Hugging Face's logo

MoritzLaurer
/
deberta-v3-base-zeroshot-v2.0-c

Zero-Shot Classification
Transformers
Safetensors
English
deberta-v2
text-classification
Inference Endpoints
Model card Files Community
MoritzLaurer HF staff commited on
Commit
ca6bdff
·
verified ·
1 Parent(s): 31f01d7

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +82 -76
README.md CHANGED
@@ -4,55 +4,50 @@ language:
4
  tags:
5
  - text-classification
6
  - zero-shot-classification
 
7
  pipeline_tag: zero-shot-classification
8
  library_name: transformers
9
  license: mit
10
- datasets:
11
- - nyu-mll/multi_nli
12
- - fever
13
  ---
14
 
15
  # Model description: deberta-v3-base-zeroshot-v2.0
16
- The model is designed for zero-shot classification with the Hugging Face pipeline.
17
 
18
- The main advantage of this `zeroshot-v2.0` series of models is that they are trained on commercially-friendly data
19
- and are fully commercially usable, while my older `zeroshot-v1.1` models included training data with non-commercially licenses data.
20
- An overview of the latest zeroshot classifiers with different sizes and licenses is available in my [Zeroshot Classifier Collection](https://huggingface.co/collections/MoritzLaurer/zeroshot-classifiers-6548b4ff407bb19ff5c3ad6f).
21
 
22
- The model can do one universal classification task: determine whether a hypothesis is "true" or "not true" given a text
 
 
 
 
 
 
 
 
23
  (`entailment` vs. `not_entailment`).
24
  This task format is based on the Natural Language Inference task (NLI).
25
  The task is so universal that any classification task can be reformulated into this task.
26
- Note that compared to other NLI models, this model predicts two classes (`entailment` vs. `not_entailment`)
27
- as opposed to three classes (entailment/neutral/contradiction).
28
 
29
 
30
  ## Training data
31
- The model is trained on two types of fully commercially-friendly data:
32
  1. Synthetic data generated with [Mixtral-8x7B-Instruct-v0.1](https://huggingface.co/mistralai/Mixtral-8x7B-Instruct-v0.1).
33
  I first created a list of 500+ diverse text classification tasks for 25 professions in conversations with Mistral-large. The data was manually curated.
34
- I then used this as seed data to generate several hundred thousand texts for the different tasks with Mixtral-8x7B-Instruct-v0.1.
35
  The final dataset used is available in the [synthetic_zeroshot_mixtral_v0.1](https://huggingface.co/datasets/MoritzLaurer/synthetic_zeroshot_mixtral_v0.1) dataset
36
- in the subset `mixtral_written_text_for_tasks_v4`. Data curation was done in multiple iterations and I will release more information on this process soon.
37
  2. Two commercially-friendly NLI datasets: ([MNLI](https://huggingface.co/datasets/nyu-mll/multi_nli), [FEVER-NLI](https://huggingface.co/datasets/fever)).
38
- These datasets were added to increase generalization. Datasets like ANLI were excluded due to their non-commercial license.
39
-
40
- The model was only trained on English data. I will release a multilingual version of this model soon.
41
- For __multilingual use-cases__,
42
- I alternatively recommend machine translating texts to English with libraries like [EasyNMT](https://github.com/UKPLab/EasyNMT).
43
- English-only models tend to perform better than multilingual models and
44
- validation with English data can be easier if you don't speak all languages in your corpus.
45
-
46
 
47
 
48
- ## How to use the model
49
  ```python
50
  #!pip install transformers[sentencepiece]
51
  from transformers import pipeline
52
  text = "Angela Merkel is a politician in Germany and leader of the CDU"
53
- hypothesis_template = "This example is about {}"
54
  classes_verbalized = ["politics", "economy", "entertainment", "environment"]
55
- zeroshot_classifier = pipeline("zero-shot-classification", model="MoritzLaurer/deberta-v3-base-zeroshot-v2.0")
56
  output = zeroshot_classifier(text, classes_verbalised, hypothesis_template=hypothesis_template, multi_label=False)
57
  print(output)
58
  ```
@@ -62,57 +57,66 @@ print(output)
62
 
