wip: respond

app.py

@@ -1,13 +1,17 @@
 import gradio as gr
 from huggingface_hub import InferenceClient
-# from transformers import AutoModelForCausalLM, AutoTokenizer
+from transformers import AutoModelForCausalLM, AutoTokenizer
 # import spaces
 
 """
 For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
 """
-client = InferenceClient("BatsResearch/bonito-v1")
+# client = InferenceClient("BatsResearch/bonito-v1")
+model = AutoModelForCausalLM.from_pretrained("BatsResearch/bonito-v1")
+tokenizer = AutoTokenizer.from_pretrained("BatsResearch/bonito-v1")
 
+# move to cuda
+model.to("cuda")
 
 def respond(
     context: str,
@@ -22,8 +26,10 @@ def respond(
                 "\n<|context|>\n" + context.strip() + "\n<|task|>\n"
     )
 
-    for token in client.text_generation(input_text, max_tokens=max_tokens, temperature=temperature, top_p=top_p, stream=True):
-        yield token
+
+
+    # for token in client.text_generation(input_text, max_tokens=max_tokens, temperature=temperature, top_p=top_p, stream=True):
+    #     yield token
 
     # messages = [{"role": "system", "content": system_message}]
 

requirements.txt

@@ -1 +1,3 @@
-huggingface_hub==0.22.2
+huggingface_hub==0.22.2
+transformers
+accelerate