add files

.ipynb_checkpoints/README-checkpoint.md

@@ -0,0 +1,12 @@
+---
+title: CodeGen Subspace
+emoji: 🔮
+colorFrom: green
+colorTo: purple
+sdk: gradio
+sdk_version: 3.0.4
+app_file: app.py
+pinned: false
+---
+
+Check out the configuration reference at https://huggingface.co/docs/hub/spaces#reference

.ipynb_checkpoints/app-checkpoint.py

@@ -0,0 +1,57 @@
+import gradio as gr
+from transformers import AutoTokenizer, AutoModelForCausalLM, set_seed
+from transformers import pipeline
+
+
+title = "CodeGen Generator"
+description = "This is a subspace to make code generation with [CodeGen](https://huggingface.co/Salesforce/codegen-16B-mono), it is used in a larger [space](https://huggingface.co/spaces/loubnabnl/Code-generation-models-v1) for model comparison. We use the 2B parameters model in this space.
+example = [
+    ["def print_hello_world():", 8, 0.6, 42],
+    ["def get_file_size(filepath):", 24, 0.6, 42],
+    ["def count_lines(filename):", 40, 0.6, 42],
+    ["def count_words(filename):", 40, 0.6, 42]]
+tokenizer = AutoTokenizer.from_pretrained("Salesforce/codegen-2B-mono")
+model = AutoModelForCausalLM.from_pretrained("Salesforce/codegen-2B-mono", low_cpu_mem_usage=True)
+
+
+def code_generation(gen_prompt, max_tokens, temperature=0.6, seed=42):
+    set_seed(seed)
+    pipe = pipeline("text-generation", model=model, tokenizer=tokenizer)
+    generated_text = pipe(gen_prompt, do_sample=True, top_p=0.95, temperature=temperature, max_new_tokens=max_tokens)[0]['generated_text']
+    return generated_text
+
+
+iface = gr.Interface(
+    fn=code_generation, 
+    inputs=[
+        gr.Textbox(lines=10, label="Input code"),
+        gr.inputs.Slider(
+            minimum=8,
+            maximum=256,
+            step=1,
+            default=8,
+            label="Number of tokens to generate",
+        ),
+        gr.inputs.Slider(
+            minimum=0,
+            maximum=2,
+            step=0.1,
+            default=0.6,
+            label="Temperature",
+        ),
+        gr.inputs.Slider(
+            minimum=0,
+            maximum=1000,
+            step=1,
+            default=42,
+            label="Random seed to use for the generation"
+        )
+    ],
+    outputs=gr.Textbox(label="Predicted code", lines=10),
+    examples=example,
+    layout="horizontal",
+    theme="peach",
+    description=description,
+    title=title
+)
+iface.launch()

.ipynb_checkpoints/requirements-checkpoint.txt

@@ -0,0 +1,2 @@
+git+https://github.com/rooa/transformers.git@add_codegen
+torch

README.md

@@ -1,13 +1,12 @@
 ---
-title: Codegen Space
-emoji: 🐠
-colorFrom: purple
+title: CodeGen Subspace
+emoji: 🔮
+colorFrom: green
 colorTo: purple
-sdk: streamlit
-sdk_version: 1.9.0
+sdk: gradio
+sdk_version: 3.0.4
 app_file: app.py
 pinned: false
-license: apache-2.0
 ---
 
 Check out the configuration reference at https://huggingface.co/docs/hub/spaces#reference

app.py

@@ -0,0 +1,57 @@
+import gradio as gr
+from transformers import AutoTokenizer, AutoModelForCausalLM, set_seed
+from transformers import pipeline
+
+
+title = "CodeGen Generator"
+description = "This is a subspace to make code generation with [CodeGen](https://huggingface.co/Salesforce/codegen-16B-mono), it is used in a larger [space](https://huggingface.co/spaces/loubnabnl/Code-generation-models-v1) for model comparison. We use the 2B parameters model in this space.
+example = [
+    ["def print_hello_world():", 8, 0.6, 42],
+    ["def get_file_size(filepath):", 24, 0.6, 42],
+    ["def count_lines(filename):", 40, 0.6, 42],
+    ["def count_words(filename):", 40, 0.6, 42]]
+tokenizer = AutoTokenizer.from_pretrained("Salesforce/codegen-2B-mono")
+model = AutoModelForCausalLM.from_pretrained("Salesforce/codegen-2B-mono", low_cpu_mem_usage=True)
+
+
+def code_generation(gen_prompt, max_tokens, temperature=0.6, seed=42):
+    set_seed(seed)
+    pipe = pipeline("text-generation", model=model, tokenizer=tokenizer)
+    generated_text = pipe(gen_prompt, do_sample=True, top_p=0.95, temperature=temperature, max_new_tokens=max_tokens)[0]['generated_text']
+    return generated_text
+
+
+iface = gr.Interface(
+    fn=code_generation, 
+    inputs=[
+        gr.Textbox(lines=10, label="Input code"),
+        gr.inputs.Slider(
+            minimum=8,
+            maximum=256,
+            step=1,
+            default=8,
+            label="Number of tokens to generate",
+        ),
+        gr.inputs.Slider(
+            minimum=0,
+            maximum=2,
+            step=0.1,
+            default=0.6,
+            label="Temperature",
+        ),
+        gr.inputs.Slider(
+            minimum=0,
+            maximum=1000,
+            step=1,
+            default=42,
+            label="Random seed to use for the generation"
+        )
+    ],
+    outputs=gr.Textbox(label="Predicted code", lines=10),
+    examples=example,
+    layout="horizontal",
+    theme="peach",
+    description=description,
+    title=title
+)
+iface.launch()

requirements.txt

@@ -0,0 +1,2 @@
+git+https://github.com/rooa/transformers.git@add_codegen
+torch