\n",
"Int64Index: 2946 entries, 1591 to 6131\n",
"Data columns (total 4 columns):\n",
" # Column Non-Null Count Dtype \n",
"--- ------ -------------- ----- \n",
" 0 tweet_id 2946 non-null object \n",
" 1 safe_text 2946 non-null object \n",
" 2 label 2946 non-null float64\n",
" 3 agreement 2946 non-null float64\n",
"dtypes: float64(2), object(2)\n",
"memory usage: 115.1+ KB\n"
]
}
]
},
{
"cell_type": "code",
"source": [
"# Save splitted subsets\n",
"\n",
"# Define file path\n",
"\n",
"file_path = '/content/drive/MyDrive/LP5/Career_Accelerator_P5-NLP'\n",
"\n",
"#\"/content/drive/MyDrive/LP5/Career_Accelerator_P5-NLP\"\n",
"\n",
"train.to_csv(os.path.join(file_path, \"train_subset.csv\"), index=False)\n",
"eval.to_csv(os.path.join(file_path, \"eval_subset.csv\"), index=False)"
],
"metadata": {
"id": "dX7PPpfWYYEH"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"source": [
"# Load the CSV files into a dataset\n",
"\n",
"dataset = load_dataset('csv', data_files={\n",
" 'train': '/content/drive/MyDrive/LP5/Career_Accelerator_P5-NLP/train_subset.csv',\n",
" 'eval': '/content/drive/MyDrive/LP5/Career_Accelerator_P5-NLP/eval_subset.csv'\n",
"}, encoding='ISO-8859-1')"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 145,
"referenced_widgets": [
"8317af457cae42a1abc4e51f51005024",
"2f69dfb451ef4d299ab8426fa13d9cc9",
"cee8221d7b604daaa0410fa43a7e02c6",
"93e0888ab59c431788bde53b47c9f1a7",
"5d6e06ad7a404ebe9989eac1ad8088b8",
"15c30df19dc34bfd838a8dab09ffc539",
"a4f9f99512df4b0e855255b723bb9a32",
"452164c70ad74ba9ba5d7a051511e7a4",
"a945817118b2490981c8c2523c0d14b7",
"39afc12f9f79453e825dc53ce47f430d",
"29290f8590e046419de099ef06f161e7",
"695b654db12041a585e86db828de5d95",
"c41e5e6cec324f5db130622308cde314",
"322b85b8208645f4985f8e9c2c49ad3f",
"c1efaccf258c4e289f18fa95163506d8",
"2e30f2e5d1a44a08b03b96afbb9f13ef",
"75246d3c183a466d9568801d029e83e4",
"fa5b047a367242b3b5752f9341075e38",
"ed2608efb1984c1da3b16a5e6a8cd721",
"288e22fe04324ce1a58f90d0eb20c364",
"404ff5cd99824b19ad9ab92259a16f6c",
"cd733dfbddad4cd897ff4de0ac59d159",
"6eb746d967ac4f8eb177ba7993195dfa",
"d34d49386cd44fccb961612f3e6927f4",
"9a43b8a0d1de4071bd5ad07e03f38ec8",
"360864a297ef455db7119b09eaf6187d",
"b3ae9602ed824d0d9404e685866b781b",
"1b54675ca28b42d0b9392d01aae335b4",
"2a42963e11534bab9a4e589afd48b088",
"ea67b81733974c20920b2f980e30fb7c",
"0b9cc75ee18e471792adb2c9cff89c84",
"d9109a472e8d47e7a3a703307dce6206",
"6e5d529bf7274b4ea671ff62c16c0b65",
"71c83f53f66a4d7c9fe937679a4e53cf",
"d0a9d85156f34ca98c39b52ddbc13fb6",
"cdd36713dac848afbfeeb32c0c77056b",
"95ea838a1cb74ab2aba597fe5360fe06",
"2ce24f7ad8724307b64f279113fb6141",
"04ae3b6e5f0b484ab742dc68f7012adc",
"74681abc93d24b538f95bb9da0d2771e",
"fd101d83e06a42ad842994c03dd5daff",
"8ff9991cd97c402694894100d7be585e",
"a5a4f8535b2e4e638260acfe5ba1dbc8",
"7fbc9531f78144bb91272299fd03fc10"
]
},
"id": "ENDoL3ObY1H6",
"outputId": "378a8c0d-2634-4f8e-de51-2b8c870d0bec"
},
"execution_count": null,
"outputs": [
{
"output_type": "display_data",
"data": {
"text/plain": [
"Downloading data files: 0%| | 0/2 [00:00\n",
" \n",
" \n",
" [406/552 39:47 < 14:22, 0.17 it/s, Epoch 2.20/3]\n",
" \n",
" \n",
" \n",
" \n",
" | Step | \n",
" Training Loss | \n",
" Validation Loss | \n",
"
\n",
" \n",
" \n",
" \n",
"
"
],
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
},
{
"output_type": "display_data",
"data": {
"text/plain": [
""
],
"text/html": [
"\n",
" \n",
" \n",
"
\n",
" [458/552 44:56 < 09:15, 0.17 it/s, Epoch 2.48/3]\n",
"
\n",
" \n",
" \n",
" | Step | \n",
" Training Loss | \n",
" Validation Loss | \n",
"
\n",
" \n",
