Hugging Face's logo

Datasets:
zenml
/
rag_qa_embedding_questions_0_60_0

Modalities:
Text
Formats:
parquet
Size:
1K - 10K
Tags:
Synthetic
distilabel
rlaif
Libraries:
Datasets
pandas
License:
Dataset card Viewer Files Community
11
strickvl commited on
Commit
9ac4bad
·
verified ·
1 Parent(s): 8795549

Upload pipeline.yaml with huggingface_hub

Browse files
Files changed (1) hide show
  1. pipeline.yaml +131 -0
pipeline.yaml ADDED
@@ -0,0 +1,131 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ distilabel:
2
+ version: 1.2.0
3
+ pipeline:
4
+ name: generate_embedding_queries
5
+ description: null
6
+ steps:
7
+ - step:
8
+ name: zenml-docs-0-60-0
9
+ input_mappings: {}
10
+ output_mappings:
11
+ page_content: anchor
12
+ batch_size: 50
13
+ repo_id: zenml/rag_qa_embedding_questions_0_60_0
14
+ split: train
15
+ config: null
16
+ streaming: false
17
+ num_examples: 10
18
+ storage_options: null
19
+ runtime_parameters_info:
20
+ - name: batch_size
21
+ optional: true
22
+ description: The number of rows that will contain the batches generated by
23
+ the step.
24
+ - name: repo_id
25
+ optional: false
26
+ description: The Hugging Face Hub repository ID of the dataset to load.
27
+ - name: split
28
+ optional: true
29
+ description: The split of the dataset to load. Defaults to 'train'.
30
+ - name: config
31
+ optional: true
32
+ description: The configuration of the dataset to load. This is optional and
33
+ only needed if the dataset has multiple configurations.
34
+ - name: streaming
35
+ optional: true
36
+ description: Whether to load the dataset in streaming mode or not. Defaults
37
+ to False.
38
+ - name: num_examples
39
+ optional: true
40
+ description: The number of examples to load from the dataset. By default will
41
+ load all examples.
42
+ type_info:
43
+ module: distilabel.steps.generators.huggingface
44
+ name: LoadDataFromHub
45
+ name: zenml-docs-0-60-0
46
+ - step:
47
+ name: generate_sentence_pair_0
48
+ input_mappings: {}
49
+ output_mappings: {}
50
+ input_batch_size: 10
51
+ llm:
52
+ generation_kwargs:
53
+ temperature: 0.7
54
+ max_new_tokens: 512
55
+ model: gpt-4o
56
+ base_url: https://api.openai.com/v1
57
+ max_retries: 6
58
+ timeout: 120
59
+ structured_output: null
60
+ type_info:
61
+ module: distilabel.llms.openai
62
+ name: OpenAILLM
63
+ group_generations: false
64
+ add_raw_output: true
65
+ num_generations: 1
66
+ triplet: true
67
+ action: query
68
+ context: The text is a chunk from ZenML's technical documentation.
69
+ runtime_parameters_info:
70
+ - name: input_batch_size
71
+ optional: true
72
+ description: The number of rows that will contain the batches processed by
73
+ the step.
74
+ - name: llm
75
+ runtime_parameters_info:
76
+ - name: generation_kwargs
77
+ description: The kwargs to be propagated to either `generate` or `agenerate`
78
+ methods within each `LLM`.
79
+ keys:
80
+ - name: max_new_tokens
81
+ optional: true
82
+ - name: frequency_penalty
83
+ optional: true
84
+ - name: presence_penalty
85
+ optional: true
86
+ - name: temperature
87
+ optional: true
88
+ - name: top_p
89
+ optional: true
90
+ - name: stop
91
+ optional: true
92
+ - name: response_format
93
+ optional: true
94
+ - name: base_url
95
+ optional: true
96
+ description: The base URL to use for the OpenAI API requests.
97
+ - name: api_key
98
+ optional: true
99
+ description: The API key to authenticate the requests to the OpenAI API.
100
+ - name: max_retries
101
+ optional: true
102
+ description: The maximum number of times to retry the request to the API
103
+ before failing.
104
+ - name: timeout
105
+ optional: true
106
+ description: The maximum time in seconds to wait for a response from the
107
+ API.
108
+ - name: structured_output
109
+ optional: true
110
+ description: The structured output format to use across all the generations.
111
+ - name: add_raw_output
112
+ optional: true
113
+ description: Whether to include the raw output of the LLM in the key `raw_output_<TASK_NAME>`
114
+ of the `distilabel_metadata` dictionary output column
115
+ - name: num_generations
116
+ optional: true
117
+ description: The number of generations to be produced per input.
118
+ type_info:
119
+ module: distilabel.steps.tasks.sentence_transformers
120
+ name: GenerateSentencePair
121
+ name: generate_sentence_pair_0
122
+ connections:
123
+ - from: zenml-docs-0-60-0
124
+ to:
125
+ - generate_sentence_pair_0
126
+ - from: generate_sentence_pair_0
127
+ to: []
128
+ routing_batch_functions: []
129
+ type_info:
130
+ module: distilabel.pipeline.local
131
+ name: Pipeline