Add app

Dockerfile

@@ -0,0 +1,11 @@
+FROM python:3.11
+
+WORKDIR /code
+
+COPY ./requirements.txt /code/requirements.txt
+
+RUN pip install --no-cache-dir --upgrade -r /code/requirements.txt
+
+COPY . .
+
+ENTRYPOINT ["solara", "run", "app.py", "--host=0.0.0.0", "--port=7860"]

LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2023 Alonso Silva Allende
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

app.py

@@ -0,0 +1,28 @@
+import solara
+import torch
+import torch.nn.functional as F
+import pandas as pd
+from transformers import AutoTokenizer, AutoModelForCausalLM
+
+tokenizer = AutoTokenizer.from_pretrained('gpt2')
+model = AutoModelForCausalLM.from_pretrained('gpt2')
+text1 = solara.reactive("Alan Turing theorized that computers would one day become")
+@solara.component
+def Page():
+  with solara.Card(margin=0):
+    solara.Markdown("#Next token prediction visualization")
+    def on_action_cell(column, row_index):
+      text1.value += tokenizer.decode(top_10.indices[0][row_index])
+    cell_actions = [solara.CellAction(icon="mdi-thumb-up", name="Select", on_click=on_action_cell)]
+    solara.InputText("Enter text:", value=text1, continuous_update=True)
+    if text1.value != "":
+      tokens = tokenizer.encode(text1.value, return_tensors="pt")
+      outputs = model.generate(tokens, max_new_tokens=1, output_scores=True, return_dict_in_generate=True, pad_token_id=tokenizer.eos_token_id)
+      scores = F.softmax(outputs.scores[0], dim=-1)
+      top_10 = torch.topk(scores, 10)
+      df = pd.DataFrame()
+      df["probs"] = top_10.values[0]
+      df["probs"] = [f"{value:.2%}" for value in df["probs"].values]
+      df["predicted next token"] = [tokenizer.decode(top_10.indices[0][i]) for i in range(10)]
+      solara.DataFrame(df, items_per_page=10, cell_actions=cell_actions)
+Page()

requirements.txt

@@ -0,0 +1,4 @@
+solara
+pandas
+torch
+transformers