FISH BREED CLASSIFICATION

Credit: AITS Cainvas Community

Photo by July PJuxa (Puchkova) on Dribbble

IMPORT LIBRARY AND DATA

!pip install wget
import numpy as np 
import pandas as pd
import os
import matplotlib.pyplot as plt
import seaborn as sns
import cv2
from sklearn.model_selection import StratifiedShuffleSplit,train_test_split
import glob
import tensorflow as tf
from sklearn import metrics
from tensorflow import keras
!wget -N 'https://cainvas-static.s3.amazonaws.com/media/user_data/cainvas-admin/fish.tar.gz'
import tarfile
file = tarfile.open('fish.tar.gz')
file.extractall('./')
file.close()

Defaulting to user installation because normal site-packages is not writeable
Requirement already satisfied: wget in ./.local/lib/python3.7/site-packages (3.2)
WARNING: You are using pip version 20.3.1; however, version 21.1.3 is available.
You should consider upgrading via the '/opt/tljh/user/bin/python -m pip install --upgrade pip' command.
--2021-07-13 10:18:37--  https://cainvas-static.s3.amazonaws.com/media/user_data/cainvas-admin/fish.tar.gz
Resolving cainvas-static.s3.amazonaws.com (cainvas-static.s3.amazonaws.com)... 52.219.66.88
Connecting to cainvas-static.s3.amazonaws.com (cainvas-static.s3.amazonaws.com)|52.219.66.88|:443... connected.
HTTP request sent, awaiting response... 304 Not Modified
File ‘fish.tar.gz’ not modified on server. Omitting download.

BASIC ANALYSIS

data = pd.read_csv('fish/data.csv')
display(data)
sns.set_style('darkgrid')
sns.countplot(data = data,y = 'label')

	path	label
0	fish/Black Sea Sprat/00033.png	Black Sea Sprat
1	fish/Black Sea Sprat/00004.png	Black Sea Sprat
2	fish/Black Sea Sprat/00023.png	Black Sea Sprat
3	fish/Black Sea Sprat/00041.png	Black Sea Sprat
4	fish/Black Sea Sprat/00024.png	Black Sea Sprat
...	...	...
425	fish/Gilt Head Bream/14.JPG	Gilt Head Bream
426	fish/Gilt Head Bream/18.JPG	Gilt Head Bream
427	fish/Gilt Head Bream/49.JPG	Gilt Head Bream
428	fish/Gilt Head Bream/2.JPG	Gilt Head Bream
429	fish/Gilt Head Bream/25.JPG	Gilt Head Bream

430 rows × 2 columns

<AxesSubplot:xlabel='count', ylabel='label'>

def get_img(index):
    img = plt.imread(data.iloc[index]['path'])
    return img
demo = pd.DataFrame(data.groupby('label').first()).reset_index()
fig,ax = plt.subplots(3,3,figsize=(15,15))
for i in range(3):
    for j in range(3):
        ax[i][j].imshow(plt.imread(demo.iloc[i*3+j].path))
        ax[i][j].tick_params(left = False, bottom = False)
        ax[i][j].set_xticks([])
        ax[i][j].set_yticks([])
        ax[i][j].set_title(demo.iloc[i*3+j].label)
display(demo)
fig.savefig('Sample.png')

	label	path
0	Black Sea Sprat	fish/Black Sea Sprat/00033.png
1	Gilt Head Bream	fish/Gilt Head Bream/45.JPG
2	Horse Mackerel	fish/Horse Mackerel/00033.png
3	Red Mullet	fish/Red Mullet/00033.png
4	Red Sea Bream	fish/Red Sea Bream/SAM_5577.JPG
5	Sea Bass	fish/Sea Bass/45.JPG
6	Shrimp	fish/Shrimp/00033.png
7	Striped Red Mullet	fish/Striped Red Mullet/00033.png
8	Trout	fish/Trout/00004.png