63
  ## Metrics
64
 
65
- The model was evaluated on 28 different text classification tasks with the [balanced_accuracy](https://scikit-learn.org/stable/modules/generated/sklearn.metrics.balanced_accuracy_score.html) metric.
66
- The main reference point is `facebook/bart-large-mnli` which is at the time of writing (27.03.24) the most used commercially-friendly 0-shot classifier.
67
- The different `zeroshot-v2.0` models were all trained with the same data and the only difference is the underlying foundation model.
68
-
69
- ![results_aggreg_v2.0](https://raw.githubusercontent.com/MoritzLaurer/zeroshot-classifier/e859471dd183ad44b705c047130433301386aab8/v2_synthetic_data/results/zeroshot-v2.0-aggreg.png)
70
-
71
- | | facebook/bart-large-mnli | roberta-base-zeroshot-v2.0 | roberta-large-zeroshot-v2.0 | deberta-v3-base-zeroshot-v2.0 | deberta-v3-large-zeroshot-v2.0 |
72
- |:---------------------------|---------------------------:|-----------------------------:|------------------------------:|--------------------------------:|---------------------------------:|
73
- | all datasets mean | 0.566 | 0.612 | 0.65 | 0.647 | 0.697 |
74
- | amazonpolarity (2) | 0.937 | 0.924 | 0.951 | 0.937 | 0.952 |
75
- | imdb (2) | 0.892 | 0.871 | 0.904 | 0.893 | 0.923 |
76
- | appreviews (2) | 0.934 | 0.913 | 0.937 | 0.938 | 0.943 |
77
- | yelpreviews (2) | 0.948 | 0.953 | 0.977 | 0.979 | 0.988 |
78
- | rottentomatoes (2) | 0.831 | 0.803 | 0.841 | 0.841 | 0.87 |
79
- | emotiondair (6) | 0.495 | 0.523 | 0.514 | 0.487 | 0.495 |
80
- | emocontext (4) | 0.605 | 0.535 | 0.609 | 0.566 | 0.687 |
81
- | empathetic (32) | 0.366 | 0.386 | 0.417 | 0.388 | 0.455 |
82
- | financialphrasebank (3) | 0.673 | 0.521 | 0.445 | 0.678 | 0.656 |
83
- | banking77 (72) | 0.327 | 0.138 | 0.297 | 0.433 | 0.542 |
84
- | massive (59) | 0.454 | 0.481 | 0.599 | 0.533 | 0.599 |
85
- | wikitoxic_toxicaggreg (2) | 0.609 | 0.752 | 0.768 | 0.752 | 0.751 |
86
- | wikitoxic_obscene (2) | 0.728 | 0.818 | 0.854 | 0.853 | 0.884 |
87
- | wikitoxic_threat (2) | 0.531 | 0.796 | 0.874 | 0.861 | 0.876 |
88
- | wikitoxic_insult (2) | 0.514 | 0.738 | 0.802 | 0.768 | 0.778 |
89
- | wikitoxic_identityhate (2) | 0.567 | 0.776 | 0.801 | 0.774 | 0.801 |
90
- | hateoffensive (3) | 0.41 | 0.497 | 0.484 | 0.539 | 0.634 |
91
- | hatexplain (3) | 0.373 | 0.423 | 0.385 | 0.441 | 0.446 |
92
- | biasframes_offensive (2) | 0.499 | 0.571 | 0.587 | 0.546 | 0.648 |
93
- | biasframes_sex (2) | 0.503 | 0.703 | 0.845 | 0.794 | 0.877 |
94
- | biasframes_intent (2) | 0.635 | 0.541 | 0.635 | 0.562 | 0.696 |
95
- | agnews (4) | 0.722 | 0.765 | 0.764 | 0.694 | 0.824 |
96
- | yahootopics (10) | 0.303 | 0.55 | 0.621 | 0.575 | 0.605 |
97
- | trueteacher (2) | 0.492 | 0.488 | 0.501 | 0.505 | 0.515 |
98
- | spam (2) | 0.523 | 0.537 | 0.528 | 0.531 | 0.698 |
99
- | wellformedquery (2) | 0.528 | 0.5 | 0.5 | 0.5 | 0.476 |
100
- | manifesto (56) | 0.088 | 0.111 | 0.206 | 0.198 | 0.277 |
101
- | capsotu (21) | 0.375 | 0.525 | 0.558 | 0.543 | 0.631 |
102
-
103
-
104
- Details on the different test datasets are available here: https://github.com/MoritzLaurer/zeroshot-classifier/blob/main/v1_human_data/datasets_overview.csv
 
