From b3d3bc849b4068aa074b0f42be104d0188b4d1a2 Mon Sep 17 00:00:00 2001 From: SkalskiP Date: Thu, 16 May 2024 16:14:10 +0200 Subject: [PATCH] add PaliGemma fine-tuning notebook --- ...etune-paligemma-on-detection-dataset.ipynb | 2460 +++++++++++++++++ 1 file changed, 2460 insertions(+) create mode 100644 notebooks/how-to-finetune-paligemma-on-detection-dataset.ipynb diff --git a/notebooks/how-to-finetune-paligemma-on-detection-dataset.ipynb b/notebooks/how-to-finetune-paligemma-on-detection-dataset.ipynb new file mode 100644 index 0000000..5a52d97 --- /dev/null +++ b/notebooks/how-to-finetune-paligemma-on-detection-dataset.ipynb @@ -0,0 +1,2460 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "source": [ + "[![Roboflow Notebooks](https://media.roboflow.com/notebooks/template/bannertest2-2.png?ik-sdk-version=javascript-1.4.3&updatedAt=1672932710194)](https://github.com/roboflow/notebooks)\n", + "\n", + "# Fine-tune PaliGemma on Object Detection Dataset\n", + "\n", + "---\n", + "\n", + "[![GitHub](https://badges.aleen42.com/src/github.svg)](https://github.com/google-research/big_vision/blob/main/big_vision/configs/proj/paligemma/README.md)\n", + "\n", + "PaliGemma is an open vision-language model (VLM) inspired by PaLI-3, built with\n", + "open components, such as\n", + "the [SigLIP vision model](https://colab.research.google.com/github/google-research/big_vision/blob/main/big_vision/configs/proj/image_text/SigLIP_demo.ipynb)\n", + "and\n", + "the [Gemma language model](https://ai.google.dev/gemma).\n", + "PaliGemma is designed as a versatile model for transfer to a wide range of\n", + "vision-language tasks such as image and short video caption, visual question\n", + "answering, text reading, object detection and object segmentation. Together with\n", + "the pretrained and transfer checkpoints at multiple resolutions, we provide a\n", + "checkpoint transferred to a mixture of tasks that can be used for off-the-shelf\n", + "exploration.\n", + "\n", + "This notebook is an extension of the [official notebook](https://colab.research.google.com/github/google-research/big_vision/blob/main/big_vision/configs/proj/paligemma/finetune_paligemma.ipynb) prepared by Google Research.\n", + "\n", + "![PaliGemma model](https://storage.cloud.google.com/com-roboflow-marketing/notebooks/examples/paligemma.png)\n", + "\n", + "To make it runnable on a T4 colab runtime with 16GB HBM and 12GB RAM, we opt to only finetune the attention layers of the language model and freeze the other parameters.\n", + "\n", + " * Download and parse Roboflow Universe dataset.\n", + " * Install deps, download model checkpoint and training data.\n", + " * Load the model onto GPU devices.\n", + " * Prepare the input to the model for training and inference.\n", + " * Finetune the model and inspect output in validation split." + ], + "metadata": { + "id": "4LqvmtZPzyY1" + } + }, + { + "cell_type": "markdown", + "source": [ + "## Download and parse Roboflow Universe dataset\n", + "\n", + "**NOTE:**\n", + "PaliGemma requires the dataset to be in the appropriate format. Let's start by parsing the dataset from the standard YOLO format to JSONL, which is compatible with the PaliGemma training pipeline." + ], + "metadata": { + "id": "FMlw3ru1YvLg" + } + }, + { + "cell_type": "code", + "source": [ + "!pip install -q roboflow supervision" + ], + "metadata": { + "id": "Wtvz4QZ9YuG8", + "outputId": "c652c668-7beb-40cd-9683-c80306c372ae", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "execution_count": 1, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "\u001b[?25l \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m0.0/74.9 kB\u001b[0m \u001b[31m?\u001b[0m eta \u001b[36m-:--:--\u001b[0m\r\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m74.9/74.9 kB\u001b[0m \u001b[31m2.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m111.0/111.0 kB\u001b[0m \u001b[31m6.7 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m158.3/158.3 kB\u001b[0m \u001b[31m9.3 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m178.7/178.7 kB\u001b[0m \u001b[31m9.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m58.8/58.8 kB\u001b[0m \u001b[31m6.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m49.1/49.1 MB\u001b[0m \u001b[31m30.7 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m54.5/54.5 kB\u001b[0m \u001b[31m6.8 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[?25h" + ] + } + ] + }, + { + "cell_type": "code", + "source": [ + "from google.colab import userdata\n", + "from roboflow import Roboflow\n", + "\n", + "ROBOFLOW_API_KEY = userdata.get('ROBOFLOW_API_KEY')\n", + "\n", + "rf = Roboflow(api_key=ROBOFLOW_API_KEY)\n", + "project = rf.workspace(\"srinithi-s-tzdkb\").project(\"fracture-detection-rhud5\")\n", + "version = project.version(3)\n", + "dataset = version.download(\"yolov8\")" + ], + "metadata": { + "id": "TGDFTYVnY4zn", + "outputId": "406c3ecb-a2f1-41b7-b2d1-79517518bcfe", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "execution_count": 2, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "loading Roboflow workspace...\n", + "loading Roboflow project...\n", + "[WARNING] we noticed you are downloading a `yolov8` datasets but you don't have `ultralytics` installed. Roboflow `.deploy` supports only models trained with `ultralytics==8.0.196`, to intall it `pip install ultralytics==8.0.196`.\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "Downloading Dataset Version Zip in fracture-detection-3 to yolov8:: 100%|██████████| 25468/25468 [00:01<00:00, 16833.18it/s]" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "\n", + "Extracting Dataset Version Zip to fracture-detection-3 in yolov8:: 100%|██████████| 2082/2082 [00:00<00:00, 9227.03it/s]\n" + ] + } + ] + }, + { + "cell_type": "code", + "source": [ + "import os\n", + "import json\n", + "import shutil\n", + "import numpy as np\n", + "import supervision as sv\n", + "from typing import List, Dict, Optional\n", + "\n", + "\n", + "def ensure_directory_exists(directory: str) -> None:\n", + " if not os.path.exists(directory):\n", + " os.makedirs(directory)\n", + "\n", + "\n", + "def extract_entries(\n", + " dataset: sv.DetectionDataset,\n", + " classes: List[str],\n", + " start_idx: Optional[int] = 0,\n", + " end_idx: Optional[int] = None\n", + ") -> List[Dict[str, str]]:\n", + " entries = []\n", + "\n", + " for index, (image_path, image, annotations) in enumerate(dataset):\n", + " if index < start_idx:\n", + " continue\n", + " if end_idx is not None and index >= end_idx:\n", + " break\n", + "\n", + " height, width, _ = image.shape\n", + "\n", + " class_names = set(classes[class_id] for class_id in annotations.class_id)\n", + " prefix = \" ; \".join(class_names)\n", + " prefix = \"detect \" + prefix\n", + "\n", + " suffix_components = []\n", + " for xyxy, class_id in zip(annotations.xyxy, annotations.class_id):\n", + " xyxy = xyxy.copy()\n", + " xyxy /= np.array([width, height, width, height])\n", + " xyxy *= 1024\n", + " suffix_component = f\" {classes[class_id]}\"\n", + " suffix_components.append(suffix_component)\n", + "\n", + " suffix = \" ; \".join(suffix_components)\n", + " image = os.path.basename(image_path)\n", + " entries.append({\"prefix\": prefix, \"suffix\": suffix, \"image\": image})\n", + "\n", + " return entries\n", + "\n", + "\n", + "def copy_images(\n", + " dataset: sv.DetectionDataset,\n", + " target_dir: str,\n", + " start_idx: Optional[int] = 0,\n", + " end_idx: Optional[int] = None\n", + ") -> None:\n", + " for index, (image_path, image, annotations) in enumerate(dataset):\n", + " if index < start_idx:\n", + " continue\n", + " if end_idx is not None and index >= end_idx:\n", + " break\n", + "\n", + " target_path = os.path.join(target_dir, os.path.basename(image_path))\n", + " shutil.copy(image_path, target_path)\n", + "\n", + "\n", + "def save_entries(entries: List[Dict[str, str]], output_path: str) -> None:\n", + " with open(output_path, 'w') as file:\n", + " for entry in entries:\n", + " file.write(json.dumps(entry) + '\\n')\n", + "\n", + "\n", + "def parse_dataset(\n", + " dataset: sv.DetectionDataset,\n", + " target_dir: str,\n", + " target_file_path: str,\n", + " classes: List[str],\n", + " start_idx: Optional[int] = 0,\n", + " end_idx: Optional[int] = None\n", + ") -> None:\n", + " ensure_directory_exists(directory=target_dir)\n", + " copy_images(\n", + " dataset=dataset,\n", + " target_dir=target_dir,\n", + " start_idx=start_idx,\n", + " end_idx=end_idx)\n", + " entries = extract_entries(\n", + " dataset=dataset,\n", + " classes=classes,\n", + " start_idx=start_idx,\n", + " end_idx=end_idx)\n", + " save_entries(entries=entries, output_path=target_file_path)" + ], + "metadata": { + "id": "reRShie2ZFcH" + }, + "execution_count": 3, + "outputs": [] + }, + { + "cell_type": "code", + "source": [ + "import supervision as sv\n", + "\n", + "train_ds = sv.DetectionDataset.from_yolo(\n", + " images_directory_path=f\"{dataset.location}/train/images\",\n", + " annotations_directory_path=f\"{dataset.location}/train/labels\",\n", + " data_yaml_path=f\"{dataset.location}/data.yaml\"\n", + ")\n", + "\n", + "valid_ds = sv.DetectionDataset.from_yolo(\n", + " images_directory_path=f\"{dataset.location}/valid/images\",\n", + " annotations_directory_path=f\"{dataset.location}/valid/labels\",\n", + " data_yaml_path=f\"{dataset.location}/data.yaml\"\n", + ")" + ], + "metadata": { + "id": "QnGTgGY0ZLxA" + }, + "execution_count": 4, + "outputs": [] + }, + { + "cell_type": "code", + "source": [ + "len(train_ds), len(valid_ds)" + ], + "metadata": { + "id": "N8xQtqC3ZOkJ", + "outputId": "749ffb4b-914c-49c9-bcdd-aae5f3225344", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "execution_count": 5, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "(780, 99)" + ] + }, + "metadata": {}, + "execution_count": 5 + } + ] + }, + { + "cell_type": "code", + "source": [ + "train_ds.classes" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "ojgIqcZv6oPq", + "outputId": "9bb5bdd2-2aa5-4c11-8eb2-5f316b967778" + }, + "execution_count": 6, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "['Bone fracture detection - v1 2023-03-05 5-51pm']" + ] + }, + "metadata": {}, + "execution_count": 6 + } + ] + }, + { + "cell_type": "markdown", + "source": [ + "**NOTE:** Before parsing the dataset, it's worth checking the names of the object classes in the original dataset. If they are of low quality, rename them." + ], + "metadata": { + "id": "V6fnTTkN6rPx" + } + }, + { + "cell_type": "code", + "source": [ + "CLASSES = ['fracture']\n", + "DATA_DIR = \"fracture\"" + ], + "metadata": { + "id": "B6RGe_y8ZRPg" + }, + "execution_count": 7, + "outputs": [] + }, + { + "cell_type": "code", + "source": [ + "parse_dataset(train_ds, DATA_DIR, f\"{DATA_DIR}/data_train.jsonl\", CLASSES)\n", + "parse_dataset(valid_ds, DATA_DIR, f\"{DATA_DIR}/data_val.jsonl\", CLASSES)" + ], + "metadata": { + "id": "sncufv4lZaLa" + }, + "execution_count": 8, + "outputs": [] + }, + { + "cell_type": "code", + "source": [ + "!head -n 5 {DATA_DIR}/data_train.jsonl" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "WLhSenP5AtQe", + "outputId": "412bedbf-df9f-4866-9001-dca13e9f096e" + }, + "execution_count": 9, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "{\"prefix\": \"detect fracture\", \"suffix\": \" fracture\", \"image\": \"rot_0_7471_png_jpg.rf.30ec1d3771a6b126e7d5f14ad0b3073b.jpg\"}\n", + "{\"prefix\": \"detect fracture\", \"suffix\": \" fracture\", \"image\": \"flip_0_5824_png_jpg.rf.abe91e2cd085f0d47e35ef7021ff8549.jpg\"}\n", + "{\"prefix\": \"detect fracture\", \"suffix\": \" fracture\", \"image\": \"all_0_8542_png_jpg.rf.6bcad49d206468d7720d727caca95724.jpg\"}\n", + "{\"prefix\": \"detect fracture\", \"suffix\": \" fracture\", \"image\": \"bri_0_592_png_jpg.rf.8d8630701ed43bb703fdad74c8765b26.jpg\"}\n", + "{\"prefix\": \"detect fracture\", \"suffix\": \" fracture ; fracture ; fracture\", \"image\": \"all_0_2435_png_jpg.rf.e602a2f82935f4fdba988b280ab11b7e.jpg\"}\n" + ] + } + ] + }, + { + "cell_type": "code", + "source": [ + "!head -n 5 {DATA_DIR}/data_val.jsonl" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "YwHY21ABA0WG", + "outputId": "4e925557-3886-44d5-eede-e2ab0c975951" + }, + "execution_count": 10, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "{\"prefix\": \"detect fracture\", \"suffix\": \" fracture\", \"image\": \"n_0_3127_png_jpg.rf.7b47b2b3ca07739d0c8c3b76035517db.jpg\"}\n", + "{\"prefix\": \"detect fracture\", \"suffix\": \" fracture\", \"image\": \"z_0_2899_png_jpg.rf.fe16767f8c9eea741b91d71bff5e35d0.jpg\"}\n", + "{\"prefix\": \"detect fracture\", \"suffix\": \" fracture ; fracture\", \"image\": \"rot_0_2535_png_jpg.rf.f8b989f8ee9992855932ea50c8a4350c.jpg\"}\n", + "{\"prefix\": \"detect fracture\", \"suffix\": \" fracture ; fracture\", \"image\": \"all_0_9008_png_jpg.rf.fd15521d8aa78110a1976658f65f682f.jpg\"}\n", + "{\"prefix\": \"detect fracture\", \"suffix\": \" fracture\", \"image\": \"bri_0_520_png_jpg.rf.68504a0b42c52a63dcb7ddd2e6843a6f.jpg\"}\n" + ] + } + ] + }, + { + "cell_type": "markdown", + "source": [ + "**NOTE:** Let's verify if the dataset conversion was successful." + ], + "metadata": { + "id": "u0SAC3oT74dz" + } + }, + { + "cell_type": "code", + "source": [ + "import re\n", + "import numpy as np\n", + "import supervision as sv\n", + "from typing import Tuple, List, Optional\n", + "\n", + "\n", + "def from_pali_gemma(\n", + " response: str,\n", + " resolution_wh: Tuple[int, int],\n", + " classes: Optional[List[str]] = None\n", + ") -> sv.Detections:\n", + " _SEGMENT_DETECT_RE = re.compile(\n", + " r'(.*?)' +\n", + " r'' * 4 + r'\\s*' +\n", + " '(?:%s)?' % (r'' * 16) +\n", + " r'\\s*([^;<>]+)? ?(?:; )?',\n", + " )\n", + "\n", + " width, height = resolution_wh\n", + " xyxy_list = []\n", + " class_name_list = []\n", + "\n", + " while response:\n", + " m = _SEGMENT_DETECT_RE.match(response)\n", + " if not m:\n", + " break\n", + "\n", + " gs = list(m.groups())\n", + " before = gs.pop(0)\n", + " name = gs.pop()\n", + " y1, x1, y2, x2 = [int(x) / 1024 for x in gs[:4]]\n", + " y1, x1, y2, x2 = map(round, (y1*height, x1*width, y2*height, x2*width))\n", + "\n", + " content = m.group()\n", + " if before:\n", + " response = response[len(before):]\n", + " content = content[len(before):]\n", + "\n", + " xyxy_list.append([x1, y1, x2, y2])\n", + " class_name_list.append(name.strip())\n", + " response = response[len(content):]\n", + "\n", + " xyxy = np.array(xyxy_list)\n", + " class_name = np.array(class_name_list)\n", + "\n", + " if classes is None:\n", + " class_id = None\n", + " else:\n", + " class_id = np.array([classes.index(name) for name in class_name])\n", + "\n", + " return sv.Detections(\n", + " xyxy=xyxy,\n", + " class_id=class_id,\n", + " data={'class_name': class_name}\n", + " )" + ], + "metadata": { + "id": "gkakRrzkJgdq" + }, + "execution_count": 11, + "outputs": [] + }, + { + "cell_type": "code", + "source": [ + "from PIL import Image\n", + "\n", + "image = Image.open(f\"{DATA_DIR}/rot_0_7471_png_jpg.rf.30ec1d3771a6b126e7d5f14ad0b3073b.jpg\")\n", + "detections = from_pali_gemma(\" fracture\", image.size, CLASSES)\n", + "sv.BoundingBoxAnnotator().annotate(image, detections)" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 657 + }, + "id": "K9pjdoSmYvqG", + "outputId": "07554a2c-9481-468d-8fcd-1196ff1bd476" + }, + "execution_count": 12, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "" + ], + "image/png": "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\n" + }, + "metadata": {}, + "execution_count": 12 + } + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "6U0QUFveqSP2" + }, + "source": [ + "## Setup" + ] + }, + { + "cell_type": "code", + "execution_count": 13, + "metadata": { + "id": "DfxKb3F839Ks", + "outputId": "b292131e-cb1a-4600-dde9-e68034a3a651", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m77.9/77.9 kB\u001b[0m \u001b[31m4.4 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[?25h Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m43.2/43.2 kB\u001b[0m \u001b[31m5.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[?25h Building wheel for ml_collections (setup.py) ... \u001b[?25l\u001b[?25hdone\n" + ] + } + ], + "source": [ + "# @title Fetch big_vision code and install dependencies.\n", + "import os\n", + "import sys\n", + "\n", + "# TPUs with\n", + "if \"COLAB_TPU_ADDR\" in os.environ:\n", + " raise \"It seems you are using Colab with remote TPUs which is not supported.\"\n", + "\n", + "# Fetch big_vision repository if python doesn't know about it and install\n", + "# dependencies needed for this notebook.\n", + "if not os.path.exists(\"big_vision_repo\"):\n", + " !git clone --quiet --branch=main --depth=1 \\\n", + " https://github.com/google-research/big_vision big_vision_repo\n", + "\n", + "# Append big_vision code to python import path\n", + "if \"big_vision_repo\" not in sys.path:\n", + " sys.path.append(\"big_vision_repo\")\n", + "\n", + "# Install missing dependencies. Assume jax~=0.4.25 with GPU available.