MODEL

def get_model(dim = (288,384,3)):
    keras.backend.clear_session()
    model = keras.models.Sequential()
    model.add(keras.layers.Conv2D(32,(3,3),activation='relu',input_shape=dim))
    model.add(keras.layers.MaxPooling2D((2,2)))
    model.add(keras.layers.Dropout(.4))
    model.add(keras.layers.Conv2D(32,(3,3),activation='relu',padding="valid"))
    model.add(keras.layers.MaxPooling2D((2,2)))
    model.add(keras.layers.Dropout(.4))
    model.add(keras.layers.Conv2D(64,(3,3),activation='relu',padding="valid"))
    model.add(keras.layers.MaxPooling2D((2,2)))
    model.add(keras.layers.Dropout(.4))
    model.add(keras.layers.Conv2D(64,(5,5),activation='relu',padding="valid"))
    model.add(keras.layers.MaxPooling2D((2,2)))
    model.add(keras.layers.Dropout(.4))
    model.add(keras.layers.Conv2D(64,(5,5),activation='relu',padding="valid"))
    model.add(keras.layers.MaxPooling2D((2,2)))
    model.add(keras.layers.Dropout(.4))
    model.add(keras.layers.Flatten()) 
    model.add(keras.layers.Dropout(.4))
    model.add(keras.layers.Dense(9,activation='softmax'))
    model.summary()
    return model
train_val,test = train_test_split(data,stratify=data['label'],test_size=45,shuffle=True)
train,val  = train_test_split(train_val,stratify=train_val['label'],test_size=.25,shuffle=True)
train.reset_index(drop=True,inplace=True)
val.reset_index(drop=True,inplace=True)
test.reset_index(drop=True,inplace=True)
print(len(train),len(val),len(test))

288 97 45

datagen = keras.preprocessing.image.ImageDataGenerator(rescale=1/255.,
                                  horizontal_flip = True,
                                  vertical_flip=True,)
datagen_scale = keras.preprocessing.image.ImageDataGenerator(rescale=1/255)
train_generator= datagen.flow_from_dataframe(train,x_col ='path',y_col='label',target_size=(288,384),batch_size=32,shuffle=True)
valid_generator= datagen.flow_from_dataframe(val,x_col ='path',y_col='label',target_size=(288,384),shuffle=False)
test_generator = datagen.flow_from_dataframe(test,x_col ='path',y_col='label',target_size=(288,384),shuffle=False)

Found 288 validated image filenames belonging to 9 classes.
Found 97 validated image filenames belonging to 9 classes.
Found 45 validated image filenames belonging to 9 classes.

model = get_model()
model.compile(loss='categorical_crossentropy',metrics='accuracy',optimizer=keras.optimizers.Adam(learning_rate=.0008))
early_stop = keras.callbacks.EarlyStopping(monitor= 'val_accuracy',patience=12,restore_best_weights=True)
red_lr = keras.callbacks.ReduceLROnPlateau(patience=4,monitor= 'val_accuracy',factor=0.75)
model_check = keras.callbacks.ModelCheckpoint(f"fishes.h5",save_best_only=True)
history = model.fit(train_generator,validation_data=valid_generator,epochs=200,callbacks = [early_stop,model_check,red_lr])