" \n",
" \n",
"
"
]
},
"metadata": {}
}
]
},
{
"cell_type": "code",
"source": [
"# Evaluate the model\n",
"eval_results = trainer_ROBERTA.evaluate()\n",
"\n",
"# Create a dictionary of the evaluation results\n",
"results_dict = {\n",
" \"Model\": \"roberta-base\",\n",
" \"Loss\": eval_results[\"eval_loss\"],\n",
" \"RMSE\": eval_results[\"eval_rmse\"],\n",
" \"Runtime\": eval_results[\"eval_runtime\"],\n",
" \"Samples Per Second\": eval_results[\"eval_samples_per_second\"],\n",
" \"Steps Per Second\": eval_results[\"eval_steps_per_second\"],\n",
" \"Epoch\": eval_results[\"epoch\"]\n",
"}\n",
"\n",
"# Create a pandas DataFrame from the dictionary\n",
"results_df = pd.DataFrame([results_dict])\n",
"\n",
"# Sort the results by \"eval_rmse\" in ascending order and get the name and state dict of the best model\n",
"best_model = results_df.loc[results_df['f1_macro'].idxmin()]\n",
"\n",
"print(best_model)"
],
"metadata": {
"id": "xrjdNpstawTP",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 235
},
"outputId": "06812a13-e52a-4a6b-9c9e-539f385e5a75"
},
"execution_count": null,
"outputs": [
{
"output_type": "error",
"ename": "NameError",
"evalue": "ignored",
"traceback": [
"\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
"\u001b[0;31mNameError\u001b[0m Traceback (most recent call last)",
"\u001b[0;32m\u001b[0m in \u001b[0;36m| \u001b[0;34m()\u001b[0m\n\u001b[1;32m 1\u001b[0m \u001b[0;31m# Evaluate the model\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 2\u001b[0;31m \u001b[0meval_results\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtrainer_ROBERTA\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mevaluate\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 3\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 4\u001b[0m \u001b[0;31m# Create a dictionary of the evaluation results\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 5\u001b[0m results_dict = {\n",
"\u001b[0;31mNameError\u001b[0m: name 'trainer_ROBERTA' is not defined"
]
}
]
},
{
"cell_type": "code",
"source": [
"# Push the final fine-tuned model to the Hugging Face model hub\n",
"\n",
"trainer_ROBERTA.push_to_hub (\"MissChloe/PQ_Roberta_Model\")"
],
"metadata": {
"id": "hCOqdoteg0aB"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"source": [
"tokenizer_ROBERTA.push_to_hub (\"MissChloe/PQ_Roberta_Model\")"
],
"metadata": {
"id": "k_uAB3cXkJnK"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"source": [
"model_ROBERTA.push_to_hub(\"MissChloe/PQ_Roberta_Model\")"
],
"metadata": {
"id": "n7IzT4d6kUsm"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"source": [
"# Load the tokenizer\n",
"tokenizer = tokenizer_ROBERTA.from_pretrained(\"slickdata/finetuned-Sentiment-classfication-ROBERTA-model\")\n",
"\n",
"# Load the fine-tuned model\n",
"model = pipeline(\"text-classification\", model=\"MissChloe/PQ_Roberta_Model\", tokenizer=tokenizer)"
],
"metadata": {
"id": "9gwaLN2hkcQd"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"source": [
"label_map = {0: \"negative\", 1: \"neutral\", 2: \"positive\"}\n",
"\n",
"# Make predictions on some example text\n",
"result = model(\"I love these covid vaccines.\")\n",
"\n",
"# Map the numerical label to the corresponding class name\n",
"result[0][\"label\"] = label_map[int(result[0][\"label\"].split(\"_\")[1])]\n",
"\n",
"# Print the predicted label and score\n",
"print(result)"
],
"metadata": {
"id": "13tRokU8kv4S"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"source": [
"!pip freeze >"
],
"metadata": {
"id": "5dwzpHGlk08E"
},
"execution_count": null,
"outputs": []
}
]
} |