\n", + "!pip3 install -q \"overrides\" \"ml_collections\" \"einops~=0.7\" \"sentencepiece\"\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "azmRZvgGyhAb" + }, + "source": [ + "### Configure your API key to access Kaggle\n", + "\n", + "To use PaliGemma, you must provide your Kaggle username and a Kaggle API key.\n", + "\n", + "1. To generate a Kaggle API key, go to the **Account** tab of your Kaggle user profile and select **Create New Token**. This will trigger the download of a `kaggle.json` file containing your API credentials.\n", + "1. In Colab, select **Secrets** (🔑) in the left pane and add your Kaggle username and Kaggle API key. Store your username under the name `KAGGLE_USERNAME` and your API key under the name `KAGGLE_KEY`.\n", + "\n", + "To be able to download, you will also need to acknowledge the Terms and Conditions of the PaliGemma on:\n", + "\n", + "* https://www.kaggle.com/models/google/paligemma/\n", + "\n" + ] + }, + { + "cell_type": "code", + "execution_count": 14, + "metadata": { + "id": "zGLIp1Cx3_CX" + }, + "outputs": [], + "source": [ + "import os\n", + "from google.colab import userdata\n", + "\n", + "# Note: `userdata.get` is a Colab API. If you're not using Colab, set the env\n", + "# vars as appropriate or make your credentials available in ~/.kaggle/kaggle.json\n", + "\n", + "os.environ[\"KAGGLE_USERNAME\"] = userdata.get('KAGGLE_USERNAME')\n", + "os.environ[\"KAGGLE_KEY\"] = userdata.get('KAGGLE_KEY')" + ] + }, + { + "cell_type": "code", + "execution_count": 15, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "gQNOTfF24AV4", + "outputId": "96cb1910-0ef9-4a0e-a338-22210e9ec658" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Downloading the checkpoint from Kaggle, this could take a few minutes....\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "Downloading from https://www.kaggle.com/api/v1/models/google/paligemma/jax/paligemma-3b-pt-224/1/download/paligemma-3b-pt-224.f16.npz...\n", + "100%|██████████| 5.45G/5.45G [02:33<00:00, 38.0MB/s]\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Model path: /root/.cache/kagglehub/models/google/paligemma/jax/paligemma-3b-pt-224/1/./paligemma-3b-pt-224.f16.npz\n", + "Downloading the model tokenizer...\n", + "Copying gs://big_vision/paligemma_tokenizer.model...\n", + "/ [1 files][ 4.1 MiB/ 4.1 MiB] \n", + "Operation completed over 1 objects/4.1 MiB. \n", + "Tokenizer path: ./paligemma_tokenizer.model\n" + ] + } + ], + "source": [ + "# @title Download checkpoint, tokenizer and dataset to local filesystem.\n", + "#\n", + "import os\n", + "import kagglehub\n", + "\n", + "MODEL_PATH = \"./paligemma-3b-pt-224.f16.npz\"\n", + "if not os.path.exists(MODEL_PATH):\n", + " print(\"Downloading the checkpoint from Kaggle, this could take a few minutes....\")\n", + " # Note: kaggle archive contains the same checkpoint in multiple formats.\n", + " # Download only the float16 model.\n", + " MODEL_PATH = kagglehub.model_download('google/paligemma/jax/paligemma-3b-pt-224', MODEL_PATH)\n", + " print(f\"Model path: {MODEL_PATH}\")\n", + "\n", + "TOKENIZER_PATH = \"./paligemma_tokenizer.model\"\n", + "if not os.path.exists(TOKENIZER_PATH):\n", + " print(\"Downloading the model tokenizer...\")\n", + " !gsutil cp gs://big_vision/paligemma_tokenizer.model {TOKENIZER_PATH}\n", + " print(f\"Tokenizer path: {TOKENIZER_PATH}\")\n", + "\n", + "# DATA_DIR=\"./longcap100\"\n", + "# if not os.path.exists(DATA_DIR):\n", + "# print(\"Downloading the dataset...\")\n", + "# !gsutil -m -q cp -n -r gs://longcap100/ .\n", + "# print(f\"Data path: {DATA_DIR}\")" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "zDoq0O77GF30" + }, + "source": [ + "## Notebook" + ] + }, + { + "cell_type": "code", + "execution_count": 16, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "dTfe2k8J4Bw0", + "outputId": "76b14d9e-b154-4825-aa6e-c52b27fbffcf" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "JAX version: 0.4.26\n", + "JAX platform: gpu\n", + "JAX devices: 1\n" + ] + } + ], + "source": [ + "import base64\n", + "import functools\n", + "import html\n", + "import io\n", + "import os\n", + "import warnings\n", + "\n", + "import jax\n", + "import jax.numpy as jnp\n", + "import numpy as np\n", + "import ml_collections\n", + "\n", + "import tensorflow as tf\n", + "import sentencepiece\n", + "\n", + "from IPython.core.display import display, HTML\n", + "from PIL import Image\n", + "\n", + "# Import model definition from big_vision\n", + "from big_vision.models.proj.paligemma import paligemma\n", + "from big_vision.trainers.proj.paligemma import predict_fns\n", + "\n", + "# Import big vision utilities\n", + "import big_vision.datasets.jsonl\n", + "import big_vision.utils\n", + "import big_vision.sharding\n", + "\n", + "# Don't let TF use the GPU or TPUs\n", + "tf.config.set_visible_devices([], \"GPU\")\n", + "tf.config.set_visible_devices([], \"TPU\")\n", + "\n", + "backend = jax.lib.xla_bridge.get_backend()\n", + "print(f\"JAX version: {jax.__version__}\")\n", + "print(f\"JAX platform: {backend.platform}\")\n", + "print(f\"JAX devices: {jax.device_count()}\")" + ] + }, + { + "cell_type": "code", + "execution_count": 17, + "metadata": { + "id": "1aghcULcEdtv" + }, + "outputs": [], + "source": [ + "# @title Construct model and load params into RAM.\n", + "\n", + "# Define model\n", + "model_config = ml_collections.FrozenConfigDict({\n", + " \"llm\": {\"vocab_size\": 257_152},\n", + " \"img\": {\"variant\": \"So400m/14\", \"pool_type\": \"none\", \"scan\": True, \"dtype_mm\": \"float16\"}\n", + "})\n", + "model = paligemma.Model(**model_config)\n", + "tokenizer = sentencepiece.SentencePieceProcessor(TOKENIZER_PATH)\n", + "\n", + "# Load params - this can take up to 1 minute in T4 colabs.\n", + "params = paligemma.load(None, MODEL_PATH, model_config)\n", + "\n", + "# Define `decode` function to sample outputs from the model.\n", + "decode_fn = predict_fns.get_all(model)['decode']\n", + "decode = functools.partial(decode_fn, devices=jax.devices(), eos_token=tokenizer.eos_id())" + ] + }, + { + "cell_type": "code", + "source": [ + "# model_config = ml_collections.FrozenConfigDict({\n", + "# \"llm\": {\"vocab_size\": 257_152},\n", + "# \"img\": {\"variant\": \"So400m/14\", \"pool_type\": \"none\", \"scan\": True, \"dtype_mm\": \"float16\"}\n", + "# })\n", + "# model = paligemma.Model(**model_config)\n", + "# tokenizer = sentencepiece.SentencePieceProcessor(TOKENIZER_PATH)\n", + "\n", + "# # Load params - this can take up to 1 minute in T4 colabs.\n", + "# params = paligemma.load(None, '/content/fine-tuned-paligemma-3b-pt-224.f16.npz', model_config)\n", + "\n", + "# # Define `decode` function to sample outputs from the model.\n", + "# decode_fn = predict_fns.get_all(model)['decode']\n", + "# decode = functools.partial(decode_fn, devices=jax.devices(), eos_token=tokenizer.eos_id())" + ], + "metadata": { + "id": "2LNRDMMwXFJ9" + }, + "execution_count": 18, + "outputs": [] + }, + { + "cell_type": "code", + "execution_count": 19, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "RWOdf_fw2SAO", + "outputId": "f666f219-5fc8-4cf7-edc3-c9c9e0235e15" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + " == Model params == \n", + "img/Transformer/encoder_norm/bias (1152,) float16\n", + "img/Transformer/encoder_norm/scale (1152,) float16\n", + "img/Transformer/encoderblock/LayerNorm_0/bias (27, 1152) float16\n", + "img/Transformer/encoderblock/LayerNorm_0/scale (27, 1152) float16\n", + "img/Transformer/encoderblock/LayerNorm_1/bias (27, 1152) float16\n", + "img/Transformer/encoderblock/LayerNorm_1/scale (27, 1152) float16\n", + "img/Transformer/encoderblock/MlpBlock_0/Dense_0/bias (27, 4304) float16\n", + "img/Transformer/encoderblock/MlpBlock_0/Dense_0/kernel (27, 1152, 4304) float16\n", + "img/Transformer/encoderblock/MlpBlock_0/Dense_1/bias (27, 1152) float16\n", + "img/Transformer/encoderblock/MlpBlock_0/Dense_1/kernel (27, 4304, 1152) float16\n", + "img/Transformer/encoderblock/MultiHeadDotProductAttention_0/key/bias (27, 16, 72) float16\n", + "img/Transformer/encoderblock/MultiHeadDotProductAttention_0/key/kernel (27, 1152, 16, 72) float16\n", + "img/Transformer/encoderblock/MultiHeadDotProductAttention_0/out/bias (27, 1152) float16\n", + "img/Transformer/encoderblock/MultiHeadDotProductAttention_0/out/kernel (27, 16, 72, 1152) float16\n", + "img/Transformer/encoderblock/MultiHeadDotProductAttention_0/query/bias (27, 16, 72) float16\n", + "img/Transformer/encoderblock/MultiHeadDotProductAttention_0/query/kernel (27, 1152, 16, 72) float16\n", + "img/Transformer/encoderblock/MultiHeadDotProductAttention_0/value/bias (27, 16, 72) float16\n", + "img/Transformer/encoderblock/MultiHeadDotProductAttention_0/value/kernel (27, 1152, 16, 72) float16\n", + "img/embedding/bias (1152,) float16\n", + "img/embedding/kernel (14, 14, 3, 1152) float16\n", + "img/head/bias (2048,) float16\n", + "img/head/kernel (1152, 2048) float16\n", + "img/pos_embedding (1, 256, 1152) float16\n", + "llm/embedder/input_embedding (257152, 2048) float16\n", + "llm/final_norm/scale (2048,) float16\n", + "llm/layers/attn/attn_vec_einsum/w (18, 8, 256, 2048) float32\n", + "llm/layers/attn/kv_einsum/w (18, 2, 1, 2048, 256) float32\n", + "llm/layers/attn/q_einsum/w (18, 8, 2048, 256) float32\n", + "llm/layers/mlp/gating_einsum (18, 2, 2048, 16384) float16\n", + "llm/layers/mlp/linear (18, 16384, 2048) float16\n", + "llm/layers/pre_attention_norm/scale (18, 2048) float16\n", + "llm/layers/pre_ffw_norm/scale (18, 2048) float16\n" + ] + } + ], + "source": [ + "# @title Move params to GPU/TPU memory.\n", + "#\n", + "# To keep HBM usage low and fit in a T4 GPU (16GB HBM) we opt to only finetune\n", + "# a part of the parameters. Additionally we keep the frozen params in float16\n", + "# and cast trainable to float32.\n", + "\n", + "# Create a pytree mask of the trainable params.\n", + "def is_trainable_param(name, param): # pylint: disable=unused-argument\n", + " if name.startswith(\"llm/layers/attn/\"): return True\n", + " if name.startswith(\"llm/\"): return False\n", + " if name.startswith(\"img/\"): return False\n", + " raise ValueError(f\"Unexpected param name {name}\")\n", + "trainable_mask = big_vision.utils.tree_map_with_names(is_trainable_param, params)\n", + "\n", + "#\n", + "# If more than one device is available (e.g. multiple GPUs) the parameters can\n", + "# be sharded across them to reduce HBM usage per device.\n", + "mesh = jax.sharding.Mesh(jax.devices(), (\"data\"))\n", + "\n", + "data_sharding = jax.sharding.NamedSharding(\n", + " mesh, jax.sharding.PartitionSpec(\"data\"))\n", + "\n", + "params_sharding = big_vision.sharding.infer_sharding(\n", + " params, strategy=[('.*', 'fsdp(axis=\"data\")')], mesh=mesh)\n", + "\n", + "# Yes: Some donated buffers are not usable.\n", + "warnings.filterwarnings(\n", + " \"ignore\", message=\"Some donated buffers were not usable\")\n", + "\n", + "@functools.partial(jax.jit, donate_argnums=(0,), static_argnums=(1,))\n", + "def maybe_cast_to_f32(params, trainable):\n", + " return jax.tree.map(lambda p, m: p.astype(jnp.float32) if m else p,\n", + " params, trainable)\n", + "\n", + "# Loading all params in simultaneous - albeit much faster and more succinct -\n", + "# requires more RAM than the T4 colab runtimes have by default (12GB RAM).\n", + "# Instead we do it param by param.\n", + "params, treedef = jax.tree.flatten(params)\n", + "sharding_leaves = jax.tree.leaves(params_sharding)\n", + "trainable_leaves = jax.tree.leaves(trainable_mask)\n", + "for idx, (sharding, trainable) in enumerate(zip(sharding_leaves, trainable_leaves)):\n", + " params[idx] = big_vision.utils.reshard(params[idx], sharding)\n", + " params[idx] = maybe_cast_to_f32(params[idx], trainable)\n", + " params[idx].block_until_ready()\n", + "params = jax.tree.unflatten(treedef, params)\n", + "\n", + "# Print params to show what the model is made of.\n", + "def parameter_overview(params):\n", + " for path, arr in big_vision.utils.tree_flatten_with_names(params)[0]:\n", + " print(f\"{path:80s} {str(arr.shape):22s} {arr.dtype}\")\n", + "\n", + "print(\" == Model params == \")\n", + "parameter_overview(params)" + ] + }, + { + "cell_type": "code", + "execution_count": 20, + "metadata": { + "id": "8SRW0NuU4UcW" + }, + "outputs": [], + "source": [ + "# @title Define preprocess functions to create inputs to the model.\n", + "\n", + "def preprocess_image(image, size=224):\n", + " # Model has been trained to handle images of different aspects ratios\n", + " # resized to 224x224 in the range [-1, 1]. Bilinear and antialias resize\n", + " # options are helpful to improve quality in some tasks.\n", + " image = np.asarray(image)\n", + " if image.ndim == 2: # Convert image without last channel into greyscale.\n", + " image = np.stack((image,)*3, axis=-1)\n", + " image = image[..., :3] # Remove alpha layer.\n", + " assert image.shape[-1] == 3\n", + "\n", + " image = tf.constant(image)\n", + " image = tf.image.resize(image, (size, size), method='bilinear', antialias=True)\n", + " return image.numpy() / 127.5 - 1.0 # [0, 255]->[-1,1]\n", + "\n", + "def preprocess_tokens(prefix, suffix=None, seqlen=None):\n", + " # Model has been trained to handle tokenized text composed of a prefix with\n", + " # full attention and a suffix with causal attention.\n", + " separator = \"\\n\"\n", + " tokens = tokenizer.encode(prefix, add_bos=True) + tokenizer.encode(separator)\n", + " mask_ar = [0] * len(tokens) # 0 to use full attention for prefix.\n", + " mask_loss = [0] * len(tokens) # 0 to not use prefix tokens in the loss.\n", + "\n", + " if suffix:\n", + " suffix = tokenizer.encode(suffix, add_eos=True)\n", + " tokens += suffix\n", + " mask_ar += [1] * len(suffix) # 1 to use causal attention for suffix.\n", + " mask_loss += [1] * len(suffix) # 1 to use suffix tokens in the loss.\n", + "\n", + " mask_input = [1] * len(tokens) # 1 if its a token, 0 if padding.\n", + " if seqlen:\n", + " padding = [0] * max(0, seqlen - len(tokens))\n", + " tokens = tokens[:seqlen] + padding\n", + " mask_ar = mask_ar[:seqlen] + padding\n", + " mask_loss = mask_loss[:seqlen] + padding\n", + " mask_input = mask_input[:seqlen] + padding\n", + "\n", + " return jax.tree.map(np.array, (tokens, mask_ar, mask_loss, mask_input))\n", + "\n", + "def postprocess_tokens(tokens):\n", + " tokens = tokens.tolist() # np.array to list[int]\n", + " try: # Remove tokens at and after EOS if any.\n", + " eos_pos = tokens.index(tokenizer.eos_id())\n", + " tokens = tokens[:eos_pos]\n", + " except ValueError:\n", + " pass\n", + " return tokenizer.decode(tokens)\n" + ] + }, + { + "cell_type": "code", + "execution_count": 21, + "metadata": { + "id": "whzWOojGOtzi" + }, + "outputs": [], + "source": [ + "# @title Function to iterate over train and validation examples.\n", + "SEQLEN = 128\n", + "\n", + "# TODO: Consider data iterators skipping big_vision and tf.data?\n", + "train_dataset = big_vision.datasets.jsonl.DataSource(\n", + " os.path.join(DATA_DIR, \"data_train.jsonl\"),\n", + " fopen_keys={\"image\": DATA_DIR})\n", + "\n", + "val_dataset = big_vision.datasets.jsonl.DataSource(\n", + " os.path.join(DATA_DIR, \"data_val.jsonl\"),\n", + " fopen_keys={\"image\": DATA_DIR})\n", + "\n", + "\n", + "def train_data_iterator():\n", + " \"\"\"Never ending iterator over training examples.\"\"\"\n", + " # Shuffle examples and repeat so one can train for many epochs.\n", + " dataset = train_dataset.get_tfdata().shuffle(1_000).repeat()\n", + " for example in dataset.as_numpy_iterator():\n", + " image = Image.open(io.BytesIO(example[\"image\"]))\n", + " image = preprocess_image(image)\n", + "\n", + " # prefix = \"caption en\" # Could also be a different prefix per example.\n", + " prefix = example[\"prefix\"].decode().lower()\n", + " suffix = example[\"suffix\"].decode().lower()\n", + " tokens, mask_ar, mask_loss, _ = preprocess_tokens(prefix, suffix, SEQLEN)\n", + "\n", + " yield {\n", + " \"image\": np.asarray(image),\n", + " \"text\": np.asarray(tokens),\n", + " \"mask_ar\": np.asarray(mask_ar),\n", + " \"mask_loss\": np.asarray(mask_loss),\n", + " }\n", + "\n", + "\n", + "def validation_data_iterator():\n", + " \"\"\"Single iterator over validation examples.\"\"\"\n", + " for example in val_dataset.get_tfdata(ordered=True).as_numpy_iterator():\n", + " image = Image.open(io.BytesIO(example[\"image\"]))\n", + " image = preprocess_image(image)\n", + "\n", + " # prefix = \"caption en\" # Could also be a different prefix per example.\n", + " prefix = example[\"prefix\"].decode().lower()\n", + " tokens, mask_ar, _, mask_input = preprocess_tokens(prefix, seqlen=SEQLEN)\n", + "\n", + " yield {\n", + " \"image\": np.asarray(image),\n", + " \"text\": np.asarray(tokens),\n", + " \"mask_ar\": np.asarray(mask_ar),\n", + " \"mask_input\": np.asarray(mask_input),\n", + " }\n" + ] + }, + { + "cell_type": "code", + "execution_count": 22, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 298 + }, + "id": "BzJfb5t0nsLq", + "outputId": "fa2b30bc-6fcc-4367-f178-1a4d72155112" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Training examples\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + "
\n", + " \n", + "

