clarifai 10.5.3__py3-none-any.whl → 10.5.4__py3-none-any.whl

clarifai/runners/deepgram_live_transcribe.py

@@ -1,98 +0,0 @@
-import os
-import httpx
-from dotenv import load_dotenv
-import threading
-
-from deepgram import (
-    DeepgramClient,
-    LiveTranscriptionEvents,
-    LiveOptions,
-)
-
-load_dotenv()
-
-# URL for the realtime streaming audio you would like to transcribe
-URL = "http://stream.live.vc.bbcmedia.co.uk/bbc_world_service"
-
-os.environ["DG_API_KEY"] = "1d5666523f0f2fbaf3e2db6ae7717f6f87280b5e"
-API_KEY = os.getenv("DG_API_KEY")
-
-
-def main():
-  try:
-    # STEP 1: Create a Deepgram client using the API key
-    deepgram = DeepgramClient(API_KEY)
-
-    # STEP 2: Create a websocket connection to Deepgram
-    dg_connection = deepgram.listen.live.v("1")
-
-    # STEP 3: Define the event handlers for the connection
-    def on_message(self, result, **kwargs):
-      sentence = result.channel.alternatives[0].transcript
-      if len(sentence) == 0:
-        return
-      print(f"speaker: {sentence}")
-
-    def on_metadata(self, metadata, **kwargs):
-      print(f"\n\n{metadata}\n\n")
-
-    def on_error(self, error, **kwargs):
-      print(f"\n\n{error}\n\n")
-
-    # STEP 4: Register the event handlers
-    dg_connection.on(LiveTranscriptionEvents.Transcript, on_message)
-    dg_connection.on(LiveTranscriptionEvents.Metadata, on_metadata)
-    dg_connection.on(LiveTranscriptionEvents.Error, on_error)
-
-    # STEP 5: Configure Deepgram options for live transcription
-    options = LiveOptions(
-        model="nova-2",
-        language="en-US",
-        smart_format=True,
-    )
-
-    # STEP 6: Start the connection
-    dg_connection.start(options)
-
-    # STEP 7: Create a lock and a flag for thread synchronization
-    lock_exit = threading.Lock()
-    exit = False
-
-    # STEP 8: Define a thread that streams the audio and sends it to Deepgram
-    def myThread():
-      with httpx.stream("GET", URL) as r:
-        for data in r.iter_bytes():
-          lock_exit.acquire()
-          if exit:
-            break
-          lock_exit.release()
-
-          dg_connection.send(data)
-
-    # STEP 9: Start the thread
-    myHttp = threading.Thread(target=myThread)
-    myHttp.start()
-
-    # STEP 10: Wait for user input to stop recording
-    input("Press Enter to stop recording...\n\n")
-
-    # STEP 11: Set the exit flag to True to stop the thread
-    lock_exit.acquire()
-    exit = True
-    lock_exit.release()
-
-    # STEP 12: Wait for the thread to finish
-    myHttp.join()
-
-    # STEP 13: Close the connection to Deepgram
-    dg_connection.finish()
-
-    print("Finished")
-
-  except Exception as e:
-    print(f"Could not open socket: {e}")
-    return
-
-
-if __name__ == "__main__":
-  main()