 
 
 
105
 
106
 
107
  ## When to use which model
108
 
109
- - deberta-v3-zeroshot vs. roberta-zeroshot: deberta-v3 performs clearly better than roberta, but it is slower.
110
  roberta is directly compatible with Hugging Face's production inference TEI containers and flash attention.
111
  These containers are a good choice for production use-cases. tl;dr: For accuracy, use a deberta-v3 model.
112
  If production inference speed is a concern, you can consider a roberta model (e.g. in a TEI container and [HF Inference Endpoints](https://ui.endpoints.huggingface.co/catalog)).
113
- - `zeroshot-v1.1` vs. `zeroshot-v2.0` models: My `zeroshot-v1.1` models (see [Zeroshot Classifier Collection](https://huggingface.co/collections/MoritzLaurer/zeroshot-classifiers-6548b4ff407bb19ff5c3ad6f)))
114
- perform better on these 28 datasets, but they are trained on several datasets with non-commercial licenses.
115
- For commercial users, I therefore recommend using a v2.0 model and non-commercial users might get better performance with a v1.1 model.
 
 
 
 
 
116
  - The latest updates on new models are always available in the [Zeroshot Classifier Collection](https://huggingface.co/collections/MoritzLaurer/zeroshot-classifiers-6548b4ff407bb19ff5c3ad6f).
117
 
118
 
@@ -121,7 +125,6 @@ For commercial users, I therefore recommend using a v2.0 model and non-commercia
121
  Reproduction code is available in the `v2_synthetic_data` directory here: https://github.com/MoritzLaurer/zeroshot-classifier/tree/main
122
 
123
 
124
-
125
  ## Limitations and bias
126
  The model can only do text classification tasks.
127
 
@@ -130,9 +133,8 @@ Biases can come from the underlying foundation model, the human NLI training dat
130
 
131
 
132
  ## License
133
- The foundation model (DeBERTa-v3) is published under the MIT license.
134
- The training data is released under different, permissive, commercially-friendly licenses
135
- ([MNLI](https://huggingface.co/datasets/nyu-mll/multi_nli), [FEVER-NLI](https://huggingface.co/datasets/fever), [synthetic_zeroshot_mixtral_v0.1](https://huggingface.co/datasets/MoritzLaurer/synthetic_zeroshot_mixtral_v0.1))
136
 
137
  ## Citation
138
 
@@ -159,7 +161,6 @@ If you use this model academically, please cite:
159
  If you have questions or ideas for cooperation, contact me at moritz{at}huggingface{dot}co or [LinkedIn](https://www.linkedin.com/in/moritz-laurer/)
160
 