<loc0330><loc0522><loc0474><loc0770> fracture ; <loc0578><loc0456><loc0671><loc0655> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0401><loc0047><loc0491><loc0164> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0564><loc0421><loc0649><loc0546> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0532><loc0200><loc0658><loc0358> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0263><loc0205><loc0415><loc0415> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0743><loc0302><loc0792><loc0434> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0396><loc0413><loc0460><loc0501> fracture ; <loc0454><loc0308><loc0526><loc0416> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0751><loc0753><loc0845><loc0876> fracture

\n", + "
\n", + " " + ] + }, + "metadata": {} + } + ], + "source": [ + "# @title Inspect training examples.\n", + "def split_and_keep_second_part(s):\n", + " parts = s.split('\\n', 1)\n", + " if len(parts) > 1:\n", + " return parts[1]\n", + " return s\n", + "\n", + "def render_inline(image, resize=(128, 128)):\n", + " \"\"\"Convert image into inline html.\"\"\"\n", + " image = Image.fromarray(image)\n", + " image.resize(resize)\n", + " with io.BytesIO() as buffer:\n", + " image.save(buffer, format='jpeg')\n", + " image_b64 = str(base64.b64encode(buffer.getvalue()), \"utf-8\")\n", + " return f\"data:image/jpeg;base64,{image_b64}\"\n", + "\n", + "def render_example(image, caption):\n", + " image = ((image + 1)/2 * 255).astype(np.uint8) # [-1,1] -> [0, 255]\n", + " h, w, _ = image.shape\n", + " try:\n", + " detections = from_pali_gemma(caption, (w, h), CLASSES)\n", + " image = sv.BoundingBoxAnnotator().annotate(image, detections)\n", + " except:\n", + " print(\"result render failed, result:\", caption)\n", + " return f\"\"\"\n", + "
\n", + " \n", + "

