Tuana commited on
Commit
653be73
1 Parent(s): b836be0

Updating readme to reflect usage in Haystack

Browse files
Files changed (1) hide show
  1. README.md +29 -37
README.md CHANGED
@@ -33,7 +33,7 @@ model-index:
33
  **Downstream-task:** Extractive QA
34
  **Training data:** SQuAD 2.0
35
  **Eval data:** SQuAD 2.0
36
- **Code:** See [example](https://github.com/deepset-ai/FARM/blob/master/examples/question_answering.py) in [FARM](https://github.com/deepset-ai/FARM/blob/master/examples/question_answering.py)
37
  **Infrastructure**: 1x Tesla v100
38
 
39
  ## Hyperparameters
@@ -68,6 +68,14 @@ Evaluated on the SQuAD 2.0 dev set with the [official eval script](https://works
68
 
69
  ## Usage
70
 
 
 
 
 
 
 
 
 
71
  ### In Transformers
72
  ```python
73
  from transformers import AutoModelForQuestionAnswering, AutoTokenizer, pipeline
@@ -87,34 +95,6 @@ model = AutoModelForQuestionAnswering.from_pretrained(model_name)
87
  tokenizer = AutoTokenizer.from_pretrained(model_name)
88
  ```
89
 
90
- ### In FARM
91
-
92
- ```python
93
- from farm.modeling.adaptive_model import AdaptiveModel
94
- from farm.modeling.tokenization import Tokenizer
95
- from farm.infer import Inferencer
96
-
97
- model_name = "deepset/minilm-uncased-squad2"
98
-
99
- # a) Get predictions
100
- nlp = Inferencer.load(model_name, task_type="question_answering")
101
- QA_input = [{"questions": ["Why is model conversion important?"],
102
- "text": "The option to convert models between FARM and transformers gives freedom to the user and let people easily switch between frameworks."}]
103
- res = nlp.inference_from_dicts(dicts=QA_input)
104
-
105
- # b) Load model & tokenizer
106
- model = AdaptiveModel.convert_from_transformers(model_name, device="cpu", task_type="question_answering")
107
- tokenizer = Tokenizer.load(model_name)
108
- ```
109
-
110
- ### In haystack
111
- For doing QA at scale (i.e. many docs instead of single paragraph), you can load the model also in [haystack](https://github.com/deepset-ai/haystack/):
112
- ```python
113
- reader = FARMReader(model_name_or_path="deepset/minilm-uncased-squad2")
114
- # or
115
- reader = TransformersReader(model="deepset/minilm-uncased-squad2",tokenizer="deepset/minilm-uncased-squad2")
116
- ```
117
-
118
 
119
  ## Authors
120
  **Vaishali Pal:** vaishali.pal@deepset.ai
@@ -124,17 +104,29 @@ reader = TransformersReader(model="deepset/minilm-uncased-squad2",tokenizer="dee
124
  **Tanay Soni:** tanay.soni@deepset.ai
125
 
126
  ## About us
127
- ![deepset logo](https://workablehr.s3.amazonaws.com/uploads/account/logo/476306/logo)
128
- We bring NLP to the industry via open source!
129
- Our focus: Industry specific language models & large scale QA systems.
 
 
 
 
 
 
 
 
130
 
131
- Some of our work:
 
132
  - [German BERT (aka "bert-base-german-cased")](https://deepset.ai/german-bert)
133
  - [GermanQuAD and GermanDPR datasets and models (aka "gelectra-base-germanquad", "gbert-base-germandpr")](https://deepset.ai/germanquad)
134
- - [FARM](https://github.com/deepset-ai/FARM)
135
- - [Haystack](https://github.com/deepset-ai/haystack/)
136
 
137
- Get in touch:
138
- [Twitter](https://twitter.com/deepset_ai) | [LinkedIn](https://www.linkedin.com/company/deepset-ai/) | [Discord](https://haystack.deepset.ai/community/join) | [GitHub Discussions](https://github.com/deepset-ai/haystack/discussions) | [Website](https://deepset.ai)
 
