglish-sentiment` for bert-base-multilingual-uncased
+
+**Note: This adapter was not trained by the AdapterHub team, but by these author(s): Meghana Bhange, Nirant K.
+See author details below.**
+
+Adapter for Hinglish Sentiment Analysis, based on SemEval 2020 Task 9
+
+**This adapter was created for usage with the [Adapters](https://github.com/Adapter-Hub/adapters) library.**
+
+## Usage
+
+First, install `adapters`:
+
+```
+pip install -U adapters
+```
+
+Now, the adapter can be loaded and activated like this:
+
+```python
+from adapters import AutoAdapterModel
+
+model = AutoAdapterModel.from_pretrained("bert-base-multilingual-uncased")
+adapter_name = model.load_adapter("AdapterHub/bert-base-multilingual-uncased-hinglish-sentiment")
+model.set_active_adapters(adapter_name)
+```
+
+## Architecture & Training
+
+- Adapter architecture: pfeiffer
+- Prediction head: classification
+- Dataset: [Hinglish Sentiment](https://ritual-uh.github.io/sentimix2020/hinglish_res)
+
+## Author Information
+
+- Author name(s): Meghana Bhange, Nirant K
+- Author email: hinglish@nirantk.com
+- Author links: [Website](https://github.com/NirantK), [GitHub](https://github.com/NirantK), [Twitter](https://twitter.com/@NirantK)
+
+
+
+## Citation
+
+```bibtex
+@article{Hinglish,
+    title={HinglishNLP: Fine-tuned Language Models for Hinglish Sentiment Detection},
+    author={Meghana Bhange,
+            Nirant Kasliwal,
+    journal={ArXiv},
+    year={2020}
+}
+
+```
+
+*This adapter has been auto-imported from https://github.com/Adapter-Hub/Hub/blob/master/adapters/nirantk/bert-base-multilingual-uncased-hinglish-sentiment.yaml*.

adapter_config.json

@@ -0,0 +1,41 @@
+{
+  "config": {
+    "adapter_residual_before_ln": false,
+    "cross_adapter": false,
+    "dropout": 0.0,
+    "factorized_phm_W": true,
+    "factorized_phm_rule": false,
+    "hypercomplex_nonlinearity": "glorot-uniform",
+    "init_weights": "bert",
+    "inv_adapter": null,
+    "inv_adapter_reduction_factor": null,
+    "is_parallel": false,
+    "learn_phm": true,
+    "leave_out": [],
+    "ln_after": false,
+    "ln_before": false,
+    "mh_adapter": false,
+    "non_linearity": "relu",
+    "original_ln_after": true,
+    "original_ln_before": true,
+    "output_adapter": true,
+    "phm_bias": true,
+    "phm_c_init": "normal",
+    "phm_dim": 4,
+    "phm_init_range": 0.0001,
+    "phm_layer": false,
+    "phm_rank": 1,
+    "reduction_factor": 16,
+    "residual_before_ln": true,
+    "scaling": 1.0,
+    "shared_W_phm": false,
+    "shared_phm_rule": true,
+    "use_gating": false
+  },
+  "hidden_size": 768,
+  "model_class": "BertAdapterModel",
+  "model_name": "bert-base-multilingual-uncased",
+  "model_type": "bert",
+  "name": "hinglish",
+  "version": "0.2.0"
+}

head_config.json

@@ -0,0 +1,22 @@
+{
+  "config": {
+    "activation_function": "tanh",
+    "bias": true,
+    "dropout_prob": null,
+    "head_type": "classification",
+    "label2id": {
+      "LABEL_0": 0,
+      "LABEL_1": 1,
+      "LABEL_2": 2
+    },
+    "layers": 2,
+    "num_labels": 3,
+    "use_pooler": false
+  },
+  "hidden_size": 768,
+  "model_class": "BertAdapterModel",
+  "model_name": "bert-base-multilingual-uncased",
+  "model_type": "bert",
+  "name": "hinglish",
+  "version": "0.2.0"
+}

pytorch_adapter.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:684c5fea3310692dc8ef8a665c0b1adbd943ee110bc8ff1bf2667b870e19f6ad
+size 3594918

pytorch_model_head.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:62b3cd45b5f4b76f308a289163c5ccb15f479f5bd53e37764f1198360895d52b
+size 2373736