{html.escape(caption)}

\n", + "
\n", + " \"\"\"\n", + "\n", + "html_out = \"\"\n", + "for idx, example in zip(range(8), train_data_iterator()):\n", + " caption = postprocess_tokens(example[\"text\"]) # detokenize model input.\n", + " caption = split_and_keep_second_part(caption)\n", + " html_out += render_example(example[\"image\"], caption)\n", + "\n", + "print(\"Training examples\")\n", + "display(HTML(html_out))" + ] + }, + { + "cell_type": "code", + "execution_count": 23, + "metadata": { + "id": "dwUV_imW3WQJ" + }, + "outputs": [], + "source": [ + "# @title Define the training step and evaluation loop.\n", + "#\n", + "# The main update_fn using simple SGD.\n", + "#\n", + "@functools.partial(jax.jit, donate_argnums=(0,))\n", + "def update_fn(params, batch, learning_rate):\n", + " imgs, txts, mask_ar = batch[\"image\"], batch[\"text\"], batch[\"mask_ar\"]\n", + "\n", + " def loss_fn(params):\n", + " text_logits, _ = model.apply({\"params\": params}, imgs, txts[:, :-1], mask_ar[:, :-1], train=True)\n", + " logp = jax.nn.log_softmax(text_logits, axis=-1)\n", + "\n", + " # The model takes as input txts[:, :-1] but the loss is defined as predicting\n", + " # next tokens txts[:, 1:]. Additionally, mask_loss[:, 1:] indicates which tokens\n", + " # are part of the loss (e.g. prefix and padded tokens are not included).\n", + " mask_loss = batch[\"mask_loss\"][:, 1:]\n", + " targets = jax.nn.one_hot(txts[:, 1:], text_logits.shape[-1])\n", + "\n", + " # Compute the loss per example. i.e. the mean of per token pplx.\n", + " # Since each example has a different number of tokens we normalize it.\n", + " token_pplx = jnp.sum(logp * targets, axis=-1) # sum across vocab_size.\n", + " example_loss = -jnp.sum(token_pplx * mask_loss, axis=-1) # sum across seq_len.\n", + " example_loss /= jnp.clip(jnp.sum(mask_loss, -1), 1) # weight by num of tokens.\n", + "\n", + " # batch_loss: mean of per example loss.\n", + " return jnp.mean(example_loss)\n", + "\n", + " loss, grads = jax.value_and_grad(loss_fn)(params)\n", + "\n", + " # Apply gradients to trainable params using SGD.\n", + " def apply_grad(param, gradient, trainable):\n", + " if not trainable: return param\n", + " return param - learning_rate * gradient\n", + "\n", + " params = jax.tree_util.tree_map(apply_grad, params, grads, trainable_mask)\n", + "\n", + " return params, loss\n", + "\n", + "# Evaluation/inference loop.\n", + "def make_predictions(data_iterator, *, num_examples=None,\n", + " batch_size=4, seqlen=SEQLEN, sampler=\"greedy\"):\n", + " outputs = []\n", + " while True:\n", + " # Construct a list of examples in the batch.\n", + " examples = []\n", + " try:\n", + " for _ in range(batch_size):\n", + " examples.append(next(data_iterator))\n", + " examples[-1][\"_mask\"] = np.array(True) # Indicates true example.\n", + " except StopIteration:\n", + " if len(examples) == 0:\n", + " return outputs\n", + "\n", + " # Not enough examples to complete a batch. Pad by repeating last example.\n", + " while len(examples) % batch_size:\n", + " examples.append(dict(examples[-1]))\n", + " examples[-1][\"_mask\"] = np.array(False) # Indicates padding example.\n", + "\n", + " # Convert list of examples into a dict of np.arrays and load onto devices.\n", + " batch = jax.tree.map(lambda *x: np.stack(x), *examples)\n", + " batch = big_vision.utils.reshard(batch, data_sharding)\n", + "\n", + " # Make model predictions\n", + " tokens = decode({\"params\": params}, batch=batch,\n", + " max_decode_len=seqlen, sampler=sampler)\n", + "\n", + " # Fetch model predictions to device and detokenize.\n", + " tokens, mask = jax.device_get((tokens, batch[\"_mask\"]))\n", + " tokens = tokens[mask] # remove padding examples.\n", + " responses = [postprocess_tokens(t) for t in tokens]\n", + "\n", + " # Append to html output.\n", + " for example, response in zip(examples, responses):\n", + " outputs.append((example[\"image\"], response))\n", + " if num_examples and len(outputs) >= num_examples:\n", + " return outputs" + ] + }, + { + "cell_type": "code", + "execution_count": 24, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 1000 + }, + "id": "067wj_6bZAG3", + "outputId": "50132bea-0b74-4b1a-9049-1a56432b32fd" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "step: 1/64 lr: 0.00167 loss: 4.6046\n", + "Model predictions at step 1\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + "
\n", + " \n", + "

<loc0000><loc0000><loc0900><loc1015> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0000><loc0000><loc1023><loc1023> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0105><loc0000><loc0964><loc1023> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0000><loc0000><loc1015><loc1023> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0373><loc0238><loc0575><loc0432> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0000><loc0000><loc1015><loc1023> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0679><loc0549><loc0735><loc0670> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0000><loc0000><loc1017><loc0767> fracture

\n", + "
\n", + " " + ] + }, + "metadata": {} + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "step: 2/64 lr: 0.00333 loss: 3.9101\n", + "step: 3/64 lr: 0.00500 loss: 3.8219\n", + "step: 4/64 lr: 0.00667 loss: 3.7913\n", + "step: 5/64 lr: 0.00833 loss: 3.9509\n", + "step: 6/64 lr: 0.01000 loss: 4.0808\n", + "step: 7/64 lr: 0.00999 loss: 3.9368\n", + "step: 8/64 lr: 0.00997 loss: 3.6630\n", + "Model predictions at step 8\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + "
\n", + " \n", + "

<loc0612><loc0549><loc0870><loc0651> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0000><loc0000><loc1023><loc1023> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0110><loc0000><loc0969><loc1023> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0000><loc0000><loc1013><loc1023> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0105><loc0425><loc0167><loc0509> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0000><loc0000><loc1015><loc1023> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0693><loc0513><loc0767><loc0826> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0517><loc0621><loc0583><loc0738> fracture

\n", + "
\n", + " " + ] + }, + "metadata": {} + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "step: 9/64 lr: 0.00994 loss: 3.7152\n", + "step: 10/64 lr: 0.00989 loss: 4.1957\n", + "step: 11/64 lr: 0.00982 loss: 3.8173\n", + "step: 12/64 lr: 0.00975 loss: 4.0603\n", + "step: 13/64 lr: 0.00966 loss: 3.6232\n", + "step: 14/64 lr: 0.00955 loss: 3.8549\n", + "step: 15/64 lr: 0.00944 loss: 3.6587\n", + "step: 16/64 lr: 0.00931 loss: 3.6792\n", + "Model predictions at step 16\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + "
\n", + " \n", + "

