File size: 22,993 Bytes
8773ff3
4b83e74
8773ff3
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
798f8ba
8773ff3
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
798f8ba
8773ff3
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
798f8ba
8773ff3
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
4b83e74
8773ff3
4b83e74
8773ff3
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
distilabel:
  version: 1.0.1
pipeline:
  name: farming
  description: null
  steps:
  - step:
      name: load_data
      input_mappings: {}
      output_mappings: {}
      batch_size: 64
      data:
      - input: animal welfare from a Family Farming perspective
      - input: animal welfare from a Agribusiness perspective
      - input: animal welfare from a Permaculture perspective
      - input: animal welfare from a Agroforestery perspective
      - input: animal welfare from a Conventional Farming perspective
      - input: economic growth from a Family Farming perspective
      - input: economic growth from a Agribusiness perspective
      - input: economic growth from a Permaculture perspective
      - input: economic growth from a Agroforestery perspective
      - input: economic growth from a Conventional Farming perspective
      - input: land from a Family Farming perspective
      - input: land from a Agribusiness perspective
      - input: land from a Permaculture perspective
      - input: land from a Agroforestery perspective
      - input: land from a Conventional Farming perspective
      - input: resources from a Family Farming perspective
      - input: resources from a Agribusiness perspective
      - input: resources from a Permaculture perspective
      - input: resources from a Agroforestery perspective
      - input: resources from a Conventional Farming perspective
      - input: efficiency from a Family Farming perspective
      - input: efficiency from a Agribusiness perspective
      - input: efficiency from a Permaculture perspective
      - input: efficiency from a Agroforestery perspective
      - input: efficiency from a Conventional Farming perspective
      runtime_parameters_info:
      - name: batch_size
        optional: true
        description: The number of rows that will contain the batches generated by
          the step.
      type_info:
        module: distilabel.steps.generators.data
        name: LoadDataFromDicts
    name: load_data
  - step:
      name: self-instruct
      input_mappings: {}
      output_mappings: {}
      input_batch_size: 8
      llm:
        generation_kwargs: {}
        model_id: null
        endpoint_name: null
        endpoint_namespace: null
        base_url: https://api-inference.huggingface.co/models/microsoft/phi-2
        tokenizer_id: null
        model_display_name: null
        use_openai_client: false
        type_info:
          module: distilabel.llms.huggingface.inference_endpoints
          name: InferenceEndpointsLLM
      group_generations: false
      num_generations: 1
      num_instructions: 5
      criteria_for_query_generation: 'Incorporate a diverse range of verbs, avoiding
        repetition.

        Ensure queries are compatible with AI model''s text generation functions and
        are limited to 1-2 sentences.

        Design queries to be self-contained and standalone.

        Blend interrogative (e.g., "What is the significance of x?") and imperative
        (e.g., "Detail the process of x.") styles.'
      application_description: 'You are an AI assistant than generates queries around
        the domain of farming.

        Your should not expect basic but profound questions from your users.

        The queries should reflect a diversity of vision and economic positions and
        political positions.

        The queries may know about different methods of farming.

        The queries can be positioned politically, economically, socially, or practically.