Model: "sequential"
_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
conv2d (Conv2D)              (None, 286, 382, 32)      896       
_________________________________________________________________
max_pooling2d (MaxPooling2D) (None, 143, 191, 32)      0         
_________________________________________________________________
dropout (Dropout)            (None, 143, 191, 32)      0         
_________________________________________________________________
conv2d_1 (Conv2D)            (None, 141, 189, 32)      9248      
_________________________________________________________________
max_pooling2d_1 (MaxPooling2 (None, 70, 94, 32)        0         
_________________________________________________________________
dropout_1 (Dropout)          (None, 70, 94, 32)        0         
_________________________________________________________________
conv2d_2 (Conv2D)            (None, 68, 92, 64)        18496     
_________________________________________________________________
max_pooling2d_2 (MaxPooling2 (None, 34, 46, 64)        0         
_________________________________________________________________
dropout_2 (Dropout)          (None, 34, 46, 64)        0         
_________________________________________________________________
conv2d_3 (Conv2D)            (None, 30, 42, 64)        102464    
_________________________________________________________________
max_pooling2d_3 (MaxPooling2 (None, 15, 21, 64)        0         
_________________________________________________________________
dropout_3 (Dropout)          (None, 15, 21, 64)        0         
_________________________________________________________________
conv2d_4 (Conv2D)            (None, 11, 17, 64)        102464    
_________________________________________________________________
max_pooling2d_4 (MaxPooling2 (None, 5, 8, 64)          0         
_________________________________________________________________
dropout_4 (Dropout)          (None, 5, 8, 64)          0         
_________________________________________________________________
flatten (Flatten)            (None, 2560)              0         
_________________________________________________________________
dropout_5 (Dropout)          (None, 2560)              0         
_________________________________________________________________
dense (Dense)                (None, 9)                 23049     
=================================================================
Total params: 256,617
Trainable params: 256,617
Non-trainable params: 0
_________________________________________________________________
Epoch 1/200
9/9 [==============================] - 2s 271ms/step - loss: 2.3557 - accuracy: 0.0764 - val_loss: 2.1962 - val_accuracy: 0.1134
Epoch 2/200
9/9 [==============================] - 2s 236ms/step - loss: 2.1913 - accuracy: 0.1319 - val_loss: 2.1966 - val_accuracy: 0.0825
Epoch 3/200
9/9 [==============================] - 2s 235ms/step - loss: 2.1976 - accuracy: 0.0833 - val_loss: 2.1962 - val_accuracy: 0.1134
Epoch 4/200
9/9 [==============================] - 2s 240ms/step - loss: 2.1853 - accuracy: 0.1250 - val_loss: 2.1957 - val_accuracy: 0.1031
Epoch 5/200
9/9 [==============================] - 2s 239ms/step - loss: 2.1833 - accuracy: 0.1354 - val_loss: 2.1944 - val_accuracy: 0.1134
Epoch 6/200
9/9 [==============================] - 2s 247ms/step - loss: 2.1776 - accuracy: 0.1701 - val_loss: 2.1945 - val_accuracy: 0.2062
Epoch 7/200
9/9 [==============================] - 2s 233ms/step - loss: 2.1771 - accuracy: 0.1771 - val_loss: 2.1948 - val_accuracy: 0.1340
Epoch 8/200
