add examples

app.py

@@ -1,6 +1,7 @@
 import streamlit as st
 from transformers import AutoTokenizer, AutoModelForCausalLM, set_seed
 from transformers import pipeline
+import json
 
 @st.cache(allow_output_mutation=True)
 def load_tokenizer(model_ckpt):
@@ -11,6 +12,12 @@ def load_model(model_ckpt):
     model = AutoModelForCausalLM.from_pretrained(model_ckpt)
     return model
 
+@st.cache()
+def load_examples():
+    with open("examples.json", "r") as f:
+        examples = json.load(f)
+    return dict([(x["name"], x["value"]) for x in examples])
+
 st.set_page_config(page_icon=':parrot:', layout="wide")
 
 default_code = '''\
@@ -20,6 +27,7 @@ def print_hello_world():\
 model_ckpt = "lvwerra/codeparrot"
 tokenizer = load_tokenizer(model_ckpt)
 model = load_model(model_ckpt)
+examples = load_examples()
 set_seed(42)
 gen_kwargs = {}
 
@@ -27,15 +35,18 @@ st.title("CodeParrot 🦜")
 st.markdown('##')
 
 pipe = pipeline('text-generation', model=model, tokenizer=tokenizer)
+st.sidebar.header("Examples:")
+selected_example = st.sidebar.selectbox("Select one of the following examples:", examples.keys())
+example_text = examples[selected_example]
 st.sidebar.header("Generation settings:")
 gen_kwargs["do_sample"] = st.sidebar.radio("Decoding strategy",  ["Greedy", "Sample"]) == "Sample"
-gen_kwargs["max_new_tokens"] = st.sidebar.slider("Number of tokens to generate", value=16, min_value=8, step=8, max_value=256)
+gen_kwargs["max_new_tokens"] = st.sidebar.slider("Number of tokens to generate", value=32, min_value=8, step=8, max_value=256)
 if gen_kwargs["do_sample"]:
     gen_kwargs["temperature"] = st.sidebar.slider("Temperature", value = 0.2, min_value = 0.0, max_value=2.0, step=0.05)
     gen_kwargs["top_k"] = st.sidebar.slider("Top-k", min_value = 0, max_value=100, value = 0)
     gen_kwargs["top_p"] = st.sidebar.slider("Top-p", min_value = 0.0, max_value=1.0, step = 0.01, value = 0.95)
 
-gen_prompt = st.text_area("Generate code with prompt:", value=default_code, height=220,).strip()
+gen_prompt = st.text_area("Generate code with prompt:", value=example_text, height=220,).strip()
 if st.button("Generate code!"):
     with st.spinner("Generating code..."):
         generated_text = pipe(gen_prompt, **gen_kwargs)[0]['generated_text']

examples.json

@@ -0,0 +1,31 @@
+[
+    {
+        "name": "Hello World!",
+        "value": "def print_hello_world():\n    \"\"\"Print 'Hello World!'.\"\"\""
+    },
+    {
+        "name": "Filesize",
+        "value": "def get_file_size(filepath):"
+    },
+    {
+        "name": "Python to Numpy",
+        "value": "# calculate mean in native Python:\ndef mean(a):\n    return sum(a)/len(a)\n\n# calculate mean numpy:\nimport numpy as np\n\ndef mean(a):"
+    },
+    {
+        "name": "unittest",
+        "value": "def is_even(value):\n    \"\"\"Returns True if value is an even number.\"\"\"\n    return value % 2 == 0\n\n# setup unit tests for is_even\nimport unittest"
+
+    },
+    {
+        "name": "Scikit-Learn",
+        "value": "import numpy as np\nfrom sklearn.ensemble import RandomForestClassifier\n\n# create training data\nX = np.random.randn(100, 100)\ny = np.random.randint(0, 1, 100)\n\n# setup train test split"
+    },
+    {
+        "name": "Pandas",
+        "value": "# load dataframe from csv\ndf = pd.read_csv(filename)\n\n# columns: \"age_group\", \"income\"\n# calculate average income per age group"
+    },
+    {
+        "name": "Transformers",
+        "value": "from transformers import AutoTokenizer, AutoModelForSequenceClassification\n\n# build a BERT classifier"
+    }
+]