Initial commit

README.md

@@ -10,4 +10,3 @@ pinned: false
 license: apache-2.0
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app.py

@@ -0,0 +1,48 @@
+import gradio as gr
+from prompt_generator import generate_prompt
+
+tokenizers = [
+    "google/gemma-7b",
+    "meta-llama/Llama-2-7b",
+    "mistralai/Mistral-7B-v0.1",
+    "facebook/opt-2.7b",
+    "microsoft/phi-2",
+    "THUDM/chatglm3-6b",
+    "Qwen/Qwen1.5-7B-Chat",
+    "bigscience/bloom-560m",
+    "ise-uiuc/Magicoder-S-DS-6.7B",
+    "google/flan-t5-base",
+    "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
+    "google-bert/bert-base-uncased",
+]
+
+
+def generate(model_id, num_tokens):
+    output_file = f"prompt_{num_tokens}.jsonl"
+    prompt = generate_prompt(model_id, int(num_tokens), silent=True, output_file=output_file)
+    return prompt, output_file
+
+
+demo = gr.Interface(
+    fn=generate,
+    title="Prompt Generator",
+    description="Generate prompts with a given length for testing transformer models. "
+    "Prompt source: https://archive.org/stream/alicesadventures19033gut/19033.txt",
+    inputs=[
+        gr.Dropdown(label="Tokenizer", choices=tokenizers, allow_custom_value=True),
+        gr.Textbox(label="Number of Tokens"),
+    ],
+    outputs=[gr.Textbox(label="prompt", show_copy_button=True), gr.File(label="Json file")],
+    examples=[
+        ["mistralai/Mistral-7B-v0.1", 32],
+        ["mistralai/Mistral-7B-v0.1", 64],
+        ["mistralai/Mistral-7B-v0.1", 128],
+        ["mistralai/Mistral-7B-v0.1", 512],
+        ["mistralai/Mistral-7B-v0.1", 1024],
+        ["mistralai/Mistral-7B-v0.1", 2048],
+    ],
+    cache_examples=False,
+    allow_flagging=False,
+)
+
+demo.launch()

requirements.txt

@@ -0,0 +1,2 @@
+transformers
+git+https://github.com/helena-intel/prompt-generator.git