upload v1687488717 model

.ipynb_checkpoints/README-checkpoint.md

@@ -0,0 +1,13 @@
+---
+title: KerasNLP GPT2 Alpaca
+emoji: 📊
+colorFrom: red
+colorTo: yellow
+sdk: gradio
+sdk_version: 3.35.2
+app_file: app.py
+pinned: false
+license: apache-2.0
+---
+
+Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

.ipynb_checkpoints/app-checkpoint.py

@@ -0,0 +1,46 @@
+from typing import Text, Any, Dict, Optional
+
+import gradio as gr
+import tensorflow as tf
+import tensorflow_text
+from tensorflow.python.saved_model import tag_constants
+from huggingface_hub import Repository
+
+local_path = "hf_model"
+
+model_version = "v1687488717"
+model_repo_id = "chansung/kerasnlp-gpt2-alpaca-pipeline"
+model_repo_url = f"https://huggingface.co/{model_repo_id}"
+
+def _clone_and_checkout(repo_url: str, local_path: str, version: str) -> Repository:
+    repository = Repository(
+        local_dir=local_path, clone_from=repo_url
+    )
+    repository.git_checkout(revision=version)
+    return repository
+
+_ = _clone_and_checkout(model_repo_url, local_path, model_version)
+model = tf.saved_model.load(local_path, tags=[tag_constants.SERVING])
+gpt_lm_predict_fn = model.signatures["serving_default"]
+
+def gen_text(prompt, max_length=512):
+    prompt = tf.constant(f"### Instruction:\n{prompt}\n\n### Response:\n")
+    max_length = tf.constant(max_length, dtype="int64")
+    
+    result = gpt_lm_predict_fn(
+        prompt=prompt,
+        max_length=max_length,
+    )
+
+    return result['result'].numpy().decode('UTF-8').split("### Response:")[-1].strip()
+
+with gr.Blocks() as demo:
+    instruction = gr.Textbox("Instruction")
+    output = gr.Textbox("Output", lines=5)
+
+    instruction.submit(
+        lambda prompt: gen_text(prompt),
+        instruction, output
+    )
+
+demo.launch()

.ipynb_checkpoints/requirements-checkpoint.txt

@@ -0,0 +1,3 @@
+tensorflow
+tensorflow_text
+huggingface_hub

README.md

@@ -1,12 +1,25 @@
 ---
-title: Kerasnlp Gpt2 Alpaca Pipeline
-emoji: 🦀
-colorFrom: green
-colorTo: gray
+title: KerasNLP GPT2 Alpaca
+emoji: 📊
+colorFrom: red
+colorTo: yellow
 sdk: gradio
 sdk_version: 3.35.2
 app_file: app.py
 pinned: false
+license: apache-2.0
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference---
+title: KerasNLP GPT2 Alpaca
+emoji: 📊
+colorFrom: red
+colorTo: yellow
+sdk: gradio
+sdk_version: 3.35.2
+app_file: app.py
+pinned: false
+license: apache-2.0
+---
+
+Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app.py

@@ -0,0 +1,46 @@
+from typing import Text, Any, Dict, Optional
+
+import gradio as gr
+import tensorflow as tf
+import tensorflow_text
+from tensorflow.python.saved_model import tag_constants
+from huggingface_hub import Repository
+
+local_path = "hf_model"
+
+model_version = "v1687488717"
+model_repo_id = "chansung/kerasnlp-gpt2-alpaca-pipeline"
+model_repo_url = f"https://huggingface.co/{model_repo_id}"
+
+def _clone_and_checkout(repo_url: str, local_path: str, version: str) -> Repository:
+    repository = Repository(
+        local_dir=local_path, clone_from=repo_url
+    )
+    repository.git_checkout(revision=version)
+    return repository
+
+_ = _clone_and_checkout(model_repo_url, local_path, model_version)
+model = tf.saved_model.load(local_path, tags=[tag_constants.SERVING])
+gpt_lm_predict_fn = model.signatures["serving_default"]
+
+def gen_text(prompt, max_length=512):
+    prompt = tf.constant(f"### Instruction:\n{prompt}\n\n### Response:\n")
+    max_length = tf.constant(max_length, dtype="int64")
+    
+    result = gpt_lm_predict_fn(
+        prompt=prompt,
+        max_length=max_length,
+    )
+
+    return result['result'].numpy().decode('UTF-8').split("### Response:")[-1].strip()
+
+with gr.Blocks() as demo:
+    instruction = gr.Textbox("Instruction")
+    output = gr.Textbox("Output", lines=5)
+
+    instruction.submit(
+        lambda prompt: gen_text(prompt),
+        instruction, output
+    )
+
+demo.launch()

requirements.txt

@@ -0,0 +1,3 @@
+tensorflow
+tensorflow_text
+huggingface_hub