9/9 [==============================] - 2s 238ms/step - loss: 2.1635 - accuracy: 0.1389 - val_loss: 2.1916 - val_accuracy: 0.1134
Epoch 9/200
9/9 [==============================] - 2s 235ms/step - loss: 2.1694 - accuracy: 0.1354 - val_loss: 2.1931 - val_accuracy: 0.1237
Epoch 10/200
9/9 [==============================] - 2s 240ms/step - loss: 2.1439 - accuracy: 0.1632 - val_loss: 2.1886 - val_accuracy: 0.1856
Epoch 11/200
9/9 [==============================] - 2s 237ms/step - loss: 2.1392 - accuracy: 0.1806 - val_loss: 2.1845 - val_accuracy: 0.1649
Epoch 12/200
9/9 [==============================] - 2s 239ms/step - loss: 2.1233 - accuracy: 0.1944 - val_loss: 2.1710 - val_accuracy: 0.1856
Epoch 13/200
9/9 [==============================] - 2s 241ms/step - loss: 2.0783 - accuracy: 0.2396 - val_loss: 2.1572 - val_accuracy: 0.2577
Epoch 14/200
9/9 [==============================] - 2s 236ms/step - loss: 2.0770 - accuracy: 0.2083 - val_loss: 2.1611 - val_accuracy: 0.2268
Epoch 15/200
9/9 [==============================] - 2s 240ms/step - loss: 1.9991 - accuracy: 0.2500 - val_loss: 2.1013 - val_accuracy: 0.3196
Epoch 16/200
9/9 [==============================] - 2s 239ms/step - loss: 1.9614 - accuracy: 0.2465 - val_loss: 2.0832 - val_accuracy: 0.2990
Epoch 17/200
9/9 [==============================] - 2s 237ms/step - loss: 1.9295 - accuracy: 0.2465 - val_loss: 2.1123 - val_accuracy: 0.3093
Epoch 18/200
9/9 [==============================] - 2s 241ms/step - loss: 1.9356 - accuracy: 0.2708 - val_loss: 2.0459 - val_accuracy: 0.2887
Epoch 19/200
9/9 [==============================] - 2s 239ms/step - loss: 1.8774 - accuracy: 0.2847 - val_loss: 2.0127 - val_accuracy: 0.3196
Epoch 20/200
9/9 [==============================] - 2s 237ms/step - loss: 1.8488 - accuracy: 0.2535 - val_loss: 2.0216 - val_accuracy: 0.3196
Epoch 21/200
9/9 [==============================] - 2s 242ms/step - loss: 1.8072 - accuracy: 0.3056 - val_loss: 2.0032 - val_accuracy: 0.3196
Epoch 22/200
9/9 [==============================] - 2s 242ms/step - loss: 1.7888 - accuracy: 0.3507 - val_loss: 1.9202 - val_accuracy: 0.3814
Epoch 23/200
9/9 [==============================] - 2s 234ms/step - loss: 1.7362 - accuracy: 0.3611 - val_loss: 1.9953 - val_accuracy: 0.3918
Epoch 24/200
9/9 [==============================] - 2s 239ms/step - loss: 1.7387 - accuracy: 0.3715 - val_loss: 1.8688 - val_accuracy: 0.4330
Epoch 25/200
9/9 [==============================] - 2s 235ms/step - loss: 1.7570 - accuracy: 0.3229 - val_loss: 1.9319 - val_accuracy: 0.4330
Epoch 26/200
9/9 [==============================] - 2s 236ms/step - loss: 1.7126 - accuracy: 0.3646 - val_loss: 1.9006 - val_accuracy: 0.3814
Epoch 27/200
9/9 [==============================] - 2s 239ms/step - loss: 1.7024 - accuracy: 0.3438 - val_loss: 1.8593 - val_accuracy: 0.4021
Epoch 28/200
9/9 [==============================] - 2s 242ms/step - loss: 1.7164 - accuracy: 0.3021 - val_loss: 1.9146 - val_accuracy: 0.3814
Epoch 29/200
9/9 [==============================] - 2s 241ms/step - loss: 1.6569 - accuracy: 0.3611 - val_loss: 1.8086 - val_accuracy: 0.3711
Epoch 30/200
9/9 [==============================] - 2s 264ms/step - loss: 1.6473 - accuracy: 0.3611 - val_loss: 1.8594 - val_accuracy: 0.4845
Epoch 31/200
9/9 [==============================] - 2s 238ms/step - loss: 1.5706 - accuracy: 0.3889 - val_loss: 1.7691 - val_accuracy: 0.4639
Epoch 32/200
9/9 [==============================] - 2s 240ms/step - loss: 1.5319 - accuracy: 0.4306 - val_loss: 1.7218 - val_accuracy: 0.3608
Epoch 33/200
9/9 [==============================] - 2s 239ms/step - loss: 1.5517 - accuracy: 0.3993 - val_loss: 1.7162 - val_accuracy: 0.4948
