Upload folder using huggingface_hub

app.py

@@ -1,31 +1,17 @@
-from langchain.prompts import PromptTemplate
-from langchain.llms import CTransformers
-import os
+from transformers import AutoModelForSeq2SeqLM, AutoTokenizer
 import gradio as gr
 
-def GetLlamaResponse(topic):
-    llm = CTransformers(
-        model_type="google-flan",
-        model="tf_model.h5",
-        config={"max_new_tokens": 64, "temperature": 0.75},
-    )
-    template = """
-    You are a helpful assistant
-    """
-    prompt = PromptTemplate(
-        input_variables=["topic", "word_count", "temperature"],
-        template=template,
-    )
-
-    response = llm(
-        prompt.format(
-            word_count=64,
-            temperature=0.4,
-            topic=topic,
-        )
-    )
-
-    return response
+model = AutoModelForSeq2SeqLM.from_pretrained("google/flan-t5-base")
+tokenizer = AutoTokenizer.from_pretrained("google/flan-t5-base")
+
+def index(prompt):
+    inputs = tokenizer(prompt, return_tensors="pt")
+    outputs = model.generate(**inputs)
+
+    print(tokenizer.batch_decode(outputs, skip_special_tokens=True))
+
+    return tokenizer.batch_decode(outputs, skip_special_tokens=True)
+
 
 inputs_image_url = [
     gr.Textbox(type="text", label="Topic Name"),
@@ -36,7 +22,7 @@ outputs_result_dict = [
 ]
 
 interface_image_url = gr.Interface(
-    fn=GetLlamaResponse,
+    fn=index,
     inputs=inputs_image_url,
     outputs=outputs_result_dict,
     title="Text Generation",

app1.py

@@ -0,0 +1,49 @@
+from langchain.prompts import PromptTemplate
+from langchain.llms import CTransformers
+import os
+import gradio as gr
+
+def GetLlamaResponse(topic):
+    llm = CTransformers(
+        model_type="google-flan",
+        model="tf_model.h5",
+        config={"max_new_tokens": 64, "temperature": 0.75},
+    )
+    template = """
+    You are a helpful assistant
+    """
+    prompt = PromptTemplate(
+        input_variables=["topic", "word_count", "temperature"],
+        template=template,
+    )
+
+    response = llm(
+        prompt.format(
+            word_count=64,
+            temperature=0.4,
+            topic=topic,
+        )
+    )
+
+    return response
+
+inputs_image_url = [
+    gr.Textbox(type="text", label="Topic Name"),
+]
+
+outputs_result_dict = [
+    gr.Textbox(type="text", label="Result"),
+]
+
+interface_image_url = gr.Interface(
+    fn=GetLlamaResponse,
+    inputs=inputs_image_url,
+    outputs=outputs_result_dict,
+    title="Text Generation",
+    cache_examples=False,
+)
+
+gr.TabbedInterface(
+    [interface_image_url],
+    tab_names=['Some inference']
+).queue().launch()

requirements.txt

@@ -3,4 +3,5 @@ uvicorn
 ctransformers
 langchain
 python-box
-gradio
+gradio
+transformers