clarifai/runners/deepgram_live_transcribe.py~

@@ -1,98 +0,0 @@
-import os
-import httpx
-from dotenv import load_dotenv
-import threading
-
-from deepgram import (
-    DeepgramClient,
-    LiveTranscriptionEvents,
-    LiveOptions,
-)
-
-load_dotenv()
-
-# URL for the realtime streaming audio you would like to transcribe
-URL = "http://stream.live.vc.bbcmedia.co.uk/bbc_world_service"
-
-os.environ.set("DG_API_KEY", "1d5666523f0f2fbaf3e2db6ae7717f6f87280b5e")
-API_KEY = os.getenv("DG_API_KEY")
-
-
-def main():
-  try:
-    # STEP 1: Create a Deepgram client using the API key
-    deepgram = DeepgramClient(API_KEY)
-
-    # STEP 2: Create a websocket connection to Deepgram
-    dg_connection = deepgram.listen.live.v("1")
-
-    # STEP 3: Define the event handlers for the connection
-    def on_message(self, result, **kwargs):
-      sentence = result.channel.alternatives[0].transcript
-      if len(sentence) == 0:
-        return
-      print(f"speaker: {sentence}")
-
-    def on_metadata(self, metadata, **kwargs):
-      print(f"\n\n{metadata}\n\n")
-
-    def on_error(self, error, **kwargs):
-      print(f"\n\n{error}\n\n")
-
-    # STEP 4: Register the event handlers
-    dg_connection.on(LiveTranscriptionEvents.Transcript, on_message)
-    dg_connection.on(LiveTranscriptionEvents.Metadata, on_metadata)
-    dg_connection.on(LiveTranscriptionEvents.Error, on_error)
-
-    # STEP 5: Configure Deepgram options for live transcription
-    options = LiveOptions(
-        model="nova-2",
-        language="en-US",
-        smart_format=True,
-    )
-
-    # STEP 6: Start the connection
-    dg_connection.start(options)
-
-    # STEP 7: Create a lock and a flag for thread synchronization
-    lock_exit = threading.Lock()
-    exit = False
-
-    # STEP 8: Define a thread that streams the audio and sends it to Deepgram
-    def myThread():
-      with httpx.stream("GET", URL) as r:
-        for data in r.iter_bytes():
-          lock_exit.acquire()
-          if exit:
-            break
-          lock_exit.release()
-
-          dg_connection.send(data)
-
-    # STEP 9: Start the thread
-    myHttp = threading.Thread(target=myThread)
-    myHttp.start()
-
-    # STEP 10: Wait for user input to stop recording
-    input("Press Enter to stop recording...\n\n")
-
-    # STEP 11: Set the exit flag to True to stop the thread
-    lock_exit.acquire()
-    exit = True
-    lock_exit.release()
-
-    # STEP 12: Wait for the thread to finish
-    myHttp.join()
-
-    # STEP 13: Close the connection to Deepgram
-    dg_connection.finish()
-
-    print("Finished")
-
-  except Exception as e:
-    print(f"Could not open socket: {e}")
-    return
-
-
-if __name__ == "__main__":
-  main()