Epoch 34/200
9/9 [==============================] - 2s 236ms/step - loss: 1.5385 - accuracy: 0.4444 - val_loss: 1.7389 - val_accuracy: 0.4845
Epoch 35/200
9/9 [==============================] - 2s 237ms/step - loss: 1.5294 - accuracy: 0.4792 - val_loss: 1.7086 - val_accuracy: 0.3918
Epoch 36/200
9/9 [==============================] - 2s 241ms/step - loss: 1.4893 - accuracy: 0.4062 - val_loss: 1.7000 - val_accuracy: 0.4845
Epoch 37/200
9/9 [==============================] - 2s 238ms/step - loss: 1.4764 - accuracy: 0.4931 - val_loss: 1.6446 - val_accuracy: 0.4948
Epoch 38/200
9/9 [==============================] - 2s 238ms/step - loss: 1.4579 - accuracy: 0.4931 - val_loss: 1.6467 - val_accuracy: 0.4433
Epoch 39/200
9/9 [==============================] - 2s 239ms/step - loss: 1.4641 - accuracy: 0.4444 - val_loss: 1.6171 - val_accuracy: 0.4845
Epoch 40/200
9/9 [==============================] - 2s 239ms/step - loss: 1.3807 - accuracy: 0.4931 - val_loss: 1.5935 - val_accuracy: 0.5052
Epoch 41/200
9/9 [==============================] - 2s 241ms/step - loss: 1.3628 - accuracy: 0.4861 - val_loss: 1.5821 - val_accuracy: 0.4330
Epoch 42/200
9/9 [==============================] - 2s 241ms/step - loss: 1.4170 - accuracy: 0.5000 - val_loss: 1.5583 - val_accuracy: 0.5155
Epoch 43/200
9/9 [==============================] - 2s 237ms/step - loss: 1.3124 - accuracy: 0.5521 - val_loss: 1.5521 - val_accuracy: 0.4639
Epoch 44/200
9/9 [==============================] - 2s 239ms/step - loss: 1.3615 - accuracy: 0.5139 - val_loss: 1.5415 - val_accuracy: 0.4639
Epoch 45/200
9/9 [==============================] - 2s 238ms/step - loss: 1.3398 - accuracy: 0.5035 - val_loss: 1.5346 - val_accuracy: 0.5155
Epoch 46/200
9/9 [==============================] - 2s 242ms/step - loss: 1.3031 - accuracy: 0.5347 - val_loss: 1.5253 - val_accuracy: 0.4433
Epoch 47/200
9/9 [==============================] - 2s 240ms/step - loss: 1.2440 - accuracy: 0.5243 - val_loss: 1.4628 - val_accuracy: 0.4948
Epoch 48/200
9/9 [==============================] - 2s 234ms/step - loss: 1.2858 - accuracy: 0.5382 - val_loss: 1.4772 - val_accuracy: 0.4742
Epoch 49/200
9/9 [==============================] - 2s 237ms/step - loss: 1.1700 - accuracy: 0.5799 - val_loss: 1.4650 - val_accuracy: 0.5361
Epoch 50/200
9/9 [==============================] - 2s 236ms/step - loss: 1.2260 - accuracy: 0.5694 - val_loss: 1.4872 - val_accuracy: 0.4948
Epoch 51/200
9/9 [==============================] - 2s 240ms/step - loss: 1.2390 - accuracy: 0.5729 - val_loss: 1.4368 - val_accuracy: 0.5052
Epoch 52/200
9/9 [==============================] - 2s 244ms/step - loss: 1.2274 - accuracy: 0.5660 - val_loss: 1.4307 - val_accuracy: 0.5773
Epoch 53/200
9/9 [==============================] - 2s 234ms/step - loss: 1.3021 - accuracy: 0.4931 - val_loss: 1.4753 - val_accuracy: 0.4845
Epoch 54/200
9/9 [==============================] - 2s 240ms/step - loss: 1.1974 - accuracy: 0.5417 - val_loss: 1.4268 - val_accuracy: 0.5464
Epoch 55/200
9/9 [==============================] - 2s 237ms/step - loss: 1.2089 - accuracy: 0.5590 - val_loss: 1.3990 - val_accuracy: 0.5361
Epoch 56/200
9/9 [==============================] - 2s 236ms/step - loss: 1.2256 - accuracy: 0.5972 - val_loss: 1.3991 - val_accuracy: 0.5670
Epoch 57/200
9/9 [==============================] - 2s 238ms/step - loss: 1.2094 - accuracy: 0.5556 - val_loss: 1.3993 - val_accuracy: 0.5464
Epoch 58/200
9/9 [==============================] - 2s 239ms/step - loss: 1.1432 - accuracy: 0.5486 - val_loss: 1.3761 - val_accuracy: 0.5567
Epoch 59/200
