major app update

app.py

@@ -0,0 +1,102 @@
+import streamlit as st
+import gradio as gr
+import pandas as pd
+import os
+
+HF_TOKEN = os.getenv('HUGGING_FACE_HUB_TOKEN')
+hf_writer = gr.HuggingFaceDatasetSaver(HF_TOKEN, "co2_submissions")
+
+st.set_page_config(
+    page_title="AI Carbon Calculator",
+    layout="wide",
+)
+
+tdp_url = "https://raw.githubusercontent.com/mlco2/impact/master/data/gpus.csv"
+compute_url = "https://raw.githubusercontent.com/mlco2/impact/master/data/impact.csv"
+
+
+electricity_url = "https://raw.githubusercontent.com/mlco2/impact/master/data/2021-10-27yearly_averages.csv"
+
+server_sheet_id = "1DqYgQnEDLQVQm5acMAhLgHLD8xXCG9BIrk-_Nv6jF3k"
+server_sheet_name = "Server%20Carbon%20Footprint"
+server_url = f"https://docs.google.com/spreadsheets/d/{server_sheet_id}/gviz/tq?tqx=out:csv&sheet={server_sheet_name}"
+
+
+embodied_gpu_sheet_id = "1DqYgQnEDLQVQm5acMAhLgHLD8xXCG9BIrk-_Nv6jF3k"
+embodied_gpu_sheet_name = "Scope%203%20Ratios"
+embodied_gpu_url = f"https://docs.google.com/spreadsheets/d/{embodied_gpu_sheet_id}/gviz/tq?tqx=out:csv&sheet={embodied_gpu_sheet_name}"
+
+TDP =pd.read_csv(tdp_url)
+
+instances = pd.read_csv(compute_url)
+providers = [p.upper() for p in instances['provider'].unique().tolist()]
+providers.append('Local/Private Infastructure')
+
+kg_per_mile = 0.348
+
+electricity = pd.read_csv(electricity_url)
+servers = pd.read_csv(server_url)
+embodied_gpu = pd.read_csv(embodied_gpu_url)
+
+
+st.title("AI Carbon Calculator")
+
+st.markdown('## Estimate your model\'s CO2 carbon footprint!')
+
+st.markdown('Building on the work of the [ML CO2 Calculator](https://mlco2.github.io/impact/), this tool allows you to consider'
+            'other aspects of your model\'s carbon footprint based on the LCA methodology.')
+
+st.markdown('We will consider 3 aspects of your model: the dynamic emissions, idle emissions embodied emissions.')
+
+st.markdown('### Dynamic Emissions')
+with st.expander("Calculate the emissions produced by energy consumption of model training"):
+    with st.form(key='dynamic_emissions'):
+        col1, col2, col3, col4 = st.columns(4)
+        with col1:
+            hardware = st.selectbox('GPU used', TDP['name'].tolist())
+            gpu_tdp = TDP['tdp_watts'][TDP['name'] == hardware].tolist()[0]
+            st.markdown("Different GPUs have different TDP (Thermal Design Power), which impacts how much energy you use.")
+        with col2:
+           training_time = st.number_input('Total number of GPU hours')
+           st.markdown('This is calculated by multiplying the number of GPUs you used by the training time: '
+                       'i.e. if you used 100 GPUs for 10 hours, this is equal to 100x10 = 1,000 GPU hours.')
+        with col3:
+           provider = st.selectbox('Provider used', providers)
+           st.markdown('If you can\'t find your provider here, select "Local/Private Infrastructure".')
+        with col4:
+            if provider != 'Local/Private Infastructure':
+                provider_instances = instances['region'][instances['provider'] == provider.lower()].unique().tolist()
+                region = st.selectbox('Provider used', provider_instances)
+                carbon_intensity = instances['impact'][(instances['provider'] == provider.lower()) & (instances['region'] == region)].tolist()[0]
+
+            else:
+                carbon_intensity = st.number_input('Carbon intensity of your energy grid, in grams of CO2 per kWh')
+                st.markdown('You can consult a resource like the [IEA](https://www.iea.org/countries) or '
+                            ' [Electricity Map](https://app.electricitymaps.com/) to get this information.')
+
+        dynamic_emissions = round(gpu_tdp * training_time * carbon_intensity/1000000)
+        st.metric(label="Dynamic emissions", value=str(dynamic_emissions)+' kilograms of CO2eq')
+        st.markdown('This is roughly equivalent to '+ str(round(dynamic_emissions/kg_per_mile,1)) + ' miles driven in an average US car'
+            ' produced in 2021. [(Source: energy.gov)](https://www.energy.gov/eere/vehicles/articles/fotw-1223-january-31-2022-average-carbon-dioxide-emissions-2021-model-year)')
+        hf_writer.setup([hardware, training_time, provider, carbon_intensity, dynamic_emissions], "dynamic_emissions")
+        st.form_submit_button(label="Share my data", help="Submit the data from your model anonymously for research purposes!",
+        onclick=hf_writer.flag([hardware, training_time, provider, carbon_intensity, dynamic_emissions]))
+
+
+
+st.markdown('### Idle Emissions')
+st.markdown('Do you know what the PUE (Power Usage Effectiveness) of your infrastructure is?')
+
+
+
+st.markdown('### Embodied Emissions')
+st.markdown('Choose your hardware, runtime and cloud provider/physical infrastructure to estimate the carbon impact of your research.')
+
+
+
+
+st.markdown('#### More information about our Methodology')
+
+st.image('images/LCA_CO2.png', caption='The LCA methodology - the parts in green are those we focus on.')
+
+modelname = st.selectbox('Choose a model to test', TDP)