161
 
162
-
163
  ### Flexible usage and "prompting"
164
  You can formulate your own hypotheses by changing the `hypothesis_template` of the zeroshot pipeline.
165
  Similar to "prompt engineering" for LLMs, you can test different formulations of your `hypothesis_template` and verbalized classes to improve performance.
@@ -167,9 +168,14 @@ Similar to "prompt engineering" for LLMs, you can test different formulations of
167
  ```python
168
  from transformers import pipeline
169
  text = "Angela Merkel is a politician in Germany and leader of the CDU"
170
- hypothesis_template = "Merkel is the leader of the party: {}"
171
- classes_verbalized = ["CDU", "SPD", "Greens"]
172
- zeroshot_classifier = pipeline("zero-shot-classification", model="MoritzLaurer/deberta-v3-base-zeroshot-v2.0")
 
 
 
 
 
173
  output = zeroshot_classifier(text, classes_verbalized, hypothesis_template=hypothesis_template, multi_label=False)
174
  print(output)
175
  ```
 
4
  tags:
5
  - text-classification
6
  - zero-shot-classification
7
+ base_model: microsoft/deberta-v3-base
8
  pipeline_tag: zero-shot-classification
9
  library_name: transformers
10
  license: mit
 
 
 
11
  ---
12
 
13
  # Model description: deberta-v3-base-zeroshot-v2.0
 
14
 
 
 
 
15
 
16
+ ## zeroshot-v2.0 series of models
17
+ Models in this series are designed for efficient zeroshot classification with the Hugging Face pipeline.
18
+ These models can do classification without training data and run on both GPUs and CPUs.
19
+ An overview of the latest zeroshot classifiers is available in my [Zeroshot Classifier Collection](https://huggingface.co/collections/MoritzLaurer/zeroshot-classifiers-6548b4ff407bb19ff5c3ad6f).
20
+
21
+ The main advantage of this `zeroshot-v2.0` series of models is that most models are trained on fully commercially-friendly data,
22
+ while my older `zeroshot-v1.1` models included training data with non-commercially licensed data.
23
+
24
+ These models can do one universal classification task: determine whether a hypothesis is "true" or "not true" given a text
25
  (`entailment` vs. `not_entailment`).
26
  This task format is based on the Natural Language Inference task (NLI).
27
  The task is so universal that any classification task can be reformulated into this task.
 
 
28
 
29
 
30
  ## Training data
31
+ These models are trained on two types of fully commercially-friendly data:
32
  1. Synthetic data generated with [Mixtral-8x7B-Instruct-v0.1](https://huggingface.co/mistralai/Mixtral-8x7B-Instruct-v0.1).
33
  I first created a list of 500+ diverse text classification tasks for 25 professions in conversations with Mistral-large. The data was manually curated.
34
+ I then used this as seed data to generate several hundred thousand texts for these tasks with Mixtral-8x7B-Instruct-v0.1.
35
  The final dataset used is available in the [synthetic_zeroshot_mixtral_v0.1](https://huggingface.co/datasets/MoritzLaurer/synthetic_zeroshot_mixtral_v0.1) dataset
36
+ in the subset `mixtral_written_text_for_tasks_v4`. Data curation was done in multiple iterations and will be improved in future iterations.
37
  2. Two commercially-friendly NLI datasets: ([MNLI](https://huggingface.co/datasets/nyu-mll/multi_nli), [FEVER-NLI](https://huggingface.co/datasets/fever)).
38
+ These datasets were added to increase generalization. Datasets like ANLI were excluded due to their non-commercial license.
39
+ 3. Models with "-nc" in the name also included a broader mix of datasets with more restrictive licenses: ANLI, WANLI, LingNLI,
40
+ and all datasets in [this list](https://github.com/MoritzLaurer/zeroshot-classifier/blob/7f82e4ab88d7aa82a4776f161b368cc9fa778001/v1_human_data/datasets_overview.csv) where `used_in_v1.1==True`.
 
 
 
 
 
41
 
42
 
43
+ ## How to use the models
44
  ```python
45
  #!pip install transformers[sentencepiece]
46
  from transformers import pipeline
47
  text = "Angela Merkel is a politician in Germany and leader of the CDU"
48
+ hypothesis_template = "This text is about {}"
49
  classes_verbalized = ["politics", "economy", "entertainment", "environment"]
50
+ zeroshot_classifier = pipeline("zero-shot-classification", model="MoritzLaurer/deberta-v3-large-zeroshot-v2.0-nc") # change the model identifier here
51
  output = zeroshot_classifier(text, classes_verbalised, hypothesis_template=hypothesis_template, multi_label=False)
52
  print(output)
53
  ```
 
57
 
58
  ## Metrics
59
 
60
+ The models were evaluated on 28 different text classification tasks with the [f1_macro](https://scikit-learn.org/stable/modules/generated/sklearn.metrics.f1_score.html) metric.
61
+ The main reference point is `facebook/bart-large-mnli` which is, at the time of writing (03.04.24), the most used commercially-friendly 0-shot classifier.
62
+
63
+ ![results_aggreg_v2.0](https://raw.githubusercontent.com/MoritzLaurer/zeroshot-classifier/main/v2_synthetic_data/results/zeroshot-v2.0-aggreg.png)
64
+
65
+
66
+ | | facebook/bart-large-mnli | roberta-base-zeroshot-v2.0 | roberta-large-zeroshot-v2.0 | deberta-v3-base-zeroshot-v2.0 | deberta-v3-base-zeroshot-v2.0-nc (fewshot) | deberta-v3-large-zeroshot-v2.0 | deberta-v3-large-zeroshot-v2.0-nc (fewshot) | bge-m3-zeroshot-v2.0 | bge-m3-zeroshot-v2.0-nc (fewshot) |
67
+ |:---------------------------|---------------------------:|-----------------------------:|------------------------------:|--------------------------------:|-----------------------------------:|---------------------------------:|------------------------------------:|-----------------------:|--------------------------:|
68
+ | all datasets mean | 0.497 | 0.587 | 0.622 | 0.619 | 0.643 (0.834) | 0.676 | 0.673 (0.846) | 0.59 | (0.803) |
69
+ | amazonpolarity (2) | 0.937 | 0.924 | 0.951 | 0.937 | 0.943 (0.961) | 0.952 | 0.956 (0.968) | 0.942 | (0.951) |
70
+ | imdb (2) | 0.892 | 0.871 | 0.904 | 0.893 | 0.899 (0.936) | 0.923 | 0.918 (0.958) | 0.873 | (0.917) |
71
+ | appreviews (2) | 0.934 | 0.913 | 0.937 | 0.938 | 0.945 (0.948) | 0.943 | 0.949 (0.962) | 0.932 | (0.954) |
72
+ | yelpreviews (2) | 0.948 | 0.953 | 0.977 | 0.979 | 0.975 (0.989) | 0.988 | 0.985 (0.994) | 0.973 | (0.978) |
73
+ | rottentomatoes (2) | 0.83 | 0.802 | 0.841 | 0.84 | 0.86 (0.902) | 0.869 | 0.868 (0.908) | 0.813 | (0.866) |
74
+ | emotiondair (6) | 0.455 | 0.482 | 0.486 | 0.459 | 0.495 (0.748) | 0.499 | 0.484 (0.688) | 0.453 | (0.697) |
75
+ | emocontext (4) | 0.497 | 0.555 | 0.63 | 0.59 | 0.592 (0.799) | 0.699 | 0.676 (0.81) | 0.61 | (0.798) |
76
+ | empathetic (32) | 0.371 | 0.374 | 0.404 | 0.378 | 0.405 (0.53) | 0.447 | 0.478 (0.555) | 0.387 | (0.455) |
77
+ | financialphrasebank (3) | 0.465 | 0.562 | 0.455 | 0.714 | 0.669 (0.906) | 0.691 | 0.582 (0.913) | 0.504 | (0.895) |
