Update app.py

app.py

@@ -11,10 +11,7 @@ def nanogpt(start:str , max_new_tokens = 500, num_samples =2):
 
   # -----------------------------------------------------------------------------
   init_from = 'resume' # either 'resume' (from an out_dir) or a gpt2 variant (e.g. 'gpt2-xl')
-  out_dir = 'out-shakespeare-char' # ignored if init_from is not 'resume'
-  #start = "God is great. I love Him." #"\n"  or "<|endoftext|>" or etc. Can also specify a file, use as: "FILE:prompt.txt"
-  #num_samples = 10 # number of samples to draw
-  #max_new_tokens = 500 # number of tokens generated in each sample
+  
   temperature = 0.8 # 1.0 = no change, < 1.0 = less random, > 1.0 = more random, in predictions
   top_k = 200 # retain only the top_k most likely tokens, clamp others to have 0 probability
   seed = 1337
@@ -85,10 +82,8 @@ def nanogpt(start:str , max_new_tokens = 500, num_samples =2):
               output = decode(y[0].tolist())
   return output
 
-INTERFACE = gr.Interface(fn=nanogpt, inputs=[gr.Textbox(label= "Prompt"),gr.Slider(300,500, "number", label= "Maximum number of tokens to be geenrated")] , outputs=gr.Text(label= "Generated Text"), title="NanoGPT",
-                 description="NanoGPT is a large transformer-based language model with 10.65 million parameters, trained on a small dataset of Shakespeare work (size: 1MB only). It is trained with character level tokeniation with a simple objective: predict the next char, given all of the previous chars within some text.",
-                 examples = [['We as the new generation AI enginners.',300,1],
-                ['A forgotten era of humility and happiness',300,2],
-                
-           ]
+INTERFACE = gr.Interface(fn=nanogpt, inputs=[gr.Textbox(label= "Prompt", value= 'My mind is tossing on the ocean.'),gr.Slider(300,500, "number",value= 250, label= "Maximum number of tokens to be generated")] , outputs=gr.Text(label= "Generated Text"), title="NanoGPT",
+                 description="NanoGPT is a transformer-based language model with only 10.65 million parameters, trained on a small dataset of Shakespeare work (size: 1MB only). It is trained with character level tokenization with a simple objective: predict the next char, given all of the previous chars within a text.",
+                 examples = [['We know what we are, but know not what we may be',300],
+                ['Sweet are the uses of adversity which, like the toad, ugly and venomous, wears yet a precious jewel in his head',300],]
                         ).launch(debug=True)