        Also take into account the impact of diverse causes on diverse domains.'
      runtime_parameters_info:
      - name: input_batch_size
        optional: true
        description: The number of rows that will contain the batches processed by
          the step.
      - name: llm
        runtime_parameters_info:
        - name: generation_kwargs
          description: The kwargs to be propagated to either `generate` or `agenerate`
            methods within each `LLM`.
          keys:
          - name: max_new_tokens
            optional: true
            description: the maximum number of new tokens that the model will generate.  Defaults
              to `128`.
          - name: frequency_penalty
            optional: true
            description: the repetition penalty to use for the generation. Defaults  to
              `0.0`. Only applies if `use_openai_client=True`.
          - name: presence_penalty
            optional: true
            description: the presence penalty to use for the generation. Defaults
              to  `0.0`. Only applies if `use_openai_client=True`.
          - name: repetition_penalty
            optional: true
            description: the repetition penalty to use for the generation. Defaults  to
              `None`. Only applies if `use_openai_client=False`.
          - name: temperature
            optional: true
            description: the temperature to use for the generation. Defaults to `1.0`.
          - name: do_sample
            optional: true
            description: whether to use sampling for the generation. Defaults to `False`.  Only
              applies if `use_openai_client=False`.
          - name: top_k
            optional: true
            description: the top-k value to use for the generation. Defaults to `0.8`,
              since neither  `0.0` nor `1.0` are valid values in TGI.
          - name: top_p
            optional: true
            description: the top-p value to use for the generation. Defaults to `1.0`.
          - name: typical_p
            optional: true
            description: the typical-p value to use for the generation. Defaults to
              `0.5`.
        - name: endpoint_name
          optional: true
          description: The name of the Inference Endpoint to use for the LLM.
        - name: endpoint_namespace
          optional: true
          description: The namespace of the Inference Endpoint to use for the LLM.
        - name: base_url
          optional: true
          description: The base URL to use for the Inference Endpoints API requests.
        - name: api_key
          optional: true
          description: The API key to authenticate the requests to the Inference Endpoints
            API.
      - name: num_generations
        optional: true
        description: The number of generations to be produced per input.
      type_info:
        module: distilabel.steps.tasks.self_instruct
        name: SelfInstruct
    name: self-instruct
  - step:
      name: evol_instruction_complexity
      input_mappings:
        instruction: question
      output_mappings: {}
      input_batch_size: 8
      llm:
        generation_kwargs: {}
        model_id: null
        endpoint_name: null
        endpoint_namespace: null
        base_url: https://api-inference.huggingface.co/models/microsoft/phi-2
        tokenizer_id: null
        model_display_name: null
        use_openai_client: false
        type_info:
          module: distilabel.llms.huggingface.inference_endpoints
          name: InferenceEndpointsLLM
      group_generations: false
      num_generations: 1
      num_evolutions: 2
      store_evolutions: true
      generate_answers: false
      include_original_instruction: true
      mutation_templates:
        CONSTRAINTS: "I want you act as a Prompt Rewriter.\n\nYour objective is to\
          \ rewrite a given prompt into a more complex version to make those famous\
          \ AI systems (e.g., chatgpt and GPT4) a bit harder to handle.\n\nBut the\
          \ rewritten prompt must be reasonable and must be understood and responded\
          \ by humans.\n\nYour rewriting cannot omit the non-text parts such as the\