<loc0603><loc0549><loc0841><loc0658> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0441><loc0230><loc0541><loc0344> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0620><loc0477><loc0714><loc0629> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0458><loc0000><loc0531><loc0127> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0559><loc0279><loc0617><loc0413> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0591><loc0036><loc0645><loc0238> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0485><loc0728><loc0541><loc0944> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0536><loc0621><loc0603><loc0738> fracture

\n", + "
\n", + " " + ] + }, + "metadata": {} + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "step: 17/64 lr: 0.00917 loss: 3.4137\n", + "step: 18/64 lr: 0.00901 loss: 3.8984\n", + "step: 19/64 lr: 0.00885 loss: 3.8873\n", + "step: 20/64 lr: 0.00867 loss: 3.7505\n", + "step: 21/64 lr: 0.00849 loss: 3.7748\n", + "step: 22/64 lr: 0.00829 loss: 3.6498\n", + "step: 23/64 lr: 0.00809 loss: 3.6461\n", + "step: 24/64 lr: 0.00787 loss: 3.7160\n", + "Model predictions at step 24\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + "
\n", + " \n", + "

<loc0603><loc0559><loc0850><loc0653> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0767><loc0505><loc1023><loc0754> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0596><loc0000><loc0726><loc0127> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0303><loc0387><loc0409><loc0529> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0559><loc0302><loc0611><loc0425> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0591><loc0000><loc0652><loc0206> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0503><loc0000><loc0561><loc0127> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0559><loc0620><loc0623><loc0754> fracture

\n", + "
\n", + " " + ] + }, + "metadata": {} + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "step: 25/64 lr: 0.00765 loss: 3.7272\n", + "step: 26/64 lr: 0.00742 loss: 3.4325\n", + "step: 27/64 lr: 0.00719 loss: 3.7337\n", + "step: 28/64 lr: 0.00694 loss: 3.5439\n", + "step: 29/64 lr: 0.00670 loss: 3.8029\n", + "step: 30/64 lr: 0.00644 loss: 3.5595\n", + "step: 31/64 lr: 0.00619 loss: 3.4872\n", + "step: 32/64 lr: 0.00593 loss: 3.4479\n", + "Model predictions at step 32\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + "
\n", + " \n", + "

<loc0575><loc0559><loc0850><loc0647> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0754><loc0485><loc1023><loc0714> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0588><loc0485><loc0677><loc0595> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0425><loc0000><loc0509><loc0100> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0525><loc0284><loc0595><loc0438> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0559><loc0000><loc0629><loc0206> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0456><loc0000><loc0536><loc0174> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0483><loc0621><loc0559><loc0735> fracture

\n", + "
\n", + " " + ] + }, + "metadata": {} + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "step: 33/64 lr: 0.00566 loss: 3.4278\n", + "step: 34/64 lr: 0.00540 loss: 3.7668\n", + "step: 35/64 lr: 0.00513 loss: 3.5646\n", + "step: 36/64 lr: 0.00487 loss: 3.2581\n", + "step: 37/64 lr: 0.00460 loss: 3.3962\n", + "step: 38/64 lr: 0.00434 loss: 3.6364\n", + "step: 39/64 lr: 0.00407 loss: 3.3160\n", + "step: 40/64 lr: 0.00381 loss: 3.5750\n", + "Model predictions at step 40\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + "
\n", + " \n", + "

<loc0589><loc0572><loc0867><loc0653> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0762><loc0670><loc1023><loc0754> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0620><loc0505><loc0703><loc0601> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0432><loc0000><loc0509><loc0113> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0547><loc0302><loc0626><loc0425> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0583><loc0000><loc0664><loc0230> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0478><loc0000><loc0549><loc0097> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0509><loc0639><loc0582><loc0735> fracture

\n", + "
\n", + " " + ] + }, + "metadata": {} + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "step: 41/64 lr: 0.00356 loss: 3.5534\n", + "step: 42/64 lr: 0.00330 loss: 3.7601\n", + "step: 43/64 lr: 0.00306 loss: 3.4201\n", + "step: 44/64 lr: 0.00281 loss: 3.2198\n", + "step: 45/64 lr: 0.00258 loss: 3.8257\n", + "step: 46/64 lr: 0.00235 loss: 3.4511\n", + "step: 47/64 lr: 0.00213 loss: 3.4098\n", + "step: 48/64 lr: 0.00191 loss: 3.5586\n", + "Model predictions at step 48\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + "
\n", + " \n", + "

<loc0575><loc0559><loc0870><loc0646> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0419><loc0230><loc0541><loc0344> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0620><loc0485><loc0703><loc0591> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0430><loc0000><loc0514><loc0100> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0529><loc0290><loc0611><loc0425> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0575><loc0000><loc0658><loc0216> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0478><loc0000><loc0561><loc0097> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0496><loc0633><loc0579><loc0738> fracture

\n", + "
\n", + " " + ] + }, + "metadata": {} + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "step: 49/64 lr: 0.00171 loss: 3.3370\n", + "step: 50/64 lr: 0.00151 loss: 3.6179\n", + "step: 51/64 lr: 0.00133 loss: 3.8233\n", + "step: 52/64 lr: 0.00115 loss: 3.4149\n", + "step: 53/64 lr: 0.00099 loss: 3.6569\n", + "step: 54/64 lr: 0.00083 loss: 3.4282\n", + "step: 55/64 lr: 0.00069 loss: 3.3576\n", + "step: 56/64 lr: 0.00056 loss: 3.5013\n", + "Model predictions at step 56\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + "
\n", + " \n", + "

<loc0583><loc0559><loc0870><loc0647> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0419><loc0230><loc0541><loc0344> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0620><loc0485><loc0703><loc0601> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0430><loc0000><loc0518><loc0104> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0529><loc0290><loc0602><loc0425> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0575><loc0000><loc0652><loc0223> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0478><loc0000><loc0561><loc0113> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0496><loc0621><loc0582><loc0754> fracture

\n", + "
\n", + " " + ] + }, + "metadata": {} + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "step: 57/64 lr: 0.00045 loss: 3.5114\n", + "step: 58/64 lr: 0.00034 loss: 3.3903\n", + "step: 59/64 lr: 0.00025 loss: 3.3435\n", + "step: 60/64 lr: 0.00018 loss: 3.2968\n", + "step: 61/64 lr: 0.00011 loss: 3.4862\n", + "step: 62/64 lr: 0.00006 loss: 3.6225\n", + "step: 63/64 lr: 0.00003 loss: 3.4152\n", + "step: 64/64 lr: 0.00001 loss: 3.6450\n", + "Model predictions at step 64\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + "
\n", + " \n", + "

<loc0583><loc0559><loc0870><loc0653> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0419><loc0230><loc0541><loc0344> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0620><loc0485><loc0703><loc0601> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0430><loc0000><loc0514><loc0104> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0529><loc0290><loc0602><loc0438> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0577><loc0000><loc0652><loc0223> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0478><loc0000><loc0561><loc0127> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0496><loc0621><loc0582><loc0754> fracture