clarifai/runners/deepgram_runner.py

@@ -1,131 +0,0 @@
-import os
-import time
-import httpx
-from dotenv import load_dotenv
-import threading
-import queue
-import hashlib
-
-from deepgram import (
-    DeepgramClient,
-    LiveTranscriptionEvents,
-    LiveOptions,
-)
-
-load_dotenv()
-
-os.environ["DG_API_KEY"] = "1d5666523f0f2fbaf3e2db6ae7717f6f87280b5e"
-API_KEY = os.getenv("DG_API_KEY")
-
-from clarifai_grpc.grpc.api import resources_pb2, service_pb2
-from collections.abc import Iterator
-from google.protobuf import json_format
-
-from clarifai.client.runner import Runner
-
-
-class MyRunner(Runner):
-  """A custom runner that adds "Hello World" to the end of the text and replaces the domain of the
-  image URL as an example.
-  """
-
-  def setup_connection(self):
-    print("ZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZz")
-    # STEP 2: Create a websocket connection to Deepgram
-    self.dg_connection = self.deepgram.listen.live.v("1")
-
-    output_q = self.output_q
-
-    # STEP 3: Define the event handlers for the connection
-    def on_message(self, result, **kwargs):
-      if result.is_final:
-        sentence = result.channel.alternatives[0].transcript
-        if len(sentence) == 0:
-          return
-        print("SSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSS")
-        print(f"speaker: {sentence}")
-        # put it on a queue as we get responses from deepgram.
-        output_q.put(sentence)
-
-    def on_metadata(self, metadata, **kwargs):
-      print(f"\n\n{metadata}\n\n")
-
-    def on_error(self, error, **kwargs):
-      print(f"\n\n{error}\n\n")
-
-    # STEP 4: Register the event handlers
-    self.dg_connection.on(LiveTranscriptionEvents.Transcript, on_message)
-    self.dg_connection.on(LiveTranscriptionEvents.Metadata, on_metadata)
-    self.dg_connection.on(LiveTranscriptionEvents.Error, on_error)
-
-    # STEP 5: Configure Deepgram options for live transcription
-    self.options = LiveOptions(
-        model="nova-2",
-        language="en-US",
-        smart_format=True,
-    )
-
-    # STEP 6: Start the connection
-    self.dg_connection.start(self.options)
-
-    print("DDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDD")
-    print(self.dg_connection._socket)
-
-  def __init__(self, *args, **kwargs):
-    print("MyRunner init")
-    # STEP 1: Create a Deepgram client using the API key
-    self.deepgram = DeepgramClient(API_KEY)
-    self.output_q = queue.Queue()
-
-    self.setup_connection()
-
-    super().__init__(*args, **kwargs)
-
-  def stream(self, request: service_pb2.PostModelOutputsRequest
-            ) -> Iterator[service_pb2.MultiOutputResponse]:
-    """Example yielding a whole batch of streamed stuff back.
-    """
-
-    assert len(request.inputs) == 1, "This runner only supports one input at a time."
-
-    # Get the next chunk of data from the incoming stream.
-
-    print("Got some audio data")
-
-    data = request.inputs[0].data.audio.base64
-    print(hashlib.md5(data).hexdigest())
-
-    # FIXME(zeiler): this doesnt' work but isn't iportant to our system.
-    if not self.dg_connection._socket:
-      #self.dg_connection.finish()
-      #self.dg_connection.start(self.options)
-      self.setup_connection()
-    self.dg_connection.send(data)
-    print("Sent it to deepgram")
-
-    while True:
-      try:
-        item = self.output_q.get(timeout=0.1)
-        output = resources_pb2.Output()
-        output.data.text.raw = item
-        output.status.code = 10000
-        print("YYYYYYYYYYYYYYYYYYYYYYYYYYYYYY")
-        print("Yielding: ", output.data.text.raw)
-        yield service_pb2.MultiOutputResponse(outputs=[
-            output,
-        ])
-      except queue.Empty:
-        print("Queue is empty, sleeping then breaking")
-        break
-
-  # # STEP 13: Close the connection to Deepgram
-  # dg_connection.finish()
-
-
-if __name__ == '__main__':
-  # Make sure you set these env vars before running the example.
-  # CLARIFAI_PAT
-  # CLARIFAI_USER_ID
-
-  # You need to first create a runner in the Clarifai API and then use the ID here.
-  MyRunner(runner_id="matt-test-runner", base_url="http://q6:32013", num_parallel_polls=1).start()