          \ table and code in #The Given Prompt#:. Also, please do not omit the input\
          \ in #The Given Prompt#.\n\nYou SHOULD complicate the given prompt using\
          \ the following method: \nPlease add one more constraints/requirements into\
          \ '#The Given Prompt#'\n\nYou should try your best not to make the #Rewritten\
          \ Prompt# become verbose, #Rewritten Prompt# can only add 10 to 20 words\
          \ into #The Given Prompt#.\n\n'#The Given Prompt#', '#Rewritten Prompt#',\
          \ 'given prompt' and 'rewritten prompt' are not allowed to appear in #Rewritten\
          \ Prompt#\n\n#The Given Prompt#:\n<PROMPT>\n#Rewritten Prompt#:\n\n"
        DEEPENING: "I want you act as a Prompt Rewriter.\n\nYour objective is to rewrite\
          \ a given prompt into a more complex version to make those famous AI systems\
          \ (e.g., chatgpt and GPT4) a bit harder to handle.\n\nBut the rewritten\
          \ prompt must be reasonable and must be understood and responded by humans.\n\
          \nYour rewriting cannot omit the non-text parts such as the table and code\
          \ in #The Given Prompt#:. Also, please do not omit the input in #The Given\
          \ Prompt#.\n\nYou SHOULD complicate the given prompt using the following\
          \ method: \nIf #The Given Prompt# contains inquiries about certain issues,\
          \ the depth and breadth of the inquiry can be increased.\n\nYou should try\
          \ your best not to make the #Rewritten Prompt# become verbose, #Rewritten\
          \ Prompt# can only add 10 to 20 words into #The Given Prompt#.\n\n'#The\
          \ Given Prompt#', '#Rewritten Prompt#', 'given prompt' and 'rewritten prompt'\
          \ are not allowed to appear in #Rewritten Prompt#\n\n#The Given Prompt#:\n\
          <PROMPT>\n#Rewritten Prompt#:\n\n"
        CONCRETIZING: "I want you act as a Prompt Rewriter.\n\nYour objective is to\
          \ rewrite a given prompt into a more complex version to make those famous\
          \ AI systems (e.g., chatgpt and GPT4) a bit harder to handle.\n\nBut the\
          \ rewritten prompt must be reasonable and must be understood and responded\
          \ by humans.\n\nYour rewriting cannot omit the non-text parts such as the\
          \ table and code in #The Given Prompt#:. Also, please do not omit the input\
          \ in #The Given Prompt#.\n\nYou SHOULD complicate the given prompt using\
          \ the following method: \nPlease replace general concepts with more specific\
          \ concepts.\n\nYou should try your best not to make the #Rewritten Prompt#\
          \ become verbose, #Rewritten Prompt# can only add 10 to 20 words into #The\
          \ Given Prompt#.\n\n'#The Given Prompt#', '#Rewritten Prompt#', 'given prompt'\
          \ and 'rewritten prompt' are not allowed to appear in #Rewritten Prompt#\n\
          \n#The Given Prompt#:\n<PROMPT>\n#Rewritten Prompt#:\n\n"
        INCREASED_REASONING_STEPS: "I want you act as a Prompt Rewriter.\n\nYour objective\
          \ is to rewrite a given prompt into a more complex version to make those\
          \ famous AI systems (e.g., chatgpt and GPT4) a bit harder to handle.\n\n\
          But the rewritten prompt must be reasonable and must be understood and responded\
          \ by humans.\n\nYour rewriting cannot omit the non-text parts such as the\
          \ table and code in #The Given Prompt#:. Also, please do not omit the input\
          \ in #The Given Prompt#.\n\nYou SHOULD complicate the given prompt using\
          \ the following method: \nIf #The Given Prompt# can be solved with just\
          \ a few simple thinking processes, you can rewrite it to explicitly request\
          \ multiple-step reasoning.\n\nYou should try your best not to make the #Rewritten\
          \ Prompt# become verbose, #Rewritten Prompt# can only add 10 to 20 words\
          \ into #The Given Prompt#.\n\n'#The Given Prompt#', '#Rewritten Prompt#',\
          \ 'given prompt' and 'rewritten prompt' are not allowed to appear in #Rewritten\
          \ Prompt#\n\n#The Given Prompt#:\n<PROMPT>\n#Rewritten Prompt#:\n\n"
        BREADTH: 'I want you act as a Prompt Creator.