\n", + "
\n", + " " + ] + }, + "metadata": {} + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "CPU times: user 11min 24s, sys: 743 ms, total: 11min 24s\n", + "Wall time: 11min 27s\n" + ] + } + ], + "source": [ + "# @title Run training loop.\n", + "#\n", + "# Run a short training loop with cosine learning rate schedule.\n", + "#\n", + "# Note: the first step can be quite slow on some machines (up to several minutes)\n", + "# due to XLA compilation of the jax.jit'd function.\n", + "#\n", + "%%time\n", + "\n", + "BATCH_SIZE = 8\n", + "TRAIN_EXAMPLES = 512\n", + "# TRAIN_EXAMPLES = 256\n", + "LEARNING_RATE = 0.01\n", + "\n", + "TRAIN_STEPS = TRAIN_EXAMPLES // BATCH_SIZE\n", + "EVAL_STEPS = TRAIN_STEPS // 8\n", + "\n", + "train_data_it = train_data_iterator()\n", + "\n", + "sched_fn = big_vision.utils.create_learning_rate_schedule(\n", + " total_steps=TRAIN_STEPS+1, base=LEARNING_RATE,\n", + " decay_type=\"cosine\", warmup_percent=0.10)\n", + "\n", + "for step in range(1, TRAIN_STEPS+1):\n", + " # Make list of N training examples.\n", + " examples = [next(train_data_it) for _ in range(BATCH_SIZE)]\n", + "\n", + " # Convert list of examples into a dict of np.arrays and load onto devices.\n", + " batch = jax.tree.map(lambda *x: np.stack(x), *examples)\n", + " batch = big_vision.utils.reshard(batch, data_sharding)\n", + "\n", + " # Training step and report training loss\n", + " learning_rate = sched_fn(step)\n", + " params, loss = update_fn(params, batch, learning_rate)\n", + "\n", + " loss = jax.device_get(loss)\n", + " print(f\"step: {step:2d}/{TRAIN_STEPS:2d} lr: {learning_rate:.5f} loss: {loss:.4f}\")\n", + "\n", + " if step == 1 or (step % EVAL_STEPS) == 0:\n", + " print(f\"Model predictions at step {step}\")\n", + " html_out = \"\"\n", + " for image, caption in make_predictions(\n", + " validation_data_iterator(), num_examples=8, batch_size=8):\n", + " html_out += render_example(image, caption)\n", + " display(HTML(html_out))\n" + ] + }, + { + "cell_type": "code", + "execution_count": 25, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 1000 + }, + "id": "hgUhEKjzPdMQ", + "outputId": "184cfc14-a20c-4898-cf0e-6c36210375b3" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Model predictions\n", + "result render failed, result: \n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + "
\n", + " \n", + "

<loc0583><loc0559><loc0870><loc0653> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0419><loc0230><loc0541><loc0344> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0620><loc0485><loc0703><loc0601> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0430><loc0000><loc0514><loc0104> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0529><loc0290><loc0602><loc0438> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0577><loc0000><loc0652><loc0223> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0478><loc0000><loc0561><loc0127> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0496><loc0621><loc0582><loc0754> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0413><loc0201><loc0582><loc0270> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0166><loc0425><loc0292><loc0726> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0321><loc0000><loc0425><loc0091> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0566><loc0284><loc0616><loc0413> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0453><loc0303><loc0527><loc0397> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0620><loc0441><loc0703><loc0575> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0559><loc0366><loc0652><loc0513> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0485><loc0366><loc0726><loc0503> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0363><loc0453><loc0575><loc0559> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0303><loc0465><loc0376><loc0629> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0366><loc0477><loc0465><loc0714> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0321><loc0302><loc0575><loc0465> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0529><loc0503><loc0675><loc0714> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0677><loc0300><loc0772><loc0453> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0413><loc0270><loc0698><loc0477> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0596><loc0200><loc0714><loc0387> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0463><loc0719><loc0543><loc0826> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0453><loc0421><loc0617><loc0509> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0409><loc0344><loc0514><loc0509> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0637><loc0293><loc0703><loc0366> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0403><loc0230><loc0812><loc0529> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0456><loc0344><loc0549><loc0485> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0238><loc0150><loc0313><loc0313> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0263><loc0366><loc0456><loc0467> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0483><loc0000><loc0629><loc0113> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0270><loc0639><loc0357><loc0735> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0279><loc0667><loc0357><loc0944> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0575><loc0754><loc0639><loc0845> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0470><loc0309><loc0575><loc0513> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0583><loc0054><loc0652><loc0211> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0366><loc0143><loc0438><loc0509> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0525><loc0000><loc0609><loc0127> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0438><loc0000><loc0596><loc0100> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0279><loc0344><loc0392><loc0559> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0559><loc0290><loc0645><loc0413> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0559><loc0313><loc0677><loc0453> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0485><loc0216><loc0595><loc0438> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0303><loc0453><loc0382><loc0653> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0431><loc0127><loc0569><loc0302> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0488><loc0471><loc0644><loc0582> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0463><loc0000><loc0541><loc0168> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0425><loc0465><loc0629><loc0559> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0344><loc0000><loc0424><loc0200> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0373><loc0425><loc0456><loc0629> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0620><loc0167><loc0703><loc0344> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0575><loc0629><loc0714><loc0797> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0575><loc0485><loc0687><loc0629> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0403><loc0601><loc0591><loc0710> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0251><loc0000><loc0409><loc0366> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0363><loc0290><loc0686><loc0582> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0403><loc0767><loc0477><loc0921> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0182><loc0113><loc0292><loc0303> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0620><loc0073><loc0726><loc0238> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0575><loc0425><loc0767><loc0513> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0465><loc0425><loc0568><loc0531> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0425><loc0579><loc0559><loc0646> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0559><loc0541><loc0662><loc0896> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0653><loc0754><loc0797><loc0870> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0363><loc0425><loc0453><loc0629> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0303><loc0442><loc0399><loc0575> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0579><loc0093><loc0658><loc0219> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0603><loc0074><loc0689><loc0267> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0456><loc0106><loc0527><loc0230> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0396><loc0297><loc0470><loc0474> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0435><loc0582><loc0505><loc0726> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0419><loc0667><loc0509><loc0896> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0175><loc0603><loc0344><loc0726> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0418><loc0152><loc0510><loc0271> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0425><loc0384><loc0582><loc0611> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0438><loc0000><loc0714><loc1023> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0509><loc0000><loc0603><loc0194> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0419><loc0667><loc0502><loc0887> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0456><loc0403><loc0559><loc0644> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0582><loc0363><loc0653><loc0505> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0363><loc0074><loc0485><loc0279> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0000><loc0617><loc0113><loc0721> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0366><loc0297><loc0438><loc0453> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0496><loc0453><loc0591><loc0549> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0408><loc0176><loc0477><loc0293> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0366><loc0505><loc0453><loc0870> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0375><loc0366><loc0467><loc0456> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0797><loc0529><loc0921><loc0754> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0453><loc0000><loc0559><loc1023> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0165><loc0735><loc0366><loc0826> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0199><loc0366><loc0425><loc0612> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0596><loc0211><loc0703><loc0344> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0366><loc0000><loc0456><loc0206> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0373><loc0211><loc0453><loc0513> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0448><loc0100><loc0527><loc0238> fracture

\n", + "
\n", + " \n", + "
\n", + " \n", + "

<loc0493><loc0620><loc0629><loc0714> fracture

\n", + "
\n", + " " + ] + }, + "metadata": {} + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "CPU times: user 59.1 s, sys: 202 ms, total: 59.3 s\n", + "Wall time: 1min 5s\n" + ] + } + ], + "source": [ + "# @title Evaluate the model on all examples.\n", + "#\n", + "# The validation data consists of 10 images in a different domain than training\n", + "# data.\n", + "%%time\n", + "\n", + "print(\"Model predictions\")\n", + "html_out = \"\"\n", + "for image, caption in make_predictions(validation_data_iterator(), batch_size=4):\n", + " html_out += render_example(image, caption)\n", + "display(HTML(html_out))\n" + ] + }, + { + "cell_type": "markdown", + "source": [ + "## Save fine-tuned model locally" + ], + "metadata": { + "id": "Hr1gTKP8trRb" + } + }, + { + "cell_type": "code", + "source": [ + "flat, _ = big_vision.utils.tree_flatten_with_names(params)\n", + "with open(\"/content/fine-tuned-paligemma-3b-pt-224.f16.npz\", \"wb\") as f:\n", + " np.savez(f, **{k: v for k, v in flat})" + ], + "metadata": { + "id": "zyVxKr2FOxPe" + }, + "execution_count": 26, + "outputs": [] + } + ], + "metadata": { + "accelerator": "GPU", + "colab": { + "gpuType": "T4", + "provenance": [], + "machine_shape": "hm" + }, + "kernelspec": { + "display_name": "Python 3", + "name": "python3" + }, + "language_info": { + "name": "python" + } + }, + "nbformat": 4, + "nbformat_minor": 0 +} \ No newline at end of file