阺hx4 |ddlmZddlZddlmZmZddlmcmZddl m Z ddl m Z ddl mZddlmZddlmZdd lmZdd dEdZdd dFdZdd dGdZ dHdIdZddddJdZddddKd!Zdddddddddd"d#dd$ dLd7Zej dHdd dMd<Zejdd dNd>Z dHdd dOdAZdBZdCZ dHdPdDZdS)Q) annotationsN)AnyLiteral) operations)get_default_model_client) model_types) operation)protobuf_helpers)field_mask_pb2clientnamemodel_types.AnyModelNameOptionsreturn*model_types.Model | model_types.TunedModelctj|}|drt||S|drt ||St d)aYGiven a model name, fetch the `types.Model` or `types.TunedModel` object. ``` import pprint model = genai.get_tuned_model(model_name): pprint.pprint(model) ``` Args: name: The name of the model to fetch. client: The client to use. Returns: A `types.Model` or `types.TunedModel` object. models/r tunedModels/z7Model names must start with `models/` or `tunedModels/`)rmake_model_name startswithget_base_modelget_tuned_model ValueError)rr s b/var/www/html/services/rag-service/venv/lib/python3.11/site-packages/google/generativeai/models.py get_modelrss$  &t , ,D y!!Td62222  ( (TtF3333RSSS model_types.BaseModelNameOptionsmodel_types.Modelc*|t}tj|}|dst d|||}t ||}tjdi|S)a.Get the `types.Model` for the given base model name. ``` import pprint model = genai.get_model('models/chat-bison-001'): pprint.pprint(model) ``` Args: name: The name of the model to fetch. client: The client to use. Returns: A `types.Model`. Nrz1Base model names must start with `models/`, got: r) rrrrrrtypeto_dictModelrr results rrr8s ~)++  &t , ,D ??9 % %USTSSTTT   4  ( (F &\\ ! !& ) )F   & &v & &&r!model_types.TunedModelNameOptionsmodel_types.TunedModelc|t}tj|}|dst d||}tj|S)aCGet the `types.TunedModel` for the given tuned model name. ``` import pprint model = genai.get_tuned_model('tunedModels/my-model-1234'): pprint.pprint(model) ``` Args: name: The name of the model to fetch. client: The client to use. Returns: A `types.TunedModel`. Nrz0Tuned model names must start with `tunedModels/`r )rrrrrrdecode_tuned_modelr%s rrrTso$~)++  &t , ,D ??> * *MKLLL  # # # . .F  )& 1 11rmodelr glm.ModelServiceClient | Nonect|tr1|drt||}|j}n|}nt|t jr|j}nt|t jr|j}nnt|tjr|j}nLt|tjr t|dd}|s |j j}ntd||S)Nrr base_modelzCannot understand model: ) isinstancestrrrr.r TunedModelr$rglmgetattrtuned_model_source TypeError)r+r r.s rget_base_model_namer6ss%=   N + + eF333E)JJJJ E;1 2 2 =% E;, - - =Z E39 % %=Z E3> * *=UL$77  =1s rrArAsu(~)++))I)>>44U ##E**,U33333344r text_inputoutput) id display_name description temperaturetop_ptop_k epoch_count batch_size learning_rate input_key output_keyr source_model training_datamodel_types.TuningDataOptionsrD str | NonerErFrG float | NonerHrIrJrKrLrMr0rN$operations.CreateTunedModelOperationc >| t} tj|}t|}|drd|i}n0|drd||di}nt d|dtj|| | }tj|| | }tj || }tj di|||||||d }| t|| }tj|S)a Launches a tuning job to create a TunedModel. Since tuning a model can take significant time, this API doesn't wait for the tuning to complete. Instead, it returns a `google.api_core.operation.Operation` object that lets you check on the status of the tuning job, or wait for it to complete, and check the result. After the job completes you can either find the resulting `TunedModel` object in `Operation.result()` or `palm.list_tuned_models` or `palm.get_tuned_model(model_id)`. ``` my_id = "my-tuned-model-id" operation = palm.create_tuned_model( id = my_id, source_model="models/text-bison-001", training_data=[{'text_input': 'example input', 'output': 'example output'},...] ) tuned_model=operation.result() # Wait for tuning to finish palm.generate_text(f"tunedModels/{my_id}", prompt="...") ``` Args: source_model: The name of the model to tune. training_data: The dataset to tune the model on. This must be either: * A `glm.Dataset`, or * An `Iterable` of: *`glm.TuningExample`, * {'text_input': text_input, 'output': output} dicts, or * `(text_input, output)` tuples. * A `Mapping` of `Iterable[str]` - use `input_key` and `output_key` to choose which columns to use as the input/output * A csv file (will be read with `pd.read_csv` and handles as a `Mapping` above). This can be: * A local path as a `str` or `pathlib.Path`. * A url for a csv file. * The url of a Google Sheets file. * A JSON file - Its contents will be handled either as an `Iterable` or `Mapping` above. This can be: * A local path as a `str` or `pathlib.Path`. id: The model identifier, used to refer to the model in the API `tunedModels/{id}`. Must be unique. display_name: A human-readable name for display. description: A description of the tuned model. temperature: The default temperature for the tuned model, see `types.Model` for details. top_p: The default `top_p` for the model, see `types.Model` for details. top_k: The default `top_k` for the model, see `types.Model` for details. epoch_count: The number of tuning epochs to run. An epoch is a pass over the whole dataset. batch_size: The number of examples to use in each training batch. learning_rate: The step size multiplier for the gradient updates. client: Which client to use. Returns: A [`google.api_core.operation.Operation`](https://googleapis.dev/python/google-api-core/latest/operation.html) Nrr.rr4) tuned_modelr.zNot understood: `source_model=`)rMrN)rJrKrL)rPhyperparameters)rErFrGrHrI tuning_task)tuned_model_idrVr!)rrrr6rrencode_tuning_datar2Hyperparameters TuningTaskr1create_tuned_modeldictrCreateTunedModelOperationfrom_core_operation)rOrPrDrErFrGrHrIrJrKrLrMrNr source_model_namebase_model_namerXrYrVr s rr^r^s|P~)++#3LAA),77O##I.. 9$&78  % %n 5 59 0-##  7|7778882zM)#O .#'K . !K))$bk*Z*Z*Z[[I  / C CI N NNrrVglm.TunedModelupdatesNonecdSNr!rVrer s rupdate_tuned_modelrj:  Drdict[str, Any]cdSrhr!ris rrjrjDrkrstr | glm.TunedModeldict[str, Any] | Nonecl|t}t|tr|}t|tst dt |||}t|}tj }| D]}|j || D]\}}t|||nt|tjrN|t#d|j}||}t'j|j|j}n t dt |d|tj||}t1j|S)zGPush updates to the tuned model. Only certain attributes are updatable.Nz]When calling `update_tuned_model(name:str, updates: dict)`, `updates` must be a `dict`. got: r zfWhen calling `update_tuned_model(tuned_model:glm.TunedModel, updates=None)`,`updates` must not be set.zxFor `update_tuned_model(tuned_model:dict|glm.TunedModel)`,`tuned_model` must be a `dict` or a `glm.TunedModel`. Got a: `rW)rV update_mask)rr/r0r_r5r"r_flatten_update_pathsr FieldMaskkeyspathsappenditems _apply_updater2r1rrr field_mask_pbrjUpdateTunedModelRequestrr*) rVrer rrypathvaluewasr&s rrjrjNs~)+++s## '4(( (W ((  ,,$,77 '00#-// LLNN * *D   # #D ) ) ) )"==?? 4 4KD% +tU 3 3 3 3 4 K 0 0  -  $$$$//%0+/JJ  bMQR]M^M^ b b b    & & # TTTF  )& 1 11rci}|D]Q\}}t|tr2t|D]\}}|||d|<L|||<R|S)N.)rwr/r_rr)re new_updateskeyr}sub_key sub_values rrrrr~sKmmoo%% U eT " " %&;E&B&B&H&H&J&J < <"2; s..W..// < %K   rc|d}|ddD]}t||}t||d|dS)Nr)splitr3setattr)thingr|r}partsparts rrxrxsU JJsOOEcrc %%t$$ E59e$$$$$rcz|t}tj|}||dS)Nr )rrrdelete_tuned_model)rVr rs rrrsA~)++  &{ 3 3D 4(((((r)rrrr)rrrr)rr'rr(rh)r+rr r,)r8r9r r,rr:)r8r9r r,rr?)rOrrPrQrDrRrErRrFrRrGrSrHrSrIr9rJr9rKr9rLrSrMr0rNr0r r,rrT)rVrdrerfr r,rr()rVr0rerlr r,rr()rVrnreror r,rr()rVr'r r,rrf) __future__rtypingrrgoogle.ai.generativelanguageaigenerativelanguager2google.generativeairgoogle.generativeai.clientrgoogle.generativeai.typesrgoogle.api_corer r google.protobufr rrrr6r=rAr^overloadrjrrrxrr!rrrs#""""" ***************??????111111%%%%%%,,,,,,******6:TTTTTT6FJ'''''':8<222222@UY6,0))))))>,0444444@#" $"!"&!,0rOrOrOrOrOrOj -1        -1       &*-2-1 -2-2-2-2-2-2`   %%%-1)))))))r