          Your goal is to draw inspiration from the #Given Prompt# to create a brand
          new prompt.


          This new prompt should belong to the same domain as the #Given Prompt# but
          be even more rare.


          The LENGTH and complexity of the #Created Prompt# should be similar to that
          of the #Given Prompt#.


          The #Created Prompt# must be reasonable and must be understood and responded
          by humans.


          ''#Given Prompt#'', ''#Created Prompt#'', ''given prompt'' and ''created
          prompt'' are not allowed to appear in #Created Prompt#


          #Given Prompt#:

          <PROMPT>

          #Created Prompt#:


          '
      seed: 42
      runtime_parameters_info:
      - name: input_batch_size
        optional: true
        description: The number of rows that will contain the batches processed by
          the step.
      - name: llm
        runtime_parameters_info:
        - name: generation_kwargs
          description: The kwargs to be propagated to either `generate` or `agenerate`
            methods within each `LLM`.
          keys:
          - name: max_new_tokens
            optional: true
            description: the maximum number of new tokens that the model will generate.  Defaults
              to `128`.
          - name: frequency_penalty
            optional: true
            description: the repetition penalty to use for the generation. Defaults  to
              `0.0`. Only applies if `use_openai_client=True`.
          - name: presence_penalty
            optional: true
            description: the presence penalty to use for the generation. Defaults
              to  `0.0`. Only applies if `use_openai_client=True`.
          - name: repetition_penalty
            optional: true
            description: the repetition penalty to use for the generation. Defaults  to
              `None`. Only applies if `use_openai_client=False`.
          - name: temperature
            optional: true
            description: the temperature to use for the generation. Defaults to `1.0`.
          - name: do_sample
            optional: true
            description: whether to use sampling for the generation. Defaults to `False`.  Only
              applies if `use_openai_client=False`.
          - name: top_k
            optional: true
            description: the top-k value to use for the generation. Defaults to `0.8`,
              since neither  `0.0` nor `1.0` are valid values in TGI.
          - name: top_p
            optional: true
            description: the top-p value to use for the generation. Defaults to `1.0`.
          - name: typical_p
            optional: true
            description: the typical-p value to use for the generation. Defaults to
              `0.5`.
        - name: endpoint_name
          optional: true
          description: The name of the Inference Endpoint to use for the LLM.
        - name: endpoint_namespace
          optional: true
          description: The namespace of the Inference Endpoint to use for the LLM.
        - name: base_url
          optional: true
          description: The base URL to use for the Inference Endpoints API requests.
        - name: api_key
          optional: true
          description: The API key to authenticate the requests to the Inference Endpoints
            API.
      - name: num_generations
        optional: true
        description: The number of generations to be produced per input.
      - name: seed
        optional: true
        description: As `numpy` is being used in order to randomly pick a mutation
          method, then is nice to seed a random seed.
      type_info:
        module: distilabel.steps.tasks.evol_instruct.base
        name: EvolInstruct
    name: evol_instruction_complexity
  - step:
      name: expand_columns
      input_mappings: {}
      output_mappings: {}
      input_batch_size: 50
      columns:
        instructions: question
      runtime_parameters_info:
      - name: input_batch_size
        optional: true
        description: The number of rows that will contain the batches processed by
          the step.
      type_info:
        module: distilabel.steps.expand
        name: ExpandColumns
    name: expand_columns
  - step:
      name: clean_numbered_list
      input_mappings: {}
      output_mappings: {}
      input_batch_size: 50
      runtime_parameters_info:
      - name: input_batch_size
        optional: true
        description: The number of rows that will contain the batches processed by
          the step.
      type_info:
        module: domain
        name: CleanNumberedList
    name: clean_numbered_list
  - step:
      name: expand_columns_evolved
      input_mappings: {}
      output_mappings: {}
      input_batch_size: 50
      columns:
        evolved_instructions: evolved_questions
      runtime_parameters_info:
      - name: input_batch_size
        optional: true
        description: The number of rows that will contain the batches processed by
          the step.
      type_info:
        module: distilabel.steps.expand
        name: ExpandColumns
    name: expand_columns_evolved
  - step:
      name: domain_expert
      input_mappings:
        instruction: evolved_questions
      output_mappings:
        generation: domain_expert_answer
      input_batch_size: 8
      llm:
        generation_kwargs: {}
        model_id: null
        endpoint_name: null
        endpoint_namespace: null
        base_url: https://api-inference.huggingface.co/models/microsoft/phi-2