 
 
 
 
139
 
140
  By the way: [we're hiring!](http://www.deepset.ai/jobs)
 
33
  **Downstream-task:** Extractive QA
34
  **Training data:** SQuAD 2.0
35
  **Eval data:** SQuAD 2.0
36
+ **Code:** See an [example QA pipeline on Haystack](https://haystack.deepset.ai/tutorials/01_basic_qa_pipeline)
37
  **Infrastructure**: 1x Tesla v100
38
 
39
  ## Hyperparameters
 
68
 
69
  ## Usage
70
 
71
+ ### In Haystack
72
+ For doing QA at scale (i.e. many docs instead of single paragraph), you can load the model also in [Haystack](https://github.com/deepset-ai/haystack/):
73
+ ```python
74
+ reader = FARMReader(model_name_or_path="deepset/minilm-uncased-squad2")
75
+ # or
76
+ reader = TransformersReader(model="deepset/minilm-uncased-squad2",tokenizer="deepset/minilm-uncased-squad2")
77
+ ```
78
+
79
  ### In Transformers
80
  ```python
81
  from transformers import AutoModelForQuestionAnswering, AutoTokenizer, pipeline
 
95
  tokenizer = AutoTokenizer.from_pretrained(model_name)
96
  ```
97
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
98
 
99
  ## Authors
100
  **Vaishali Pal:** vaishali.pal@deepset.ai
 
104
  **Tanay Soni:** tanay.soni@deepset.ai
105
 
106
  ## About us
107
+ <div class="grid lg:grid-cols-2 gap-x-4 gap-y-3">
108
+ <div class="w-full h-40 object-cover mb-2 rounded-lg flex items-center justify-center">
109
+ <img alt="" src="https://raw.githubusercontent.com/deepset-ai/.github/main/deepset-logo-colored.png" class="w-40"/>
110
+ </div>
111
+ <div class="w-full h-40 object-cover mb-2 rounded-lg flex items-center justify-center">
112
+ <img alt="" src="https://raw.githubusercontent.com/deepset-ai/.github/main/haystack-logo-colored.png" class="w-40"/>
113
+ </div>
114
+ </div>
115
+
116
+ [deepset](http://deepset.ai/) is the company behind the open-source NLP framework [Haystack](https://haystack.deepset.ai/) which is designed to help you build production ready NLP systems that use: Question answering, summarization, ranking etc.
117
+
118
 
119
+ Some of our other work:
120
+ - [Distilled roberta-base-squad2 (aka "tinyroberta-squad2")]([https://huggingface.co/deepset/tinyroberta-squad2)
121
  - [German BERT (aka "bert-base-german-cased")](https://deepset.ai/german-bert)
122
  - [GermanQuAD and GermanDPR datasets and models (aka "gelectra-base-germanquad", "gbert-base-germandpr")](https://deepset.ai/germanquad)
 
 
123
 
124
+ ## Get in touch and join the Haystack community
125
+
126
+ <p>For more info on Haystack, visit our <strong><a href="https://github.com/deepset-ai/haystack">GitHub</a></strong> repo and <strong><a href="https://docs.haystack.deepset.ai">Documentation</a></strong>.
127
+
128
+ We also have a <strong><a class="h-7" href="https://haystack.deepset.ai/community">Discord community open to everyone!</a></strong></p>
129
+
130
+ [Twitter](https://twitter.com/deepset_ai) | [LinkedIn](https://www.linkedin.com/company/deepset-ai/) | [Discord](https://haystack.deepset.ai/community) | [GitHub Discussions](https://github.com/deepset-ai/haystack/discussions) | [Website](https://deepset.ai)
131
 
132
  By the way: [we're hiring!](http://www.deepset.ai/jobs)