9/9 [==============================] - 2s 241ms/step - loss: 1.1738 - accuracy: 0.5694 - val_loss: 1.3717 - val_accuracy: 0.5773
Epoch 60/200
9/9 [==============================] - 2s 254ms/step - loss: 1.1903 - accuracy: 0.5590 - val_loss: 1.3548 - val_accuracy: 0.5567
Epoch 61/200
9/9 [==============================] - 2s 247ms/step - loss: 1.0896 - accuracy: 0.5903 - val_loss: 1.3593 - val_accuracy: 0.5464
Epoch 62/200
9/9 [==============================] - 2s 255ms/step - loss: 1.1350 - accuracy: 0.5799 - val_loss: 1.3537 - val_accuracy: 0.5567
Epoch 63/200
9/9 [==============================] - 2s 250ms/step - loss: 1.1040 - accuracy: 0.5833 - val_loss: 1.3186 - val_accuracy: 0.5979
Epoch 64/200
9/9 [==============================] - 2s 238ms/step - loss: 1.0624 - accuracy: 0.5972 - val_loss: 1.3240 - val_accuracy: 0.5464
Epoch 65/200
9/9 [==============================] - 2s 248ms/step - loss: 1.0363 - accuracy: 0.6354 - val_loss: 1.3031 - val_accuracy: 0.5464
Epoch 66/200
9/9 [==============================] - 2s 248ms/step - loss: 1.0541 - accuracy: 0.6042 - val_loss: 1.3171 - val_accuracy: 0.5258
Epoch 67/200
9/9 [==============================] - 2s 253ms/step - loss: 1.0706 - accuracy: 0.6146 - val_loss: 1.2592 - val_accuracy: 0.5876
Epoch 68/200
9/9 [==============================] - 2s 248ms/step - loss: 1.0427 - accuracy: 0.6285 - val_loss: 1.2837 - val_accuracy: 0.5979
Epoch 69/200
9/9 [==============================] - 2s 239ms/step - loss: 1.0323 - accuracy: 0.6250 - val_loss: 1.3140 - val_accuracy: 0.5361
Epoch 70/200
9/9 [==============================] - 2s 250ms/step - loss: 1.0680 - accuracy: 0.6181 - val_loss: 1.2819 - val_accuracy: 0.5670
Epoch 71/200
9/9 [==============================] - 2s 246ms/step - loss: 1.0589 - accuracy: 0.6250 - val_loss: 1.2708 - val_accuracy: 0.5361
Epoch 72/200
9/9 [==============================] - 2s 236ms/step - loss: 1.0410 - accuracy: 0.5972 - val_loss: 1.2671 - val_accuracy: 0.6392
Epoch 73/200
9/9 [==============================] - 2s 237ms/step - loss: 1.0728 - accuracy: 0.6424 - val_loss: 1.3063 - val_accuracy: 0.5464
Epoch 74/200
9/9 [==============================] - 2s 236ms/step - loss: 1.1118 - accuracy: 0.5938 - val_loss: 1.2966 - val_accuracy: 0.5773
Epoch 75/200
9/9 [==============================] - 2s 239ms/step - loss: 1.0058 - accuracy: 0.6319 - val_loss: 1.2693 - val_accuracy: 0.5670
Epoch 76/200
9/9 [==============================] - 2s 234ms/step - loss: 1.0212 - accuracy: 0.6354 - val_loss: 1.2658 - val_accuracy: 0.5567
Epoch 77/200
9/9 [==============================] - 2s 237ms/step - loss: 0.9979 - accuracy: 0.6458 - val_loss: 1.2684 - val_accuracy: 0.5464
Epoch 78/200
9/9 [==============================] - 2s 238ms/step - loss: 1.0301 - accuracy: 0.6354 - val_loss: 1.2605 - val_accuracy: 0.5670
Epoch 79/200
9/9 [==============================] - 2s 238ms/step - loss: 1.0601 - accuracy: 0.6007 - val_loss: 1.2794 - val_accuracy: 0.5670
Epoch 80/200
9/9 [==============================] - 2s 237ms/step - loss: 1.0265 - accuracy: 0.6111 - val_loss: 1.2683 - val_accuracy: 0.5876
Epoch 81/200
9/9 [==============================] - 2s 241ms/step - loss: 1.0152 - accuracy: 0.6701 - val_loss: 1.2544 - val_accuracy: 0.5979
Epoch 82/200
9/9 [==============================] - 2s 237ms/step - loss: 1.1278 - accuracy: 0.6076 - val_loss: 1.2576 - val_accuracy: 0.5876
Epoch 83/200
9/9 [==============================] - 2s 239ms/step - loss: 1.0711 - accuracy: 0.6111 - val_loss: 1.2639 - val_accuracy: 0.5876
Epoch 84/200
9/9 [==============================] - 2s 239ms/step - loss: 0.9647 - accuracy: 0.6597 - val_loss: 1.2562 - val_accuracy: 0.6186

