Update app.py

app.py

@@ -168,11 +168,27 @@ st.latex("A \in \mathbb{R}^{MxK}, B \in R^{KxN}, C \in \mathbb{R}^{MxN}")
 
 st.markdown('''
 To execute this operation on the GPU, we need to 
-- Read A, B from memory
-- Perform math operations
-- Write C to memory
+1. Read A, B from memory
+2. Perform math operations
+3. Write C to memory
 ''')
 
+st.latex('''
+For float16 operations (2 bytes), we can estimate the memory access time of A as follows:
+T_mem(A) = 2*M*K / BW_mem
+where BW_mem is the memory bandwidth of the GPU (e.g. 1935 GB/s for A100)
+''')
+
+st.latex('''
+For float16 operations (2 bytes), we can estimate the memory access time of A as follows:
+T_mem(A) = 2*M*K / BW_mem
+where BW_mem is the memory bandwidth of the GPU (e.g. 1935 GB/s for A100)
+''')
+
+
+
+
+
 
 breakdown = st.checkbox("Show breakdown per operation")
 if breakdown: