Add application file

app.py

@@ -1,64 +1,78 @@
-# !usr/bin/env python
-# -*- coding:utf-8 -*-
+# # !usr/bin/env python
+# # -*- coding:utf-8 -*-
 
-'''
- Description  : 
- Version      : 1.0
- Author       : Chaofan Tao
- Mail         : [email protected]
- Github       : https://github.com/sail-sg/scaling-with-vocab
- Date         : 2024-08-09 00:25
- Copyright (C) 2024 Chaofan Tao. All rights reserved.
-'''
-import gradio as gr
-from utils import approach1_isoflops, approach2_derivative, approach3_isoloss
+# '''
+#  Description  : 
+#  Version      : 1.0
+#  Author       : Chaofan Tao
+#  Mail         : [email protected]
+#  Github       : https://github.com/sail-sg/scaling-with-vocab
+#  Date         : 2024-08-09 00:25
+#  Copyright (C) 2024 Chaofan Tao. All rights reserved.
+# '''
+# import gradio as gr
+# from utils import approach1_isoflops, approach2_derivative, approach3_isoloss
 
 
-def compute_optimal_vocab(Nnv: float,
-                        flops: float, 
-                        ):
+# def compute_optimal_vocab(Nnv: float,
+#                         flops: float, 
+#                         ):
     
-    if flops is None:
-        Vopt_app1 = approach1_isoflops(Nnv)
-        Vopt_app2 = approach2_derivative(Nnv)
-        Vopt_app3 = approach3_isoloss(Nnv) 
-    else:
-        Vopt_app1,  Vopt_app2 = None, None
-        Vopt_app3 = approach3_isoloss(Nnv, flops) 
+#     if flops is None:
+#         Vopt_app1 = approach1_isoflops(Nnv)
+#         Vopt_app2 = approach2_derivative(Nnv)
+#         Vopt_app3 = approach3_isoloss(Nnv) 
+#     else:
+#         Vopt_app1,  Vopt_app2 = None, None
+#         Vopt_app3 = approach3_isoloss(Nnv, flops) 
     
-    results = f"## The optimal vocabulary size for non-vocabulary parameters {Nnv:1e} is:\nApproach 1: {Vopt_app1}\nApproach 2: {Vopt_app2}Approach 3: {Vopt_app3}"
-    return results
+#     results = f"## The optimal vocabulary size for non-vocabulary parameters {Nnv:1e} is:\nApproach 1: {Vopt_app1}\nApproach 2: {Vopt_app2}Approach 3: {Vopt_app3}"
+#     return results
 
 
-with gr.Blocks() as demo:
-    with gr.Column():
-        gr.Markdown(
-            """<img src="https://raw.githubusercontent.com/MrYxJ/calculate-flops.pytorch/main/screenshot/calflops_hf3.png?raw=true" style="float: left;" width="250" height="250"><h1> ⛽️Model(Transformers) FLOPs and Parameter Calculator</h1>
-    This tool is used to predict the optimal vocabulary size <h1> given the non-vocabulary parameters $N_{nv}$</h1>.
-    We provide 3 ways for prediction:
+# with gr.Blocks() as demo:
+#     with gr.Column():
+#         gr.Markdown(
+#             """<img src="https://raw.githubusercontent.com/MrYxJ/calculate-flops.pytorch/main/screenshot/calflops_hf3.png?raw=true" style="float: left;" width="250" height="250"><h1> ⛽️Model(Transformers) FLOPs and Parameter Calculator</h1>
+#     This tool is used to predict the optimal vocabulary size <h1> given the non-vocabulary parameters $N_{nv}$</h1>.
+#     We provide 3 ways for prediction:
 
-     - Approach 1: Build the relationship between studied attributes and FLOPs: Build the relationship between the optimal data points (the points that reach the lowest loss under the same FLOPs budget) and the FLOPs.
-     - Approach 2: Derivative-Based Estimation: Fast calculation method using the derivative of FLOPs with respect to the vocabulary size.
-     - Approach 3: Parametric Fit of Loss Formula: Design a loss formula that considers the effect of vocabulary size and utilizes the loss to make prediction.
+#      - Approach 1: Build the relationship between studied attributes and FLOPs: Build the relationship between the optimal data points (the points that reach the lowest loss under the same FLOPs budget) and the FLOPs.
+#      - Approach 2: Derivative-Based Estimation: Fast calculation method using the derivative of FLOPs with respect to the vocabulary size.
+#      - Approach 3: Parametric Fit of Loss Formula: Design a loss formula that considers the effect of vocabulary size and utilizes the loss to make prediction.
     
-    Approach 1 and 2 can only be used to compute the optimal vocabulary size when the compute is optimally allocated to non-vocabulary parameters, vocabulary parameters and data jointly.
-    Approach 3 will not only consider the case above, but also consider the case when the amount of data does not satisfy the optimal compute allocation, and can calculate the optimal vocabulary size with specified $N_{nv}$ and FLOPs.
+#     Approach 1 and 2 can only be used to compute the optimal vocabulary size when the compute is optimally allocated to non-vocabulary parameters, vocabulary parameters and data jointly.
+#     Approach 3 will not only consider the case above, but also consider the case when the amount of data does not satisfy the optimal compute allocation, and can calculate the optimal vocabulary size with specified $N_{nv}$ and FLOPs.
 
-    Thanks for trying 🌟🌟🌟!
-    """)
+#     Thanks for trying 🌟🌟🌟!
+#     """)
         
-        with gr.Row():
-            Nnv = gr.Textbox(label="Non-vocabulary Parameters", value=7*10**9)
-            flops = gr.Textbox(label="FLOPs", placeholder="Optional (e.g. 7.05*10**21)")
-            output_text = gr.Textbox(label='output')
-        with gr.Row():
-            btn = gr.Button("Compute the optimal vocabulary size")
+#         with gr.Row():
+#             Nnv = gr.Textbox(label="Non-vocabulary Parameters", value=7*10**9)
+#             flops = gr.Textbox(label="FLOPs", placeholder="Optional (e.g. 7.05*10**21)")
+#             output_text = gr.Textbox(label='output')
+#         with gr.Row():
+#             btn = gr.Button("Compute the optimal vocabulary size")
 
   
-        demo = gr.Interface(fn=compute_optimal_vocab, inputs=[Nnv, flops], outputs=output_text)
-        btn.click(
-                compute_optimal_vocab,
-                inputs=[Nnv, flops],
-                outputs=output_text
-            )
-demo.launch(share=True)
+#         demo = gr.Interface(fn=compute_optimal_vocab, inputs=[Nnv, flops], outputs=output_text)
+#         btn.click(
+#                 compute_optimal_vocab,
+#                 inputs=[Nnv, flops],
+#                 outputs=output_text
+#             )
+# demo.launch(share=True)
+
+import gradio as gr
+def update(name):
+    return f"Welcome to Gradio, {name}!"
+
+with gr.Blocks() as demo:
+    gr.Markdown("Start typing below and then click **Run** to see the output.")
+    with gr.Row():
+        inp = gr.Textbox(placeholder="What is your name?")
+        out = gr.Textbox()
+    btn = gr.Button("Run")
+    btn.click(fn=update, inputs=inp, outputs=out)
+
+demo.launch()