        tokenizer_id: null
        model_display_name: null
        use_openai_client: false
        type_info:
          module: distilabel.llms.huggingface.inference_endpoints
          name: InferenceEndpointsLLM
      group_generations: false
      num_generations: 1
      runtime_parameters_info:
      - name: input_batch_size
        optional: true
        description: The number of rows that will contain the batches processed by
          the step.
      - name: llm
        runtime_parameters_info:
        - name: generation_kwargs
          description: The kwargs to be propagated to either `generate` or `agenerate`
            methods within each `LLM`.
          keys:
          - name: max_new_tokens
            optional: true
            description: the maximum number of new tokens that the model will generate.  Defaults
              to `128`.
          - name: frequency_penalty
            optional: true
            description: the repetition penalty to use for the generation. Defaults  to
              `0.0`. Only applies if `use_openai_client=True`.
          - name: presence_penalty
            optional: true
            description: the presence penalty to use for the generation. Defaults
              to  `0.0`. Only applies if `use_openai_client=True`.
          - name: repetition_penalty
            optional: true
            description: the repetition penalty to use for the generation. Defaults  to
              `None`. Only applies if `use_openai_client=False`.
          - name: temperature
            optional: true
            description: the temperature to use for the generation. Defaults to `1.0`.
          - name: do_sample
            optional: true
            description: whether to use sampling for the generation. Defaults to `False`.  Only
              applies if `use_openai_client=False`.
          - name: top_k
            optional: true
            description: the top-k value to use for the generation. Defaults to `0.8`,
              since neither  `0.0` nor `1.0` are valid values in TGI.
          - name: top_p
            optional: true
            description: the top-p value to use for the generation. Defaults to `1.0`.
          - name: typical_p
            optional: true
            description: the typical-p value to use for the generation. Defaults to
              `0.5`.
        - name: endpoint_name
          optional: true
          description: The name of the Inference Endpoint to use for the LLM.
        - name: endpoint_namespace
          optional: true
          description: The namespace of the Inference Endpoint to use for the LLM.
        - name: base_url
          optional: true
          description: The base URL to use for the Inference Endpoints API requests.
        - name: api_key
          optional: true
          description: The API key to authenticate the requests to the Inference Endpoints
            API.
      - name: num_generations
        optional: true
        description: The number of generations to be produced per input.
      type_info:
        module: domain
        name: DomainExpert
    name: domain_expert
  - step:
      name: keep_columns
      input_mappings: {}
      output_mappings: {}
      input_batch_size: 50
      columns:
      - model_name
      - evolved_questions
      - domain_expert_answer
      runtime_parameters_info:
      - name: input_batch_size
        optional: true
        description: The number of rows that will contain the batches processed by
          the step.
      type_info:
        module: distilabel.steps.keep
        name: KeepColumns
    name: keep_columns
  - step:
      name: text_generation_to_argilla
      input_mappings:
        instruction: evolved_questions
        generation: domain_expert_answer
      output_mappings: {}
      input_batch_size: 50
      dataset_name: test_3
      dataset_workspace: admin
      api_url: https://burtenshaw-test-3-argilla-space.hf.space
      runtime_parameters_info:
      - name: input_batch_size
        optional: true
        description: The number of rows that will contain the batches processed by
          the step.
      - name: dataset_name
        optional: false
        description: The name of the dataset in Argilla.
      - name: dataset_workspace
        optional: true
        description: The workspace where the dataset will be created in Argilla. Defaultsto
          `None` which means it will be created in the default workspace.
      - name: api_url
        optional: true
        description: The base URL to use for the Argilla API requests.
      - name: api_key
        optional: true
        description: The API key to authenticate the requests to the Argilla API.
      type_info:
        module: distilabel.steps.argilla.text_generation
        name: TextGenerationToArgilla
    name: text_generation_to_argilla
  connections:
  - from: load_data
    to:
    - self-instruct
  - from: self-instruct
    to:
    - expand_columns
  - from: evol_instruction_complexity
    to:
    - expand_columns_evolved
  - from: expand_columns
    to:
    - clean_numbered_list
  - from: clean_numbered_list
    to:
    - evol_instruction_complexity
  - from: expand_columns_evolved
    to:
    - domain_expert
  - from: domain_expert
    to:
    - keep_columns
  - from: keep_columns
    to:
    - text_generation_to_argilla
  - from: text_generation_to_argilla
    to: []
  type_info:
    module: distilabel.pipeline.local
    name: Pipeline