Update README.md

README.md

@@ -1,70 +1,46 @@
 ---
-license: mit
 base_model: microsoft/deberta-v3-xsmall
+language:
+- en
 tags:
-- generated_from_trainer
-metrics:
-- accuracy
-model-index:
-- name: deberta-v3-xsmall-zeroshot-v1.1-none
-  results: []
+- text-classification
+- zero-shot-classification
+pipeline_tag: zero-shot-classification
+library_name: transformers
+license: mit
 ---
 
-<!-- This model card has been generated automatically according to the information the Trainer had access to. You
-should probably proofread and complete it, then remove this comment. -->
-
-# deberta-v3-xsmall-zeroshot-v1.1-none
-
-This model is a fine-tuned version of [microsoft/deberta-v3-xsmall](https://huggingface.co/microsoft/deberta-v3-xsmall) on an unknown dataset.
-It achieves the following results on the evaluation set:
-- Loss: 0.2072
-- F1 Macro: 0.6369
-- F1 Micro: 0.7013
-- Accuracy Balanced: 0.6751
-- Accuracy: 0.7013
-- Precision Macro: 0.6439
-- Recall Macro: 0.6751
-- Precision Micro: 0.7013
-- Recall Micro: 0.7013
-
-## Model description
-
-More information needed
-
-## Intended uses & limitations
 
-More information needed
+# deberta-v3-xsmall-zeroshot-v1.1-all-33 
 
-## Training and evaluation data
+This model was fine-tuned using the same pipeline as described in 
+the model card for [MoritzLaurer/deberta-v3-large-zeroshot-v1.1-all-33](https://huggingface.co/MoritzLaurer/deberta-v3-large-zeroshot-v1.1-all-33)
+and in this [paper](https://arxiv.org/pdf/2312.17543.pdf).
+ 
+The foundation model is [microsoft/deberta-v3-xsmall](https://huggingface.co/microsoft/deberta-v3-xsmall).
+The model only has 22 million backbone parameters and 128 million vocabulary parameters. 
+The backbone parameters are the main parameters active during inference, providing a significant speedup over larger models. 
+The model is 241 MB small.
 
-More information needed
+This model was trained to provide a small and highly efficient zeroshot option, 
+especially for edge devices or in-browser use-cases with transformers.js.
 
-## Training procedure
+## Usage and other details
+For usage instructions and other details refer to 
+this model card [MoritzLaurer/deberta-v3-large-zeroshot-v1.1-all-33](https://huggingface.co/MoritzLaurer/deberta-v3-large-zeroshot-v1.1-all-33)
+and this [paper](https://arxiv.org/pdf/2312.17543.pdf).
 
-### Training hyperparameters
+## Metrics:
 
-The following hyperparameters were used during training:
-- learning_rate: 2e-05
-- train_batch_size: 32
-- eval_batch_size: 128
-- seed: 42
-- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
-- lr_scheduler_type: linear
-- lr_scheduler_warmup_ratio: 0.06
-- num_epochs: 3
+I didn't not do zeroshot evaluation for this model to save time and compute. 
+The table below shows standard accuracy for all datasets the model was trained on (note that the NLI datasets are binary).
 
-### Training results
+General takeaway: the model is much more efficient than its larger sisters, but it performs less well. 
 
-| Training Loss | Epoch | Step  | Validation Loss | F1 Macro | F1 Micro | Accuracy Balanced | Accuracy | Precision Macro | Recall Macro | Precision Micro | Recall Micro |
-|:-------------:|:-----:|:-----:|:---------------:|:--------:|:--------:|:-----------------:|:--------:|:---------------:|:------------:|:---------------:|:------------:|
-| 0.2532        | 1.0   | 30790 | 0.4006          | 0.8198   | 0.8384   | 0.8151            | 0.8384   | 0.8257          | 0.8151       | 0.8384          | 0.8384       |
-| 0.2113        | 2.0   | 61580 | 0.3907          | 0.8254   | 0.8439   | 0.8198            | 0.8439   | 0.8326          | 0.8198       | 0.8439          | 0.8439       |
-| 0.1727        | 3.0   | 92370 | 0.4228          | 0.8306   | 0.8461   | 0.8297            | 0.8461   | 0.8315          | 0.8297       | 0.8461          | 0.8461       |
+|Datasets|mnli_m|mnli_mm|fevernli|anli_r1|anli_r2|anli_r3|wanli|lingnli|wellformedquery|rottentomatoes|amazonpolarity|imdb|yelpreviews|hatexplain|massive|banking77|emotiondair|emocontext|empathetic|agnews|yahootopics|biasframes_sex|biasframes_offensive|biasframes_intent|financialphrasebank|appreviews|hateoffensive|trueteacher|spam|wikitoxic_toxicaggregated|wikitoxic_obscene|wikitoxic_identityhate|wikitoxic_threat|wikitoxic_insult|manifesto|capsotu|
+| :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: |
+|Accuracy|0.925|0.923|0.886|0.732|0.633|0.661|0.814|0.887|0.722|0.872|0.944|0.925|0.967|0.774|0.734|0.627|0.762|0.745|0.465|0.888|0.702|0.94|0.853|0.863|0.914|0.926|0.921|0.635|0.968|0.897|0.918|0.915|0.935|0.9|0.505|0.701|
+|Inference text/sec (A100, batch=128)|1573.0|1630.0|683.0|1282.0|1352.0|1072.0|2325.0|2008.0|4781.0|2743.0|677.0|228.0|238.0|2357.0|5027.0|4323.0|3247.0|3129.0|941.0|1643.0|335.0|1517.0|1452.0|1498.0|2367.0|974.0|2634.0|353.0|2284.0|260.0|252.0|256.0|254.0|259.0|1941.0|2080.0|
 
 
-### Framework versions
 
-- Transformers 4.33.3
-- Pytorch 2.1.2+cu121
-- Datasets 2.14.7
-- Tokenizers 0.13.3