78
+ | banking77 (72) | 0.312 | 0.124 | 0.29 | 0.421 | 0.446 (0.751) | 0.513 | 0.567 (0.766) | 0.387 | (0.715) |
79
+ | massive (59) | 0.43 | 0.428 | 0.543 | 0.512 | 0.52 (0.755) | 0.526 | 0.518 (0.789) | 0.414 | (0.692) |
80
+ | wikitoxic_toxicaggreg (2) | 0.547 | 0.751 | 0.766 | 0.751 | 0.769 (0.904) | 0.741 | 0.787 (0.911) | 0.736 | (0.9) |
81
+ | wikitoxic_obscene (2) | 0.713 | 0.817 | 0.854 | 0.853 | 0.869 (0.922) | 0.883 | 0.893 (0.933) | 0.783 | (0.914) |
82
+ | wikitoxic_threat (2) | 0.295 | 0.71 | 0.817 | 0.813 | 0.87 (0.946) | 0.827 | 0.879 (0.952) | 0.68 | (0.947) |
83
+ | wikitoxic_insult (2) | 0.372 | 0.724 | 0.798 | 0.759 | 0.811 (0.912) | 0.77 | 0.779 (0.924) | 0.783 | (0.915) |
84
+ | wikitoxic_identityhate (2) | 0.473 | 0.774 | 0.798 | 0.774 | 0.765 (0.938) | 0.797 | 0.806 (0.948) | 0.761 | (0.931) |
85
+ | hateoffensive (3) | 0.161 | 0.352 | 0.29 | 0.315 | 0.371 (0.862) | 0.47 | 0.461 (0.847) | 0.291 | (0.823) |
86
+ | hatexplain (3) | 0.239 | 0.396 | 0.314 | 0.376 | 0.369 (0.765) | 0.378 | 0.389 (0.764) | 0.29 | (0.729) |
87
+ | biasframes_offensive (2) | 0.336 | 0.571 | 0.583 | 0.544 | 0.601 (0.867) | 0.644 | 0.656 (0.883) | 0.541 | (0.855) |
88
+ | biasframes_sex (2) | 0.263 | 0.617 | 0.835 | 0.741 | 0.809 (0.922) | 0.846 | 0.815 (0.946) | 0.748 | (0.905) |
89
+ | biasframes_intent (2) | 0.616 | 0.531 | 0.635 | 0.554 | 0.61 (0.881) | 0.696 | 0.687 (0.891) | 0.467 | (0.868) |
90
+ | agnews (4) | 0.703 | 0.758 | 0.745 | 0.68 | 0.742 (0.898) | 0.819 | 0.771 (0.898) | 0.687 | (0.892) |
91
+ | yahootopics (10) | 0.299 | 0.543 | 0.62 | 0.578 | 0.564 (0.722) | 0.621 | 0.613 (0.738) | 0.587 | (0.711) |
92
+ | trueteacher (2) | 0.491 | 0.469 | 0.402 | 0.431 | 0.479 (0.82) | 0.459 | 0.538 (0.846) | 0.471 | (0.518) |
93
+ | spam (2) | 0.505 | 0.528 | 0.504 | 0.507 | 0.464 (0.973) | 0.74 | 0.597 (0.983) | 0.441 | (0.978) |
94
+ | wellformedquery (2) | 0.407 | 0.333 | 0.333 | 0.335 | 0.491 (0.769) | 0.334 | 0.429 (0.815) | 0.361 | (0.718) |
95
+ | manifesto (56) | 0.084 | 0.102 | 0.182 | 0.17 | 0.187 (0.376) | 0.258 | 0.256 (0.408) | 0.147 | (0.331) |
96
+ | capsotu (21) | 0.34 | 0.479 | 0.523 | 0.502 | 0.477 (0.664) | 0.603 | 0.502 (0.686) | 0.472 | (0.644) |
97
+
98
+
99
+ These numbers indicate zeroshot performance, as no data from these datasets was added in the training mix.
100
+ Note that models with `-nc` in the title were evaluated twice: one run without any data from the 28 datasets to test pure zeroshot performance (the first number in the respective column) and
101
+ the final run including up to 500 training data points per class from each of the 28 datasets (the second number in brackets in the column). No model was trained on test data.
102
+
103
+ Details on the different datasets are available here: https://github.com/MoritzLaurer/zeroshot-classifier/blob/main/v1_human_data/datasets_overview.csv
104
 