clarifai/runners/deepgram_runner.py~

@@ -1,130 +0,0 @@
-import os
-import time
-import httpx
-from dotenv import load_dotenv
-import threading
-import queue
-import hashlib
-
-from deepgram import (
-    DeepgramClient,
-    LiveTranscriptionEvents,
-    LiveOptions,
-)
-
-load_dotenv()
-
-os.environ["DG_API_KEY"] = "1d5666523f0f2fbaf3e2db6ae7717f6f87280b5e"
-API_KEY = os.getenv("DG_API_KEY")
-
-from clarifai_grpc.grpc.api import resources_pb2, service_pb2
-from collections.abc import Iterator
-from google.protobuf import json_format
-
-from clarifai.client.runner import Runner
-
-
-class MyRunner(Runner):
-  """A custom runner that adds "Hello World" to the end of the text and replaces the domain of the
-  image URL as an example.
-  """
-
-  def setup_connection(self):
-    print("ZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZz")
-    # STEP 2: Create a websocket connection to Deepgram
-    self.dg_connection = self.deepgram.listen.live.v("1")
-
-    output_q = self.output_q
-
-    # STEP 3: Define the event handlers for the connection
-    def on_message(self, result, **kwargs):
-      if result.is_final:
-        sentence = result.channel.alternatives[0].transcript
-        if len(sentence) == 0:
-          return
-        print("SSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSS")
-        print(f"speaker: {sentence}")
-        # put it on a queue as we get responses from deepgram.
-        output_q.put(sentence)
-
-    def on_metadata(self, metadata, **kwargs):
-      print(f"\n\n{metadata}\n\n")
-
-    def on_error(self, error, **kwargs):
-      print(f"\n\n{error}\n\n")
-
-    # STEP 4: Register the event handlers
-    self.dg_connection.on(LiveTranscriptionEvents.Transcript, on_message)
-    self.dg_connection.on(LiveTranscriptionEvents.Metadata, on_metadata)
-    self.dg_connection.on(LiveTranscriptionEvents.Error, on_error)
-
-    # STEP 5: Configure Deepgram options for live transcription
-    self.options = LiveOptions(
-        model="nova-2",
-        language="en-US",
-        smart_format=True,
-    )
-
-    # STEP 6: Start the connection
-    self.dg_connection.start(self.options)
-
-    print("DDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDD")
-    print(self.dg_connection._socket)
-
-  def __init__(self, *args, **kwargs):
-    print("MyRunner init")
-    # STEP 1: Create a Deepgram client using the API key
-    self.deepgram = DeepgramClient(API_KEY)
-    self.output_q = queue.Queue()
-
-    self.setup_connection()
-
-    super().__init__(*args, **kwargs)
-
-  def stream(self, request: service_pb2.PostModelOutputsRequest
-            ) -> Iterator[service_pb2.MultiOutputResponse]:
-    """Example yielding a whole batch of streamed stuff back.
-    """
-
-    assert len(request.inputs) == 1, "This runner only supports one input at a time."
-
-    # Get the next chunk of data from the incoming stream.
-
-    print("Got some audio data")
-
-    data = request.inputs[0].data.audio.base64
-    print(hashlib.md5(data).hexdigest())
-
-    if not self.dg_connection._socket:
-      self.dg_connection.finish()
-      self.dg_connection.start(self.options)
-      #self.setup_connection()
-    self.dg_connection.send(data)
-    print("Sent it to deepgram")
-
-    while True:
-      try:
-        item = self.output_q.get(timeout=0.1)
-        output = resources_pb2.Output()
-        output.data.text.raw = item
-        output.status.code = 10000
-        print("YYYYYYYYYYYYYYYYYYYYYYYYYYYYYY")
-        print("Yielding: ", output.data.text.raw)
-        yield service_pb2.MultiOutputResponse(outputs=[
-            output,
-        ])
-      except queue.Empty:
-        print("Queue is empty, sleeping then breaking")
-        break
-
-  # # STEP 13: Close the connection to Deepgram
-  # dg_connection.finish()
-
-
-if __name__ == '__main__':
-  # Make sure you set these env vars before running the example.
-  # CLARIFAI_PAT
-  # CLARIFAI_USER_ID
-
-  # You need to first create a runner in the Clarifai API and then use the ID here.
-  MyRunner(runner_id="matt-test-runner", base_url="http://q6:32013", num_parallel_polls=1).start()

clarifai/runners/example_llama2.py~

@@ -1,72 +0,0 @@
-from clarifai.client.runner import Runner
-from clarifai_grpc.grpc.api import resources_pb2
-
-# This example requires to run the following before running this example:
-# pip install transformers
-
-# https://huggingface.co/TheBloke/Llama-2-70B-chat-GPTQ
-model_name_or_path = "TheBloke/Llama-2-7B-chat-GPTQ"
-model_basename = "model"
-
-use_triton = False
-
-
-class Llama2Runner(Runner):
-  """A custom runner that runs the LLama2 LLM.
-  """
-
-  def __init__(self, *args, **kwargs):
-    print("Starting to load the model...")
-    st = time.time()
-    self.tokenizer = AutoTokenizer.from_pretrained(model_name_or_path, use_fast=True)
-    self.model = AutoModelForCausalLM.from_pretrained(model_name_or_path, device_map='auto')
-
-    self.logger.info("Loading model complete in (%f seconds), ready to loop for requests." %
-                     (time.time() - st))
-    super(MyRunner, self).__init__(*args, **kwargs)
-
-  def run_input(self, input: resources_pb2.Input,
-                output_info: resources_pb2.OutputInfo) -> resources_pb2.Output:
-    """This is the method that will be called when the runner is run. It takes in an input and
-    returns an output.
-    """
-
-    output = resources_pb2.Output()
-    data = input.data
-    if data.text.raw != "":
-      input_text = data.text.raw
-    elif data.text.url != "":
-      input_text = str(requests.get(data.text.url).text)
-    else:
-      raise Exception("Need to include data.text.raw or data.text.url in your inputs.")
-
-    st = time.time()
-    max_tokens = 4096
-    # # Method 1
-    # input_ids = self.tokenizer(input_text, return_tensors='pt').input_ids.cuda()
-    # out = self.model.generate(inputs=input_ids, temperature=0.7, max_new_tokens=max_tokens)
-    # out_text = self.tokenizer.decode(out[0], skip_special_tokens=True)
-    # output.data.text.raw = out_text.replace(input_text, '')
-
-    # # Method 2
-    pipe = pipeline(
-        "text-generation",
-        model=self.model,
-        tokenizer=self.tokenizer,
-        max_new_tokens=max_tokens,
-        temperature=0.7,
-        top_p=0.95,
-        repetition_penalty=1.15,
-        return_full_text=False)
-    a = pipe(input_text)
-    output.data.text.raw = a[0]['generated_text']
-    return output
-
-
-if __name__ == '__main__':
-  # Make sure you set these env vars before running the example.
-  # CLARIFAI_PAT
-  # CLARIFAI_USER_ID
-
-  # You need to first create a runner in the Clarifai API and then use the ID here.
-  Llama2Runner(runner_id="sdk-llama2-runner").start()