fig = plt.figure()
plt.plot(history.history['accuracy'])
plt.plot(history.history['val_accuracy'])
plt.title('model accuracy')
plt.ylabel('accuracy')
plt.xlabel('epoch')
plt.legend(['train', 'validation'], loc='upper left')
plt.show()
fig.savefig('Accuracy vs Epoch.png')

plt.plot(history.history['loss'])
plt.plot(history.history['val_loss'])
plt.title('model loss')
plt.ylabel('loss')
plt.xlabel('epoch')
plt.legend(['train', 'validation'], loc='upper left')
plt.show()
fig.savefig('Loss vs Epoch.png')

EVALUATE

datagen_scale = keras.preprocessing.image.ImageDataGenerator(rescale=1/255)
train_generator= datagen_scale.flow_from_dataframe(train,x_col ='path',y_col='label',target_size=(288,384),batch_size=48,shuffle=True)
valid_generator= datagen_scale.flow_from_dataframe(val,x_col ='path',y_col='label',target_size=(288,384),shuffle=False)
test_generator = datagen_scale.flow_from_dataframe(test,x_col ='path',y_col='label',target_size=(288,384),shuffle=False)
model.evaluate(train_generator)
model.evaluate(valid_generator)
model.evaluate(test_generator)

Found 288 validated image filenames belonging to 9 classes.
Found 97 validated image filenames belonging to 9 classes.
Found 45 validated image filenames belonging to 9 classes.
6/6 [==============================] - 1s 190ms/step - loss: 1.1375 - accuracy: 0.6493
4/4 [==============================] - 0s 93ms/step - loss: 1.2547 - accuracy: 0.6392
2/2 [==============================] - 0s 103ms/step - loss: 1.2398 - accuracy: 0.5333

[1.2397706508636475, 0.5333333611488342]

PREDICT

demo = pd.DataFrame(test.groupby('label').first()).reset_index()
datagen_scale = keras.preprocessing.image.ImageDataGenerator(rescale=1/255)
test_generator = datagen_scale.flow_from_dataframe(demo,x_col ='path',y_col='label',target_size=(288,384),shuffle=False)
demo = pd.concat((demo,pd.Series(np.argmax(model.predict(test_generator),axis=1),name='Prediction').map({a:b for b,a in test_generator.class_indices.items()})),axis=1)
fig,ax = plt.subplots(3,3,figsize=(15,10))
plt.tick_params(left = False, bottom = False)
for i in range(3):
    for j in range(3):
        ax[i][j].imshow(plt.imread(demo.iloc[i*3+j].path))
        ax[i][j].tick_params(left = False, bottom = False)
        ax[i][j].set_xticks([])
        ax[i][j].set_yticks([])
        ax[i][j].set_title("Truth Label "+demo.iloc[i*3+j].label+"   \n PREDICTED   "+demo.iloc[i*3+j].Prediction)
fig.savefig('Predict.png')

Found 9 validated image filenames belonging to 9 classes.

model.save("Fish_breed_classification.h5")

DeepCC

!deepCC Fish_breed_classification.h5