o j6h+>@sddlmZddlZddlmZmZddlmmZddl m Z ddl m Z ddl m Z ddlmZddlmZdd lmZdd lmZdd lmZdd lmZddd dSddZddd dTddZddd dUddZ dVdWd!d"Zd#ddd$dXd(d)Zd#ddd$dYd+d,Zdddddddddd-d.ddd/ dZdBdCZej dVddd d[dHdIZ ejddd d\dKdIZ dVddd d]dNdIZ dOdPZ!  d^d_dQdRZ"dS)`) annotationsN)AnyLiteral)protos) operations)get_default_model_client) model_types) helper_types) operation)protobuf_helpers)field_mask_pb2)flatten_update_pathsclientrequest_optionsnamemodel_types.AnyModelNameOptionsr&helper_types.RequestOptionsType | Nonereturn*model_types.Model | model_types.TunedModelcCsHt|}|drt|||dS|drt|||dStd|)aoCalls the API to fetch a model by name. ``` import pprint model = genai.get_model('models/gemini-1.5-flash') pprint.pprint(model) ``` Args: name: The name of the model to fetch. Should start with `models/` client: The client to use. request_options: Options for the request. Returns: A `types.Model` models/r tunedModels/zWInvalid model name: Model names must start with `models/` or `tunedModels/`. Received: )rmake_model_name startswithget_base_modelget_tuned_model ValueError)rrrri/var/www/html/chefvision.cloud.itp360.com/venv/lib/python3.10/site-packages/google/generativeai/models.py get_model!s   r model_types.BaseModelNameOptionsmodel_types.ModelcCsn|duri}|dur t}t|}|dstd||jdd|i|}t||}tjdi|S)axCalls the API to fetch a base model by name. ``` import pprint model = genai.get_base_model('models/chat-bison-001') pprint.pprint(model) ``` Args: name: The name of the model to fetch. Should start with `models/` client: The client to use. request_options: Options for the request. Returns: A `types.Model`. NrzJInvalid model name: Base model names must start with `models/`. Received: rr) rrrrrrtypeto_dictModelrrrresultrrrrBs  r!model_types.TunedModelNameOptionsmodel_types.TunedModelcCsZ|duri}|dur t}t|}|dstd||jdd|i|}t|S)aCalls the API to fetch a tuned model by name. ``` import pprint model = genai.get_tuned_model('tunedModels/gemini-1.5-flash') pprint.pprint(model) ``` Args: name: The name of the model to fetch. Should start with `tunedModels/` client: The client to use. request_options: Options for the request. Returns: A `types.TunedModel`. NrzPInvalid model name: Tuned model names must start with `tunedModels/`. Received: rr)rrrrrrdecode_tuned_modelr%rrrris   rmodelrglm.ModelServiceClient | NonecCst|tr|drt||d}|j}|S|}|St|tjr$|j}|St|tjr/|j}|St|t jr:|j}|St|t jrNt |dd}|sL|j j}|St d|d)z6Calls the API to fetch the base model name of a model.r)r base_modelNz#Invalid model: The provided model 'z' is not recognized or supported. Supported types are: str, model_types.TunedModel, model_types.Model, protos.Model, and protos.TunedModel.) isinstancestrrrr,r TunedModelr$rrgetattrtuned_model_source TypeError)r*rr,rrrget_base_model_names0        r32) page_sizerrr5 int | Nonemodel_types.ModelsIterableccsZ|duri}|durt}|jdd|i|D]}t||}tjdi|VqdS)aCalls the API to list all available models. ``` import pprint for model in genai.list_models(): pprint.pprint(model) ``` Args: page_size: How many `types.Models` to fetch per page (api call). client: You may pass a `glm.ModelServiceClient` instead of using the default client. request_options: Options for the request. Yields: `types.Model` objects. Nr5r)r list_modelsr"r#rr$r5rrr*rrrr8sr8model_types.TunedModelsIterableccsT|duri}|durt}|jdd|i|D]}t||}t|VqdS)aCalls the API to list all tuned models. ``` import pprint for model in