clarifai/runners/matt_example.py

@@ -1,89 +0,0 @@
-from clarifai_grpc.grpc.api import resources_pb2, service_pb2
-from collections.abc import Iterator
-from google.protobuf import json_format
-
-from clarifai.client.runner import Runner
-
-
-class MyRunner(Runner):
-  """A custom runner that adds "Hello World" to the end of the text and replaces the domain of the
-  image URL as an example.
-  """
-
-  def run_input(self, input: resources_pb2.Input, output_info: resources_pb2.OutputInfo,
-                **kwargs) -> resources_pb2.Output:
-    """This is the method that will be called when the runner is run. It takes in an input and
-    returns an output.
-    """
-
-    output = resources_pb2.Output()
-
-    data = input.data
-
-    # Optional use of output_info
-    params_dict = {}
-    if "params" in output_info:
-      params_dict = output_info["params"]
-
-    if data.text.raw != "":
-      output.data.text.raw = data.text.raw + "Hello World" + params_dict.get(
-          "hello", "") + kwargs.get("extra", "")
-    if data.image.url != "":
-      output.data.text.raw = data.image.url.replace("samples.clarifai.com",
-                                                    "newdomain.com" + params_dict.get("domain",))
-    return output
-
-  def generate(self, request: service_pb2.PostModelOutputsRequest
-              ) -> Iterator[service_pb2.MultiOutputResponse]:
-    """Example yielding a whole batch of streamed stuff back.
-    """
-
-    model = request.model
-    output_info = None
-    if request.model.model_version.id != "":
-      output_info = json_format.MessageToDict(
-          model.model_version.output_info, preserving_proto_field_name=True)
-
-    for i in range(10):  # fake something iterating generating 10 times.
-
-      outputs = []
-      for input in request.inputs:
-        # output = self.run_input(input, output_info, extra=f" {i}")
-        output = resources_pb2.Output()
-        output.data.text.raw = f"Generate Hello World {i}"
-        outputs.append(output)
-      resp = service_pb2.MultiOutputResponse(outputs=outputs,)
-      yield resp
-
-  def stream(self, request: service_pb2.PostModelOutputsRequest
-            ) -> Iterator[service_pb2.MultiOutputResponse]:
-    """Example yielding a whole batch of streamed stuff back.
-    """
-
-    model = request.model
-    output_info = None
-    if request.model.model_version.id != "":
-      output_info = json_format.MessageToDict(
-          model.model_version.output_info, preserving_proto_field_name=True)
-
-    for i in range(10):  # fake something iterating generating 10 times.
-
-      outputs = []
-      for input in request.inputs:
-        # output = self.run_input(input, output_info, extra=f" {i}")
-        output = resources_pb2.Output()
-        out_text = input.data.text.raw + f"Stream Hello World {i}"
-        print(out_text)
-        output.data.text.raw = out_text
-        outputs.append(output)
-      resp = service_pb2.MultiOutputResponse(outputs=outputs,)
-      yield resp
-
-
-if __name__ == '__main__':
-  # Make sure you set these env vars before running the example.
-  # CLARIFAI_PAT
-  # CLARIFAI_USER_ID
-
-  # You need to first create a runner in the Clarifai API and then use the ID here.
-  MyRunner(runner_id="matt-test-runner", base_url="http://q6:32013", num_parallel_polls=1).start()