105
 
106
  ## When to use which model
107
 
108
+ - **deberta-v3-zeroshot vs. roberta-zeroshot**: deberta-v3 performs clearly better than roberta, but it is a bit slower.
109
  roberta is directly compatible with Hugging Face's production inference TEI containers and flash attention.
110
  These containers are a good choice for production use-cases. tl;dr: For accuracy, use a deberta-v3 model.
111
  If production inference speed is a concern, you can consider a roberta model (e.g. in a TEI container and [HF Inference Endpoints](https://ui.endpoints.huggingface.co/catalog)).
112
+ - **commercial use-cases**: models with `-nc` in the title were trained on more data and perform better, but include data with non-commercial licenses.
113
+ It's not fully established how this training data affects the license of the trained model. For users with strict legal requirements,
114
+ the models without `-nc` in the title are recommended.
115
+ - **Multilingual/non-English use-cases**: use [bge-m3-zeroshot-v2.0](https://huggingface.co/MoritzLaurer/bge-m3-zeroshot-v2.0) or [bge-m3-zeroshot-v2.0-nc](https://huggingface.co/MoritzLaurer/bge-m3-zeroshot-v2.0-nc).
116
+ Note that multilingual models perform worse than English-only models. You can therefore also first machine translate your texts to English with libraries like [EasyNMT](https://github.com/UKPLab/EasyNMT)
117
+ and then apply any English-only model to the translated data. Machine translation also facilitates validation in case your team does not speak all languages in the data.
118
+ - **context window**: The `bge-m3` models can process up to 8192 tokens. The other models can process up to 512. Note that longer text inputs both make the
119
+ mode slower and decrease performance, so if you're only working with texts of up to 400~ words / 1 page, use e.g. a deberta model for better performance.
120
  - The latest updates on new models are always available in the [Zeroshot Classifier Collection](https://huggingface.co/collections/MoritzLaurer/zeroshot-classifiers-6548b4ff407bb19ff5c3ad6f).
121
 
122
 
 
125
  Reproduction code is available in the `v2_synthetic_data` directory here: https://github.com/MoritzLaurer/zeroshot-classifier/tree/main
126
 
127
 
 
128
  ## Limitations and bias
129
  The model can only do text classification tasks.
130
 
 
133
 
134
 
135
  ## License
136
+ The foundation model is published under the MIT license.
137
+ The the licenses of the training data vary depending on the model, see above.
 
138
 
139
  ## Citation
140
 
 
161
  If you have questions or ideas for cooperation, contact me at moritz{at}huggingface{dot}co or [LinkedIn](https://www.linkedin.com/in/moritz-laurer/)
162
 
163
 
 
164
  ### Flexible usage and "prompting"
165
  You can formulate your own hypotheses by changing the `hypothesis_template` of the zeroshot pipeline.
166
  Similar to "prompt engineering" for LLMs, you can test different formulations of your `hypothesis_template` and verbalized classes to improve performance.
 
168
  ```python
169
  from transformers import pipeline
170
  text = "Angela Merkel is a politician in Germany and leader of the CDU"
171
+ # formulation 1
172
+ hypothesis_template = "This text is about {}"
173
+ classes_verbalized = ["politics", "economy", "entertainment", "environment"]
174
+ # formulation 2
175
+ hypothesis_template = "The topic of this text is {}"
176
+ classes_verbalized = ["politics", "economic policy", "entertainment or music", "environmental protection"]
177
+ # test different formulations
178
+ zeroshot_classifier = pipeline("zero-shot-classification", model="MoritzLaurer/deberta-v3-large-zeroshot-v2.0-nc") # change the model identifier here
179
  output = zeroshot_classifier(text, classes_verbalized, hypothesis_template=hypothesis_template, multi_label=False)
180
  print(output)
181
  ```