genai.list_tuned_models(): pprint.pprint(model) ``` Args: page_size: How many `types.Models` to fetch per page (api call). client: You may pass a `glm.ModelServiceClient` instead of using the default client. request_options: Options for the request. Yields: `types.TunedModel` objects. Nr5r)rlist_tuned_modelsr"r#rr)r9rrrr;s r; text_inputoutput) id display_name description temperaturetop_ptop_k epoch_count batch_size learning_rate input_key output_keyrr source_model training_datamodel_types.TuningDataOptionsr> str | Noner?r@rA float | NonerBrCrDrErFrGr.rH$operations.CreateTunedModelOperationc  Cs|duri}| dur t} t|}t|}|dr d|i}n|dr-d||di}ntd|dtj|| | d }tj|| | d }tj ||d }tj di|||||||d }| j t ||d fi|}t j|S)aB Calls the API to initiate a tuning process that optimizes a model for specific data, returning an operation object to track and manage the tuning progress. 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 `protos.Dataset`, or * An `Iterable` of: *`protos.TuningExample`, * `{'text_input': text_input, 'output': output}` dicts * `(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. request_options: Options for the request. Returns: A [`google.api_core.operation.Operation`](https://googleapis.dev/python/google-api-core/latest/operation.html) Nrr,rr1) tuned_modelr,z(Invalid model name: The provided model 'zM' does not match any known model patterns such as 'models/' or 'tunedModels/')rGrH)rDrErF)rJhyperparameters)r?r@rArBrC tuning_task)tuned_model_idrOr)rrrr3rrencode_tuning_datarHyperparameters TuningTaskr/create_tuned_modeldictrCreateTunedModelOperationfrom_core_operation)rIrJr>r?r@rArBrCrDrErFrGrHrrsource_model_namebase_model_namerPrQrOr rrrrVsZI        rVrOprotos.TunedModelupdatesNonecCdSNrrOr]rrrrrupdate_tuned_modelsrbdict[str, Any]cCr_r`rrarrrrb~rcstr | protos.TunedModeldict[str, Any] | Nonec Cs"|duri}|dur t}t|trS|}t|ts$tdt|jd|jd d|i|}t|}t }| D]}|j |q:|D] \}}t|||qGn+t|tjrs|duratd|j}|j|d}t|j|j}n tdt|jd|jtj||dfi|} t| S) zeCalls the API to push updates to a specified tuned model where only certain attributes are updatable.NzInvalid argument type: In the function `update_tuned_model(name:str, updates: dict)`, the `updates` argument must be of type `dict`. Received type: .rzInvalid argument: When calling `update_tuned_model(tuned_model:protos.TunedModel, updates=None)`, the `updates` argument must not be set.)rzInvalid argument type: In the function `update_tuned_model(tuned_model:dict|protos.TunedModel)`, the `tuned_model` argument must be of type `dict` or `protos.TunedModel`. Received type: )rO update_maskr)rr-r.rWr2r"__name__rr r FieldMaskkeyspathsappenditems _apply_updaterr/rrr field_mask_pbrbUpdateTunedModelRequestrr)) rOr]rrrrppathvaluewasr&rrrrbsJ        cCs:|d}|ddD]}t||}q t||d|dS)Nrg)splitr0setattr)thingrsrtpartspartrrrros  rocCs<|duri}|dur t}t|}|jdd|i|dS)z/Calls the API to delete a specified tuned modelNrr)rrrdelete_tuned_model)rOrrrrrrr|s  r|)rrrrrr)rr rrrr!)rr'rrrr(r`)r*rrr+)r5r6rr+rrrr7)r5r6rr+rrrr:) rIrrJrKr>rLr?rLr@rLrArMrBrMrCr6rDr6rEr6rFrMrGr.rHr.rr+rrrrN) rOr\r]r^rr+rrrr() rOr.r]rdrr+rrrr() rOrer]rfrr+rrrr()NN)rOr'rr+rrrr^)# __future__rtypingrrgoogle.ai.generativelanguageaigenerativelanguageglmgoogle.generativeairrgoogle.generativeai.clientrgoogle.generativeai.typesrr google.api_corer r google.protobufr google.generativeai.utilsr rrrr3r8r;rVoverloadrbror|rrrrs          $*)  $(~  5